- Email: [email protected]
The role of exendin-4-conjugated superparamagnetic iron oxide nanoparticles in beta-cell-targeted MRI
Biomaterials 34 (2013) 5843e5852
Contents lists available at SciVerse ScienceDirect
Biomaterials journal homepage: www.elsevier.com/locate/biomaterials
The role of exendin-4-conjugated superparamagnetic iron oxide nanoparticles in beta-cell-targeted MRI Bo Zhang a, Bin Yang a, Chuanxin Zhai b, Biao Jiang c, Yulian Wu a, * a
Department of Surgery, 2nd Affiliated Hospital, School of Medicine, Zhejiang University, 88 Jiefang Road, Zhejiang Province, Hangzhou 310009, PR China State Key Lab of Silicon Materials and Department of Materials Science and Engineering, Zhejiang University, Hangzhou 310027, PR China c Department of Radiology, 2nd Affiliated Hospital, School of Medicine, Zhejiang University, Hangzhou 310009, PR China b
a r t i c l e i n f o
a b s t r a c t
Article history: Received 27 March 2013 Accepted 10 April 2013 Available online 2 May 2013
Noninvasive targeted visualization of pancreatic beta cells or islets is becoming the focus of molecular imaging application in diabetes and islet transplantation studies, but it is currently unsuccessful due to the lack of specific beta cell biomarkers. Glucagon-like peptide 1 receptor (GLP-1R) is highly expressed in beta cells and considered as a promising target. We here developed a targeted superparamagnetic iron oxide (SPIO) nanoparticle using GLP-1 analog-exendin-4 which is conjugated to polyethylene glycol coated SPIO (PEG-SPIO). The results demonstrated that exendin-4 functionalized SPIO was able to specifically bind to and internalized by GLP-1R-expressing INS-1 cells, with the higher labeling efficiency than non-targeted nanoparticles. Notably, SPIO-exendin4 could differentially label islets in pancreatic slices or beta cell grafts in vitro. Systemic delivery of SPIO-exendin4 into nude mice bearing s.c. insulinomas (derived from INS-1 cells) leads to the accumulation of the nanoparticles in tumors, generating a strong magnetic resonance imaging contrast detectable by a clinical MRI scanner at field strength of 3.0 T, and the iron deposition in tumors was further confirmed by Prussian blue staining. Furthermore, preliminary biodistribution study indicated that SPIO-exendin4 had a tendency to accumulate in pancreas. Toxicity assessments demonstrated good biocompatibility in vivo. These results suggest that SPIOexendin4 has potential as molecularly targeted imaging agents for in vivo imaging of insulinoma, and possibly for future beta cell imaging. Ó 2013 Elsevier Ltd. All rights reserved.
Keywords: Glucagon-like peptide 1 receptor Exendin-4 MRI Beta cells Insulinoma
1. Introduction Direct and reliable imaging of beta cells could be a valuable tool for evaluating in vivo beta-cell mass/islet grafts and their function after islet transplantation or under medical therapies in diabetes mellitus [1]. Beta-cell specific imaging could also be served as a diagnostic method for detecting both primary and metastatic betacell derived tumor-insulinoma. However, despite of great advances in modern imaging devices, such imaging technique is currently not available due to the lack of a beta-cell specific contrast agent. Identification of beta-cell-specific biomarker is prerequisite for developing such contrast agents. The ideal beta-cell targeted probe should at least meet the following three requests: 1, the marker should be exclusively
Abbrevations: GLP-1R, glucagon-like peptide 1 receptor; MRI, magnetic resonance imaging; PEG, polyethylene glycol; PFA, paraformaldehyde; SPIO, superparamagnetic iron oxide; TEM, transmission electron microscope. * Corresponding author. Tel.: þ86 (0)571 87783602; fax: þ86 (0)571 87784604. E-mail addresses: [email protected], [email protected] (Y. Wu). 0142-9612/$ e see front matter Ó 2013 Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.biomaterials.2013.04.021
expressed on beta cells but not on others, or at least much higher than other tissues; 2, sufficient probes could accumulate in the pancreas to be available for beta cells’ uptake; 3, tissues around islets, such as pancreatic exocrine tissues and liver/kidney (the most common sites where islets are transplanted) should be devoid of this probe to achieve better signal-to-noise ratio. Up to now, a series of potential markers were adopted as imaging targets, such as using radio-labeled compounds and ligands, targeting the sulfonylurea receptor (SUR) [2], the presynaptic vesicular acetylcholine transporter [3], dopamine uptake in synaptic vesicles [4] and vesicular monoamine transporter 2 (VMAT2) [5] et al. to imaging pancreatic islets by means of positron emission tomography (PET) and SPECT. Unfortunately, none of these targets or tracer probes has allowed for islet imaging successfully in humans as yet, mainly due to the failure to achieve the required endocrine-to-exocrine binding ratio (>100:1) (in other words, unfavorable signal-tobackground ratio). Schneider S. et al. [6] used 125I-labeled single chain antibody (SCA), generated by phage-display technology, as specific imaging agents to highly selective labeling of the beta cells. But the precise site/receptor targeted by SCA has not been
5844
B. Zhang et al. / Biomaterials 34 (2013) 5843e5852
identified clearly. Besides, these SCA isolated from rodents might not be specific or suitable for human islets. The glucagon-like peptide 1 receptor (GLP-1R) is highly expressed in rat, mouse and human pancreatic beta cells but not in alpha, delta and pp cells [7]. Rare expression of the GLP-1R was observed in the pancreatic exocrine tissues and other organs (duodenum, stomach, lung, heart and brain). There are several studies reporting that radio-labeled exendin-4 (GLP-1 analog) was high uptake in pancreas and in subcutaneous insulinomas detected by SPECT/PET [8e11]. Notably, Francois Pattou [12] reported direct visualization of islet grafts transplanted in the muscle in a 48-yearold woman using 111In-labeled exendin-4, concluding that functional human beta-cells could be imaged in vivo with a tracer specific to the GLP-1 receptor. Although the nuclear imaging techniques (PET/SPECT) have high detective sensitivity, many unavoidable defects, such as low spatial resolution, poor metabolism of the tracers, hampered their further applications. MRI is a promising imaging approach, which has superior spatial resolution by using positive or negative contrast agents, even can be used to successfully monitor signal of a single cell in vivo [13]. Superparamagnetic iron oxide nanoparticle (SPIO) has made important contributions for developing negative contrast agent. It has been developed in many studies to conjugate antibody, peptide, and nucleotide to iron nanoparticles forming targeting contrast agent [14]. So far, there is no report of GLP-1 analog functionalized SPIO and its possible application in beta cell/insulinoma specific imaging. In the present study, we conjugated GLP-1 analog, exendin-4, to PEG-SPIO constructing the potential beta-cell-specific MRI probe(PEG)SPIO-exendin4, and investigated the binding and targeting properties of (PEG)SPIO-exendin4 to beta cells and insulinoma with high GLP-1R expression. 2. Materials and methods 2.1. Synthesis of amino-terminated PEG-SPIO PEG-SPIO nanoparticles were synthesized according to the procedure of Sun et al. [15] with slight modification (Fig. 1). Briefly, 0.1 g FeCl2(H2O)4 and 0.27 g FeCl3(H2O)6 (molar ratio: Fe3þ/Fe2þ ¼ 2:1) were dissolved in 50 ml PEG (molecular weight 600), and the mixture was heated to 125 C under Ar protection. 0.9 ml DI water was put in, followed by adding 10 ml EDEA. The temperature was maintained at 125 C for 30 min. Then the pallet would be cooled down to room temperature and be washed and centrifuged by toluene. The pellet was washed by magnetic isolation for three times and resuspended in 100 ml toluene. Next, 2 ml (3aminopropyl)tri-methoxysilane (APS) and 400 ml titanium(IV) isopropoxide Ti(OPr)4 were added to the solution for silanization reaction at 80 C for 12 h. The nanoparticles would be washed 3 times by toluene to remove excess APS and Ti(OPr)4. 2 ml poly(ethylene glycol)bis(carboxymethyl) ether and 1.25 g N,N0 -dicyclohexylcarbodiimide (DCC) were added to the APS-grafted nanoparticles with the reaction at 40 C for 3 h under Ar protection. 10 ml EDEA was added in, and the reaction was continued for another 5 h under the same condition. Finally, PEG-
coated nanoparticles were purified by washing three times with ethanol and PBS respectively. As synthesized PEG-SPIO turned out to be stable in aqueous solution without precipitates (Fig. 1e). The physiochemical properties of PEG-SPIO were characterized by Fourier Transform Infrared (FTIR) spectra, Vibrating Sample Magnetometer (VSM), Dynamic Light Scattering (DLS) and transmission electron microscope (TEM) examinations as described previously [16]. 2.2. Surface modification of amine-terminal PEG-SPIO with exendin-4 Exendin-4 was synthesized by Hua’an Biotechnology Cor. (Hangzhou) with the purity greater than 98%. Functionalization of PEG-SPIO with exendin-4 was achieved using a modification of carbodiimide (NHS/EDC) method (Fig. 2). Briefly, PEG-SPIO was activated by incubation with 37 ml 1-ethyl-3-(3-dimethylaminopropyl)-carbodiimide hydrochloride (EDC) and 50 mg N-hydroxysuccinimide (NHS) in 10 ml 0.1 M MES buffer (pH 6.3) at room temperature on a rotating shaker. After being activated, the PEG-SPIO was added to exendin-4 (1 ml of 1 mg/ml in PBS buffer) and the conjugation reaction was achieved by incubation for 24 h on a rotating shaker at room temperature. Then, the synthesized (PEG)SPIO-exendin4 was further isolated magnetically and purified by hyper-filtration to remove excess residues. 2.3. Cell culture and islet isolation INS-1 cells, a rat insulinoma cell line, were grown in monolayer cultures in regular RPMI-1640 medium supplemented with 10 mmol/l HEPES, 10% heatinactivated FCS, 2 mmol/l L-glutamine, 1 mmol/l sodium pyruvate, 50 mmol/l bmercaptoethanol, 100 IU/ml penicillin and 100 mg/ml streptomycin at 37 C in a humid atmosphere (5% CO2, 95% air). Primary islets were isolated by collagenase digestion from C57Bl/6 mice and separated using density gradient centrifugation as previously described [16]. Islets were incubated at 37 C in 2 ml RPMI-1640 medium at a density of 100e150 islets per well in 6-well plates. 2.4. In vitro cell labeling and imaging INS-1 cells were labeled by (PEG)SPIO-exendin4 or PEG-SPIO for comparing cellular uptake of targeting or non-targeting nanoparticles. Briefly, INS-1 cells were seeded in 6-well plates, 18e24 h later, cells were further incubated with labeling medium containing (PEG)SPIO-exendin4 or PEG-SPIO at 150 or 300 ug/ml iron concentration. 24 h later, the labeling medium was discarded, and cells were washed by PBS for three times. The same number of labeled cells in different groups were collected for following MRI scanning. In vitro imaging was performed using a clinical 3.0-T MR scanner (GE Sigma Excite HD, Milwaukee, WI, USA). Acquisition of T2weighted spin echo pulse sequences was based on previously described protocol [17]: TR/TE ¼ 4000/80 ms, echo train length ¼ 16, inter-echo time ¼ 76.1 ms, slice thickness/gap ¼ 1.0 mm/0.1 mm and the field of view (FOV) ¼ 10 6 cm2, NEX ¼ 4. Furthermore, INS-1 cells were also labeled by quantum dots (QDs) or exendin4conjugated QDs (QD-exendin4). INS-1 cells were plated on glass cover slides and grown for 24 h at 37 C in a humid atmosphere with 5% CO2. Cells were washed with PBS and incubated with cell culture medium containing QD-exendin4 or QDs for 24 h at 37 C. After incubation, cells were washed with PBS for three times, and the slides were mounted with DAPI-containing mounting medium, and examined under fluorescent microscope. 2.5. Prussian blue (PB) staining and quantification of intracellular iron content The existence of iron particle within beta cells or tissues was confirmed by PB staining as described previously [18]. The intracellular iron concentrations were quantified using inductively coupled plasma atomic absorption spectroscopy (ICPAAS, Hitachi 180-50). The labeled cell pellet was dissolved in 37% HCl solution at 70e
Fig. 1. (aed) Schematic illustration of the synthetic process for PEG-coated SPIO. (e) PEG-SPIO in aqueous solution.
B. Zhang et al. / Biomaterials 34 (2013) 5843e5852
5845
Fig. 2. Synthesis of exendin-4-conjugated PEG-SPIO. A,B. The molecular structure and sequence of exendin-4. C. Schematic diagram outlining the synthesis of (PEG)SPIO-exendin4 using NHS/EDC method.
80 C for 30 min. The samples were diluted to final iron concentrations of 1.0e 4.0 mg/l.
number of RBC, WBC and aspartate aminotransferase (AST)/alanine aminotransferase (ALT) levels) were analyzed.
2.6. In vivo studies
3. Results All animal experiments were carried out in accordance with the ethical guidelines of Animal Experimentation Committee in College of Medicine, Zhejiang University. 4e6 weeks old male athymic nu/nu mice were used for tumor implantation. 8e10 106 INS-1 cells in Logarithmic growth phase were collected in 100e200 ml PBS and injected subcutaneously in the right axilla of the nude mice to establish xenograft tumor model. The (PEG)SPIO-exendin4 or PEG-SPIO was injected into the tumor-bearing mice via tail vein (3 mg iron/kg); MRI scanning was performed before or at 24 h after nanoparticles injection. A series of T2-weighted coronal images were acquired using T2-weighted fast spin echo sequences (TR/TE ¼ 4000/80 ms) and special animal coils. The spatial resolution parameters were as follows: FOV ¼ 6 4.2 cm2, matrix size ¼ 288 256, slice thickness/gap ¼ 1.0 mm/0.1 mm, scan time ¼ 200 s, NEX ¼ 4. 2.7. Biodistribution of PEG-SPIO and (PEG)SPIO-exendin4 ICR mice were used for assessing the biodistribution of nanoparticles in the whole body. Mice (about 25 g weights) were injected with (PEG)SPIO-exendin4 or PEG-SPIO via tail vein (3 mg iron/kg). 24 h later, the mice were sacrificed, and the liver, lung, spleen, kidney, stomach, intestine, muscle, pancreas were dissected, fixed with 4% paraformaldehyde (PFA) and embedded in paraffin wax. Prussian blue staining was performed on slices of each organ to observe the existence/accumulation of iron.
3.1. Syntheses and characterization of PEG-SPIO Fig. 3A shows the TEM image of the PEG-SPIO. The nanoparticle displayed good dispersibility in aqueous solution, and the core sizes were about 5e10 nm. The chemical bond of PEG-SPIO was assessed by FTIR: amine terminal PEG with methylene signature peaks at 2930 cm1, carbonyl bands at 1628 cm1 corresponding to the amide bonds linking the PEG to the silane. A peak at 1104 cm1 corresponding to the SieO bond demonstrates the bonding between the silane and the iron oxide core (Fig. 3B). The mean hydrodynamic size of PEG-SPIO in PBS was 25 nm as determined by DLS (Fig. 3C). The magnetic property of PEG-SPIO was examined via VSM. Fig. 3D displayed these nanoparticles exhibited superparamagnetic behavior at 300 K with the saturation magnetization to be 29 emu/g at 2 T. 3.2. Immunofluorescent staining for GLP-1R and in vitro targeting assay
2.8. Biocompatibility of (PEG)SPIO-exendin4 ICR mice (about 25 g weight) were injected with (PEG)SPIO-exendin4 via tail vein (3 mg iron/kg). 24e48 h later, the whole organs (liver, lung, kidney, and spleen) were removed, fixed with 10% formalin, and embedded in paraffin wax. The slices of each organ were stained by hematoxylin and eosin (H&E), and evaluated to observe whether there was any acute toxicity existed in these organs. The whole blood and separated serum were also collected from the mice and a panel of parameters (the
Pancreatic slices from mice were stained with anti-GLP-1R antibody. Figure S1 demonstrated that islets were positive for GLP-1R, with the expression higher than surrounding tissues. To test the functionality of the exendin4-conjugated PEG-SPIO, an in vitro cell uptake assay was performed using INS-1 cells and primary islet. After being incubated with PEG-SPIO or (PEG)SPIO-
5846
B. Zhang et al. / Biomaterials 34 (2013) 5843e5852
Fig. 3. Characterization of PEG-SPIO. TEM (A), FTIR (B), DLS (C) and VSM (D) analysis of PEG-SPIO.
Fig. 4. A, Prussian blue staining of INS-1 cells labeled with PEG-SPIO or (PEG)SPIO-exendin4. B, MRI scanning of INS-1 cells labeled by respective nanoparticles (T2*-weighted image). C, determination of intracellular iron content in PEG-SPIO or (PEG)SPIO-exendin4-labeled cells by ICP-AAS.
B. Zhang et al. / Biomaterials 34 (2013) 5843e5852
5847
Fig. 5. A, Prussian blue staining of INS-1 cells labeled with PEG-SPIO or (PEG)SPIO-exendin4 after different incubation time (labeling after cell fixation). B, Prussian blue staining of isolated islet labeled with (PEG)SPIO-exendin4.
Exendin4 for 4 h (100 mg/ml iron concentration) at 37 C, INS-1 cells were fixed by 4% PFA and Prussian blue staining was performed. Fig. 4A shows that more percentage of cells was labeled by (PEG) SPIO-Exendin4 compared with non-targeted nanoparticle group. Then 5 105 INS-1 cells labeled by (PEG)SPIO-exendin4 or PEGSPIO (200 mg/ml iron) for 12 h were collected for MRI scanning. As shown in Fig. 4B, cells labeled by (PEG)SPIO-exendin4 displayed negative contrast-enhancement in T2*-weighted MR images than cells labeled by PEG-SPIO, which indicated that (PEG)SPIO-exendin4 could be endocytosed into INS-1 cells more easily via ligande receptor conjugation. Such conclusion was further supported by measuring intracellular iron content via ICP-AAS analysis (Fig. 4c, 14.42 6.26 pg/cell in SPIO-exendin4 group vs 3.56 1.07 pg/cell in PEG-SPIO group). Next we used the fixed cells to observe the targeting properties of (PEG)SPIO-Exendin4. INS-1 cells were fixed by 4% PFA followed by incubation with (PEG)SPIO-Exendin4 or PEG-SPIO (200 mg/ml iron) for 1, 2 or 4 h at 37 C. PB staining demonstrated that (PEG) SPIO-Exendin4 could also label fixed cells, whereas PEG-SPIO could not, even for longer incubation time (Fig. 5A). PFA-fixed primary islets could be labeled by (PEG)SPIO-exendin4 as well (Fig. 5B). For further directly verifying the targeting ability of exendin-4, we conjugated it to quantum dots (QDs). After being incubated with QD-exendin4 for 1 h, INS-1 cells were fixed by 4% PFA and observed under fluorescent microscope (Fig. 6). Cells labeled by
Fig. 6. Fluorescent imaging of living or fixed INS-1 cells incubated by QD-exendin4 or QDs. Upper, living cells; lower, fixed cells.
5848
B. Zhang et al. / Biomaterials 34 (2013) 5843e5852
Fig. 7. HE (A) and PB staining (BeG) of pancreatic slice. Pancreatic slices were incubated with (PEG)SPIO-exendin4 (200 mg/ml iron concentration) for 2 h at 37 C, and washed by PBS for three times, followed by PB staining. Islets could be labeled by (PEG)SPIO-exendin4, which could differentiate islets from pancreatic exocrine tissues in slice.
QD-exendin4 displayed red fluorescence in cytoplasm, whereas in QDs labeling group, less cells were labeled. Similar results were observed in fixed-cell labeling experiment. These results indicated that functionalization with exendin-4 could significantly improve the ability of cellular uptake of nanoparticles. 3.3. (PEG)SPIO-exendin4 binding to in situ islets and beta-cell grafts in vitro The purpose of developing beta-cell-targeted contrast agent is to visualize islets in situ or beta-cell grafts in vivo (in other words, differentiating islets from pancreatic exocrine tissue or other organs) by certain imaging techniques. We first analyzed whether targeted nanoparticle could differentiate islets from exocrine tissues in pancreatic slice. As shown in Fig. 7, islets could be notably labeled by (PEG)SPIO-exendin4 (PB staining), whereas surrounding exocrine tissues were not labeled. Furthermore, we transplanted beta cells (INS-1 cells) under renal capsule of the nude mice. The slices containing beta-cell
grafts were incubated with non-targeted SPIO or (PEG)SPIO-exendin4 as same as above procedure. Similarly, beta-cell grafts could be labeled by targeted nanoparticles, but surrounding tissues (renal capsule, renal tubule etc.) with less expression of GLP-1R could not (Fig. 8). These initial findings provide us that exendin-4-conjugated iron nanoparticles could specifically bind to islets/beta-cells in vitro, which is the prerequisite for further in vivo studies. 3.4. In vivo targeting and MR imaging of insulinomas using (PEG) SPIO-exendin4 In nude mice which bearing s.c. insulinomas derived from INS-1 cells, (PEG)SPIO-exendin4 could selectively accumulated in the tumors after 24 h post-injection in T2-weighted gradient echo imaging (signal decreased in tumor region), whereas there was no significant signal change in the tumors of the mice receiving nontargeted iron oxide particles (Fig. 9). To further confirm the distribution of non-targeted and (PEG)SPIO-exendin4 in tumor tissues, PB staining was performed on tumor sections from mice injected
Fig. 8. A, HE staining of beta-cell grafts (c, capsule of the kidney; g: grafts; k, kidney tubule); B,C, PB staining of beta-cell grafts in PEG-SPIO-incubated slices. C is the amplification of square area (graft region) in B. DeF, PB staining in (PEG)SPIO-exendin4-incubated slices, D and F are amplification of square area in E. D, renal tubule region; F, graft region.
B. Zhang et al. / Biomaterials 34 (2013) 5843e5852
5849
Fig. 9. Representative MR images of the insulinoma-bearing mice 24 h after intravenous administration of saline, non-targeted PEG-SPIO or (PEG)SPIO-exendin4.
with different contrast agent. Iron was proved only in the targeting tumor sections but not in non-targeting sections (Fig. 10). 3.5. Biodistribution and biocompatibility of PEG-SPIO and (PEG) SPIO-exendin4 in vivo We next investigated the biodistribution of PEG-SPIO and (PEG) SPIO-exendin4 in ICR mice strain. Nanoparticles were injected into mice via tail vein (iron content 3 mg/kg). At 24 h post-injection, mice were sacrificed. As shown in Fig. 11, iron was accumulated in liver, spleen, lung and kidney in non-targeted agent injection group, whereas it was absent in pancreatic (both exocrine and endocrine) tissues. In mice injected with (PEG)SPIO-exendin4, nanoparticles were unavoidably captured and accumulated in liver and spleen (Fig. 12). But notably, we observed the presence of iron in pancreatic tissue. In some areas, scattered iron deposited around
or in islets (Fig. 12). These findings manifest the beta-cell-targeted properties of exendin-4-conjugated nanoparticles in vivo. In addition, we performed histological analysis on various organs (liver, spleen, lung, and kidney) of mice injected with (PEG) SPIO-exendin4 to observe the probable acute toxicity reaction. There were no pathological changes in sections of above organs (data not show). Furthermore, there was no significant difference on blood cell panels and ALT/AST levels between (PEG)SPIO-exendin4-injected mice and untreated control (Table 1). 4. Discussion GLP-1 is an incretin hormone, which regulates glucose levels through binding to a G protein-coupled receptor (GLP-1R) and stimulating glucose dependent insulin secretion or biosynthesis, inducing glucagon secretion suppress, gastric emptying delay and
Fig. 10. PB staining of tumor sections from mice injected with PEG-SPIO or (PEG)SPIO-exendin4. A,B, PEG-SPIO-injected group; C,D, (PEG)SPIO-exendin4-injected group. B and D are the amplification of the region of interest (ROI) in A and C, and are counter-stained with nuclear fast red. Images confirm the deposits of oxide iron nanoparticles in targeted tumor tissues.
5850
B. Zhang et al. / Biomaterials 34 (2013) 5843e5852
Fig. 11. Histological analysis of biodistribution of PEG-SPIO. Organs were collected at 24 h post-injection. PB staining displayed that iron deposited in liver, kidney, spleen and lung, but not in pancreas (both exocrine and endocrine tissues).
Fig. 12. Biodistribution studies of (PEG)SPIO-exendin4. Nanoparticles were inevitably accumulated in liver and spleen. The presence of iron in pancreatic tissue and around/within islet were observed. Islets were circled and the deposited iron was marked by arrow.
B. Zhang et al. / Biomaterials 34 (2013) 5843e5852 Table 1 Serum liver enzyme levels and blood cell panels for mice administered with (PEG) SPIO-exendin4 and for mice receiving no injection. SPIO-PEG-exendin4 group ALT (U/l) AST (U/l) WBC (103/ml) RBC (106/ml) PLT (103/ml) Hb (g/dl)
27.67 72.67 3.47 8.31 1129.67 14.5
5.51 11.59 2.11 0.16 55.77 0.17
Untreated group 28.33 82 2.4 6.97 1144.33 12.7
4.51 6.24 0.52 1.17 219.81 1.42
promoting satiety [19]. GLP-1R is highly overexpressed in human insulinomas, gastrinomas and islets [20]. In islets, GLP-1R immunoreactivity is reported to be restricted to endocrine beta cells. Therefore, the ligand of GLP-1R could be an ideal probe for pancreatic beta-cell imaging. However, due to rapid degradation by the enzyme dipeptidyl peptidase-4 in the circulation, the half life of GLP-1 is very short, which makes it not suitable for serving as a conjugated substrate [21]. Exendin-4 is a GLP-1 analog, which shares 53% homology, displays similar biological properties and is much more stable for targeting imaging than GLP-1 metabolically. To date, there have been several reports of using radionuclidelabeled exendin-4 or exendin-3 for GLP-1R targeted imaging. Researchers developed 111In, 123I, 125I, 99mTc-exendin-4 conjugate for targeting GLP-1 receptor-positive tumors imaging and biodistribution studies by SPECT system, or using 18F, 68Ga-labeled exendin-4 for PET imaging. Besides of exendin-4, some researchers use 68Ga to label exendin-3 for PET imaging of GLP-1R-expressing insulinomas [22]. Interestingly, Mukai et al. [23] observed that 125 I-conjugated exendin (9e39), a truncated version of exendin with high affinity for the recpetor, could also achieve GLP-1R targeted imaging with the pancreas being the second organ possessing highest level of radioactivity. In the present study, we found that exendin-4-conjugated SPIO, compared with PEG-SPIO, could label both living and fixed beta cells more specifically, which was probably attributed to the ligandereceptor binding. With the ideal labeling efficiency, (PEG) SPIO-exendin4 might be serving as a good beta cell/islet-labeling contrast agent that lower iron concentration achieve superior detective sensitivity. To confirm the specificity properties of exendin4-based nanoparticles to beta cells, we prepared exendin4functionlized quantum dots to label INS-1 cells. Similarly, QDexendin4 displayed higher labeling efficiency in cells, either alive or fixed. These results suggest that efficient internalization of the ligand/receptor complex increased the concentration of the nanoparticles in beta cells, which possibly enhanced the effect of following MR/optical imaging. The ultimate goal of targeting beta-cell imaging is tracking the beta-cell grafts and visualizing pancreatic islets in situ by imaging technique after system administration of special contrast agent. Very recently, Wu et al. [24] used a PET probe-18F-labeled exendin4 to non-invasively evaluate islet graft mass in rodent intrahepatic transplantation model. In our study, it is interesting to note that our (PEG)SPIO-exendin4 could specifically bind to beta-cell grafts and islet in renal (containing grafts) and pancreatic slices. The differential binding to beta cells and the surrounding tissues by (PEG) SPIO-exendin4 highlights its potential role in tracking and imaging beta-cell grafts/islet in vivo by MRI. Another application of GLP-1R-targeted probe is to detect (GLP1R-positive) insulinoma. In clinical settings, the majority of insulinomas are presented as intrapancreatic, scattered and multiple small-sized (<2 cm) tumors. Especially, about 10e27% insulinomas are occult [25]. These make great challenges for accurate tumor localization (both primary and metastatic lesions) based on current
5851
imaging techniques if without specific targeting substrates. Receptor-targeted molecular imaging provides the possibility of disease diagnosis in above conditions and subsequent targeting therapy. Sowa-Staszczak et al. [8], for the first time, utilized 99mTClabeled exendin-4 to localize small insulinoma tumors in 11 patients, highlighting its promising application in clinical practice. In our insulinoma model, (PEG)SPIO-exendin4 also displayed tumortargeted properties evidenced by MR imaging and PB staining. It is worth further exploration that whether (PEG)SPIO-exendin4 could be serving as a GLP-1R-positive insulinoma specific diagnostic agent for clinical application. In our experiments, despite that (PEG)SPIO-exendin4 was unavoidably captured by reticulo-endothelial system (RES) in liver and spleen during biodistribution study, we are delighted to find that iron particles started to appear in and around islets in vivo. The improvement and modification of SPIO to avoid excessive consumption by RES in liver and spleen might provide opportunity for pancreatic islets to uptake more iron nanoparticles and to generate better signal. 5. Conclusion In summary, we here present the evidence that GLP-1 analog functionalized iron oxide nanoparticle probe is capable of targeting GLP-1 receptor-expressed beta cells and insulinoma in vitro and enable to achieve receptor-targeted MR imaging of the insulinoma in vivo. These results manifest that (PEG)SPIO-exendin4 might be served as a promising diagnostic probe for molecular targeting MR imaging of insulinoma in clinics. In addition, higher labeling efficiency of (PEG)SPIO-exendin4 to beta cells suggested that it might be an ideal beta-cell labeling agent. The properties of targeting to in situ beta cells also indicated its potential application in direct visualization of native islets in the near future. Acknowledgment This study was supported by grants from the National Natural Science Foundation of China (nos. 81100549, 81172158, 81001094, and 81272332) and the Ministry of Science and Technology of People’s Republic of China (no. 2007AA02Z476). Appendix ASupplementary data Supplementary data related to this article can be found at http:// dx.doi.org/10.1016/j.biomaterials.2013.04.021. References [1] Malaisse WJ, Louchami K, Sener A. Noninvasive imaging of pancreatic beta cells. Nat Rev Endocrinol 2009;5:394e400. [2] Oh CS, Kohanim S, Kong FL, Song HC, Huynh N, Mendez R, et al. Sulfonylurea receptor as a target for molecular imaging of pancreas beta cells with (99m) Tc-DTPA-glipizide. Ann Nucl Med 2012;26:253e61. [3] Clark PB, Gage HD, Brown-Proctor C, Buchheimer N, Calles-Escandon J, Mach RH, et al. Neurofunctional imaging of the pancreas utilizing the cholinergic PET radioligand [18F]4-fluorobenzyltrozamicol. Eur J Nucl Med Mol Imaging 2004;31:258e60. [4] Otonkoski T, Nanto-Salonen K, Seppanen M, Veijola R, Huopio H, Hussain K, et al. Noninvasive diagnosis of focal hyperinsulinism of infancy with [18F]DOPA positron emission tomography. Diabetes 2006;55:13e8. [5] Goland R, Freeby M, Parsey R, Saisho Y, Kumar D, Simpson N, et al. 11Cdihydrotetrabenazine PET of the pancreas in subjects with long-standing type 1 diabetes and in healthy controls. J Nucl Med 2009;50:382e9. [6] Ueberberg S, Meier JJ, Waengler C, Schechinger W, Dietrich JW, Tannapfel A, et al. Generation of novel single-chain antibodies by phage-display technology to direct imaging agents highly selective to pancreatic beta- or alpha-cells in vivo. Diabetes 2009;58:2324e34. [7] Zhang Y, Chen W. Radiolabeled glucagon-like peptide-1 analogues: a new pancreatic beta-cell imaging agent. Nucl Med Commun 2012;33:223e7.
5852
B. Zhang et al. / Biomaterials 34 (2013) 5843e5852
[8] Sowa-Staszczak A, Pach D, Mikolajczak R, Macke H, Jabrocka-Hybel A, Stefanska A, et al. Glucagon-like peptide-1 receptor imaging with [Lys(40)(Ahx-HYNIC- (99m)Tc/EDDA)NH 2]-exendin-4 for the detection of insulinoma. Eur J Nucl Med Mol Imaging 2013;40:524e31. [9] Kiesewetter DO, Gao H, Ma Y, Niu G, Quan Q, Guo N, et al. 18F-radiolabeled analogs of exendin-4 for PET imaging of GLP-1 in insulinoma. Eur J Nucl Med Mol Imaging 2012;39:463e73. [10] Wild D, Behe M, Wicki A, Storch D, Waser B, Gotthardt M, et al. [Lys40(AhxDTPA-111In)NH2]exendin-4, a very promising ligand for glucagon-like peptide-1 (GLP-1) receptor targeting. J Nucl Med 2006;47:2025e33. [11] Christ E, Wild D, Forrer F, Brandle M, Sahli R, Clerici T, et al. Glucagon-like peptide-1 receptor imaging for localization of insulinomas. J Clin Endocrinol Metab 2009;94:4398e405. [12] Pattou F, Kerr-Conte J, Wild D. GLP-1-receptor scanning for imaging of human beta cells transplanted in muscle. N Engl J Med 2010;363:1289e90. [13] Rumenapp C, Gleich B, Haase A. Magnetic nanoparticles in magnetic resonance imaging and diagnostics. Pharm Res 2012;29:1165e79. [14] Rosen JE, Chan L, Shieh DB, Gu FX. Iron oxide nanoparticles for targeted cancer imaging and diagnostics. Nanomedicine 2012;8:275e90. [15] Sun C, Du K, Fang C, Bhattarai N, Veiseh O, Kievit F, et al. PEG-mediated synthesis of highly dispersive multifunctional superparamagnetic nanoparticles: their physicochemical properties and function in vivo. ACS Nano 2010;4:2402e10. [16] Huang H, Xie Q, Kang M, Zhang B, Zhang H, Chen J, et al. Labeling transplanted mice islet with polyvinylpyrrolidone coated superparamagnetic iron oxide nanoparticles for in vivo detection by magnetic resonance imaging. Nanotechnology 2009;20:365101.
[17] Zhang B, Jiang B, Chen Y, Huang H, Xie Q, Kang M, et al. Detection of viability of transplanted beta cells labeled with a novel contrast agent polyvinylpyrrolidone-coated superparamagnetic iron oxide nanoparticles by magnetic resonance imaging. Contrast Media Mol Imaging 2012;7: 35e44. [18] Tai JH, Foster P, Rosales A, Feng B, Hasilo C, Martinez V, et al. Imaging islets labeled with magnetic nanoparticles at 1.5 Tesla. Diabetes 2006;55:2931e8. [19] Holst JJ, Deacon CF, Vilsboll T, Krarup T, Madsbad S. Glucagon-like peptide-1, glucose homeostasis and diabetes. Trends Mol Med 2008;14:161e8. [20] Kiesewetter DO, Guo N, Guo J, Gao H, Zhu L, Ma Y, et al. Evaluation of an [(18) F]AlF-NOTA Analog of Exendin-4 for imaging of GLP-1 receptor in insulinoma. Theranostics 2012;2:999e1009. [21] Deacon CF, Johnsen AH, Holst JJ. Degradation of glucagon-like peptide-1 by human plasma in vitro yields an N-terminally truncated peptide that is a major endogenous metabolite in vivo. J Clin Endocrinol Metab 1995;80: 952e7. [22] Brom M, Oyen WJ, Joosten L, Gotthardt M, Boerman OC. 68Ga-labelled exendin-3, a new agent for the detection of insulinomas with PET. Eur J Nucl Med Mol Imaging 2010;37:1345e55. [23] Mukai E, Toyoda K, Kimura H, Kawashima H, Fujimoto H, Ueda M, et al. GLP-1 receptor antagonist as a potential probe for pancreatic beta-cell imaging. Biochem Biophys Res Commun 2009;389:523e6. [24] Wu Z, Liu S, Hassink M, Nair I, Park R, Li L, et al. Development and evaluation of 18F-TTCO-Cys40-Exendin-4: a PET probe for imaging transplanted islets. J Nucl Med 2013;54:244e51. [25] Abboud B, Boujaoude J. Occult sporadic insulinoma: localization and surgical strategy. World J Gastroenterol 2008;14:657e65.
Contents lists available at SciVerse ScienceDirect
Biomaterials journal homepage: www.elsevier.com/locate/biomaterials
The role of exendin-4-conjugated superparamagnetic iron oxide nanoparticles in beta-cell-targeted MRI Bo Zhang a, Bin Yang a, Chuanxin Zhai b, Biao Jiang c, Yulian Wu a, * a
Department of Surgery, 2nd Affiliated Hospital, School of Medicine, Zhejiang University, 88 Jiefang Road, Zhejiang Province, Hangzhou 310009, PR China State Key Lab of Silicon Materials and Department of Materials Science and Engineering, Zhejiang University, Hangzhou 310027, PR China c Department of Radiology, 2nd Affiliated Hospital, School of Medicine, Zhejiang University, Hangzhou 310009, PR China b
a r t i c l e i n f o
a b s t r a c t
Article history: Received 27 March 2013 Accepted 10 April 2013 Available online 2 May 2013
Noninvasive targeted visualization of pancreatic beta cells or islets is becoming the focus of molecular imaging application in diabetes and islet transplantation studies, but it is currently unsuccessful due to the lack of specific beta cell biomarkers. Glucagon-like peptide 1 receptor (GLP-1R) is highly expressed in beta cells and considered as a promising target. We here developed a targeted superparamagnetic iron oxide (SPIO) nanoparticle using GLP-1 analog-exendin-4 which is conjugated to polyethylene glycol coated SPIO (PEG-SPIO). The results demonstrated that exendin-4 functionalized SPIO was able to specifically bind to and internalized by GLP-1R-expressing INS-1 cells, with the higher labeling efficiency than non-targeted nanoparticles. Notably, SPIO-exendin4 could differentially label islets in pancreatic slices or beta cell grafts in vitro. Systemic delivery of SPIO-exendin4 into nude mice bearing s.c. insulinomas (derived from INS-1 cells) leads to the accumulation of the nanoparticles in tumors, generating a strong magnetic resonance imaging contrast detectable by a clinical MRI scanner at field strength of 3.0 T, and the iron deposition in tumors was further confirmed by Prussian blue staining. Furthermore, preliminary biodistribution study indicated that SPIO-exendin4 had a tendency to accumulate in pancreas. Toxicity assessments demonstrated good biocompatibility in vivo. These results suggest that SPIOexendin4 has potential as molecularly targeted imaging agents for in vivo imaging of insulinoma, and possibly for future beta cell imaging. Ó 2013 Elsevier Ltd. All rights reserved.
Keywords: Glucagon-like peptide 1 receptor Exendin-4 MRI Beta cells Insulinoma
1. Introduction Direct and reliable imaging of beta cells could be a valuable tool for evaluating in vivo beta-cell mass/islet grafts and their function after islet transplantation or under medical therapies in diabetes mellitus [1]. Beta-cell specific imaging could also be served as a diagnostic method for detecting both primary and metastatic betacell derived tumor-insulinoma. However, despite of great advances in modern imaging devices, such imaging technique is currently not available due to the lack of a beta-cell specific contrast agent. Identification of beta-cell-specific biomarker is prerequisite for developing such contrast agents. The ideal beta-cell targeted probe should at least meet the following three requests: 1, the marker should be exclusively
Abbrevations: GLP-1R, glucagon-like peptide 1 receptor; MRI, magnetic resonance imaging; PEG, polyethylene glycol; PFA, paraformaldehyde; SPIO, superparamagnetic iron oxide; TEM, transmission electron microscope. * Corresponding author. Tel.: þ86 (0)571 87783602; fax: þ86 (0)571 87784604. E-mail addresses: [email protected], [email protected] (Y. Wu). 0142-9612/$ e see front matter Ó 2013 Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.biomaterials.2013.04.021
expressed on beta cells but not on others, or at least much higher than other tissues; 2, sufficient probes could accumulate in the pancreas to be available for beta cells’ uptake; 3, tissues around islets, such as pancreatic exocrine tissues and liver/kidney (the most common sites where islets are transplanted) should be devoid of this probe to achieve better signal-to-noise ratio. Up to now, a series of potential markers were adopted as imaging targets, such as using radio-labeled compounds and ligands, targeting the sulfonylurea receptor (SUR) [2], the presynaptic vesicular acetylcholine transporter [3], dopamine uptake in synaptic vesicles [4] and vesicular monoamine transporter 2 (VMAT2) [5] et al. to imaging pancreatic islets by means of positron emission tomography (PET) and SPECT. Unfortunately, none of these targets or tracer probes has allowed for islet imaging successfully in humans as yet, mainly due to the failure to achieve the required endocrine-to-exocrine binding ratio (>100:1) (in other words, unfavorable signal-tobackground ratio). Schneider S. et al. [6] used 125I-labeled single chain antibody (SCA), generated by phage-display technology, as specific imaging agents to highly selective labeling of the beta cells. But the precise site/receptor targeted by SCA has not been
5844
B. Zhang et al. / Biomaterials 34 (2013) 5843e5852
identified clearly. Besides, these SCA isolated from rodents might not be specific or suitable for human islets. The glucagon-like peptide 1 receptor (GLP-1R) is highly expressed in rat, mouse and human pancreatic beta cells but not in alpha, delta and pp cells [7]. Rare expression of the GLP-1R was observed in the pancreatic exocrine tissues and other organs (duodenum, stomach, lung, heart and brain). There are several studies reporting that radio-labeled exendin-4 (GLP-1 analog) was high uptake in pancreas and in subcutaneous insulinomas detected by SPECT/PET [8e11]. Notably, Francois Pattou [12] reported direct visualization of islet grafts transplanted in the muscle in a 48-yearold woman using 111In-labeled exendin-4, concluding that functional human beta-cells could be imaged in vivo with a tracer specific to the GLP-1 receptor. Although the nuclear imaging techniques (PET/SPECT) have high detective sensitivity, many unavoidable defects, such as low spatial resolution, poor metabolism of the tracers, hampered their further applications. MRI is a promising imaging approach, which has superior spatial resolution by using positive or negative contrast agents, even can be used to successfully monitor signal of a single cell in vivo [13]. Superparamagnetic iron oxide nanoparticle (SPIO) has made important contributions for developing negative contrast agent. It has been developed in many studies to conjugate antibody, peptide, and nucleotide to iron nanoparticles forming targeting contrast agent [14]. So far, there is no report of GLP-1 analog functionalized SPIO and its possible application in beta cell/insulinoma specific imaging. In the present study, we conjugated GLP-1 analog, exendin-4, to PEG-SPIO constructing the potential beta-cell-specific MRI probe(PEG)SPIO-exendin4, and investigated the binding and targeting properties of (PEG)SPIO-exendin4 to beta cells and insulinoma with high GLP-1R expression. 2. Materials and methods 2.1. Synthesis of amino-terminated PEG-SPIO PEG-SPIO nanoparticles were synthesized according to the procedure of Sun et al. [15] with slight modification (Fig. 1). Briefly, 0.1 g FeCl2(H2O)4 and 0.27 g FeCl3(H2O)6 (molar ratio: Fe3þ/Fe2þ ¼ 2:1) were dissolved in 50 ml PEG (molecular weight 600), and the mixture was heated to 125 C under Ar protection. 0.9 ml DI water was put in, followed by adding 10 ml EDEA. The temperature was maintained at 125 C for 30 min. Then the pallet would be cooled down to room temperature and be washed and centrifuged by toluene. The pellet was washed by magnetic isolation for three times and resuspended in 100 ml toluene. Next, 2 ml (3aminopropyl)tri-methoxysilane (APS) and 400 ml titanium(IV) isopropoxide Ti(OPr)4 were added to the solution for silanization reaction at 80 C for 12 h. The nanoparticles would be washed 3 times by toluene to remove excess APS and Ti(OPr)4. 2 ml poly(ethylene glycol)bis(carboxymethyl) ether and 1.25 g N,N0 -dicyclohexylcarbodiimide (DCC) were added to the APS-grafted nanoparticles with the reaction at 40 C for 3 h under Ar protection. 10 ml EDEA was added in, and the reaction was continued for another 5 h under the same condition. Finally, PEG-
coated nanoparticles were purified by washing three times with ethanol and PBS respectively. As synthesized PEG-SPIO turned out to be stable in aqueous solution without precipitates (Fig. 1e). The physiochemical properties of PEG-SPIO were characterized by Fourier Transform Infrared (FTIR) spectra, Vibrating Sample Magnetometer (VSM), Dynamic Light Scattering (DLS) and transmission electron microscope (TEM) examinations as described previously [16]. 2.2. Surface modification of amine-terminal PEG-SPIO with exendin-4 Exendin-4 was synthesized by Hua’an Biotechnology Cor. (Hangzhou) with the purity greater than 98%. Functionalization of PEG-SPIO with exendin-4 was achieved using a modification of carbodiimide (NHS/EDC) method (Fig. 2). Briefly, PEG-SPIO was activated by incubation with 37 ml 1-ethyl-3-(3-dimethylaminopropyl)-carbodiimide hydrochloride (EDC) and 50 mg N-hydroxysuccinimide (NHS) in 10 ml 0.1 M MES buffer (pH 6.3) at room temperature on a rotating shaker. After being activated, the PEG-SPIO was added to exendin-4 (1 ml of 1 mg/ml in PBS buffer) and the conjugation reaction was achieved by incubation for 24 h on a rotating shaker at room temperature. Then, the synthesized (PEG)SPIO-exendin4 was further isolated magnetically and purified by hyper-filtration to remove excess residues. 2.3. Cell culture and islet isolation INS-1 cells, a rat insulinoma cell line, were grown in monolayer cultures in regular RPMI-1640 medium supplemented with 10 mmol/l HEPES, 10% heatinactivated FCS, 2 mmol/l L-glutamine, 1 mmol/l sodium pyruvate, 50 mmol/l bmercaptoethanol, 100 IU/ml penicillin and 100 mg/ml streptomycin at 37 C in a humid atmosphere (5% CO2, 95% air). Primary islets were isolated by collagenase digestion from C57Bl/6 mice and separated using density gradient centrifugation as previously described [16]. Islets were incubated at 37 C in 2 ml RPMI-1640 medium at a density of 100e150 islets per well in 6-well plates. 2.4. In vitro cell labeling and imaging INS-1 cells were labeled by (PEG)SPIO-exendin4 or PEG-SPIO for comparing cellular uptake of targeting or non-targeting nanoparticles. Briefly, INS-1 cells were seeded in 6-well plates, 18e24 h later, cells were further incubated with labeling medium containing (PEG)SPIO-exendin4 or PEG-SPIO at 150 or 300 ug/ml iron concentration. 24 h later, the labeling medium was discarded, and cells were washed by PBS for three times. The same number of labeled cells in different groups were collected for following MRI scanning. In vitro imaging was performed using a clinical 3.0-T MR scanner (GE Sigma Excite HD, Milwaukee, WI, USA). Acquisition of T2weighted spin echo pulse sequences was based on previously described protocol [17]: TR/TE ¼ 4000/80 ms, echo train length ¼ 16, inter-echo time ¼ 76.1 ms, slice thickness/gap ¼ 1.0 mm/0.1 mm and the field of view (FOV) ¼ 10 6 cm2, NEX ¼ 4. Furthermore, INS-1 cells were also labeled by quantum dots (QDs) or exendin4conjugated QDs (QD-exendin4). INS-1 cells were plated on glass cover slides and grown for 24 h at 37 C in a humid atmosphere with 5% CO2. Cells were washed with PBS and incubated with cell culture medium containing QD-exendin4 or QDs for 24 h at 37 C. After incubation, cells were washed with PBS for three times, and the slides were mounted with DAPI-containing mounting medium, and examined under fluorescent microscope. 2.5. Prussian blue (PB) staining and quantification of intracellular iron content The existence of iron particle within beta cells or tissues was confirmed by PB staining as described previously [18]. The intracellular iron concentrations were quantified using inductively coupled plasma atomic absorption spectroscopy (ICPAAS, Hitachi 180-50). The labeled cell pellet was dissolved in 37% HCl solution at 70e
Fig. 1. (aed) Schematic illustration of the synthetic process for PEG-coated SPIO. (e) PEG-SPIO in aqueous solution.
B. Zhang et al. / Biomaterials 34 (2013) 5843e5852
5845
Fig. 2. Synthesis of exendin-4-conjugated PEG-SPIO. A,B. The molecular structure and sequence of exendin-4. C. Schematic diagram outlining the synthesis of (PEG)SPIO-exendin4 using NHS/EDC method.
80 C for 30 min. The samples were diluted to final iron concentrations of 1.0e 4.0 mg/l.
number of RBC, WBC and aspartate aminotransferase (AST)/alanine aminotransferase (ALT) levels) were analyzed.
2.6. In vivo studies
3. Results All animal experiments were carried out in accordance with the ethical guidelines of Animal Experimentation Committee in College of Medicine, Zhejiang University. 4e6 weeks old male athymic nu/nu mice were used for tumor implantation. 8e10 106 INS-1 cells in Logarithmic growth phase were collected in 100e200 ml PBS and injected subcutaneously in the right axilla of the nude mice to establish xenograft tumor model. The (PEG)SPIO-exendin4 or PEG-SPIO was injected into the tumor-bearing mice via tail vein (3 mg iron/kg); MRI scanning was performed before or at 24 h after nanoparticles injection. A series of T2-weighted coronal images were acquired using T2-weighted fast spin echo sequences (TR/TE ¼ 4000/80 ms) and special animal coils. The spatial resolution parameters were as follows: FOV ¼ 6 4.2 cm2, matrix size ¼ 288 256, slice thickness/gap ¼ 1.0 mm/0.1 mm, scan time ¼ 200 s, NEX ¼ 4. 2.7. Biodistribution of PEG-SPIO and (PEG)SPIO-exendin4 ICR mice were used for assessing the biodistribution of nanoparticles in the whole body. Mice (about 25 g weights) were injected with (PEG)SPIO-exendin4 or PEG-SPIO via tail vein (3 mg iron/kg). 24 h later, the mice were sacrificed, and the liver, lung, spleen, kidney, stomach, intestine, muscle, pancreas were dissected, fixed with 4% paraformaldehyde (PFA) and embedded in paraffin wax. Prussian blue staining was performed on slices of each organ to observe the existence/accumulation of iron.
3.1. Syntheses and characterization of PEG-SPIO Fig. 3A shows the TEM image of the PEG-SPIO. The nanoparticle displayed good dispersibility in aqueous solution, and the core sizes were about 5e10 nm. The chemical bond of PEG-SPIO was assessed by FTIR: amine terminal PEG with methylene signature peaks at 2930 cm1, carbonyl bands at 1628 cm1 corresponding to the amide bonds linking the PEG to the silane. A peak at 1104 cm1 corresponding to the SieO bond demonstrates the bonding between the silane and the iron oxide core (Fig. 3B). The mean hydrodynamic size of PEG-SPIO in PBS was 25 nm as determined by DLS (Fig. 3C). The magnetic property of PEG-SPIO was examined via VSM. Fig. 3D displayed these nanoparticles exhibited superparamagnetic behavior at 300 K with the saturation magnetization to be 29 emu/g at 2 T. 3.2. Immunofluorescent staining for GLP-1R and in vitro targeting assay
2.8. Biocompatibility of (PEG)SPIO-exendin4 ICR mice (about 25 g weight) were injected with (PEG)SPIO-exendin4 via tail vein (3 mg iron/kg). 24e48 h later, the whole organs (liver, lung, kidney, and spleen) were removed, fixed with 10% formalin, and embedded in paraffin wax. The slices of each organ were stained by hematoxylin and eosin (H&E), and evaluated to observe whether there was any acute toxicity existed in these organs. The whole blood and separated serum were also collected from the mice and a panel of parameters (the
Pancreatic slices from mice were stained with anti-GLP-1R antibody. Figure S1 demonstrated that islets were positive for GLP-1R, with the expression higher than surrounding tissues. To test the functionality of the exendin4-conjugated PEG-SPIO, an in vitro cell uptake assay was performed using INS-1 cells and primary islet. After being incubated with PEG-SPIO or (PEG)SPIO-
5846
B. Zhang et al. / Biomaterials 34 (2013) 5843e5852
Fig. 3. Characterization of PEG-SPIO. TEM (A), FTIR (B), DLS (C) and VSM (D) analysis of PEG-SPIO.
Fig. 4. A, Prussian blue staining of INS-1 cells labeled with PEG-SPIO or (PEG)SPIO-exendin4. B, MRI scanning of INS-1 cells labeled by respective nanoparticles (T2*-weighted image). C, determination of intracellular iron content in PEG-SPIO or (PEG)SPIO-exendin4-labeled cells by ICP-AAS.
B. Zhang et al. / Biomaterials 34 (2013) 5843e5852
5847
Fig. 5. A, Prussian blue staining of INS-1 cells labeled with PEG-SPIO or (PEG)SPIO-exendin4 after different incubation time (labeling after cell fixation). B, Prussian blue staining of isolated islet labeled with (PEG)SPIO-exendin4.
Exendin4 for 4 h (100 mg/ml iron concentration) at 37 C, INS-1 cells were fixed by 4% PFA and Prussian blue staining was performed. Fig. 4A shows that more percentage of cells was labeled by (PEG) SPIO-Exendin4 compared with non-targeted nanoparticle group. Then 5 105 INS-1 cells labeled by (PEG)SPIO-exendin4 or PEGSPIO (200 mg/ml iron) for 12 h were collected for MRI scanning. As shown in Fig. 4B, cells labeled by (PEG)SPIO-exendin4 displayed negative contrast-enhancement in T2*-weighted MR images than cells labeled by PEG-SPIO, which indicated that (PEG)SPIO-exendin4 could be endocytosed into INS-1 cells more easily via ligande receptor conjugation. Such conclusion was further supported by measuring intracellular iron content via ICP-AAS analysis (Fig. 4c, 14.42 6.26 pg/cell in SPIO-exendin4 group vs 3.56 1.07 pg/cell in PEG-SPIO group). Next we used the fixed cells to observe the targeting properties of (PEG)SPIO-Exendin4. INS-1 cells were fixed by 4% PFA followed by incubation with (PEG)SPIO-Exendin4 or PEG-SPIO (200 mg/ml iron) for 1, 2 or 4 h at 37 C. PB staining demonstrated that (PEG) SPIO-Exendin4 could also label fixed cells, whereas PEG-SPIO could not, even for longer incubation time (Fig. 5A). PFA-fixed primary islets could be labeled by (PEG)SPIO-exendin4 as well (Fig. 5B). For further directly verifying the targeting ability of exendin-4, we conjugated it to quantum dots (QDs). After being incubated with QD-exendin4 for 1 h, INS-1 cells were fixed by 4% PFA and observed under fluorescent microscope (Fig. 6). Cells labeled by
Fig. 6. Fluorescent imaging of living or fixed INS-1 cells incubated by QD-exendin4 or QDs. Upper, living cells; lower, fixed cells.
5848
B. Zhang et al. / Biomaterials 34 (2013) 5843e5852
Fig. 7. HE (A) and PB staining (BeG) of pancreatic slice. Pancreatic slices were incubated with (PEG)SPIO-exendin4 (200 mg/ml iron concentration) for 2 h at 37 C, and washed by PBS for three times, followed by PB staining. Islets could be labeled by (PEG)SPIO-exendin4, which could differentiate islets from pancreatic exocrine tissues in slice.
QD-exendin4 displayed red fluorescence in cytoplasm, whereas in QDs labeling group, less cells were labeled. Similar results were observed in fixed-cell labeling experiment. These results indicated that functionalization with exendin-4 could significantly improve the ability of cellular uptake of nanoparticles. 3.3. (PEG)SPIO-exendin4 binding to in situ islets and beta-cell grafts in vitro The purpose of developing beta-cell-targeted contrast agent is to visualize islets in situ or beta-cell grafts in vivo (in other words, differentiating islets from pancreatic exocrine tissue or other organs) by certain imaging techniques. We first analyzed whether targeted nanoparticle could differentiate islets from exocrine tissues in pancreatic slice. As shown in Fig. 7, islets could be notably labeled by (PEG)SPIO-exendin4 (PB staining), whereas surrounding exocrine tissues were not labeled. Furthermore, we transplanted beta cells (INS-1 cells) under renal capsule of the nude mice. The slices containing beta-cell
grafts were incubated with non-targeted SPIO or (PEG)SPIO-exendin4 as same as above procedure. Similarly, beta-cell grafts could be labeled by targeted nanoparticles, but surrounding tissues (renal capsule, renal tubule etc.) with less expression of GLP-1R could not (Fig. 8). These initial findings provide us that exendin-4-conjugated iron nanoparticles could specifically bind to islets/beta-cells in vitro, which is the prerequisite for further in vivo studies. 3.4. In vivo targeting and MR imaging of insulinomas using (PEG) SPIO-exendin4 In nude mice which bearing s.c. insulinomas derived from INS-1 cells, (PEG)SPIO-exendin4 could selectively accumulated in the tumors after 24 h post-injection in T2-weighted gradient echo imaging (signal decreased in tumor region), whereas there was no significant signal change in the tumors of the mice receiving nontargeted iron oxide particles (Fig. 9). To further confirm the distribution of non-targeted and (PEG)SPIO-exendin4 in tumor tissues, PB staining was performed on tumor sections from mice injected
Fig. 8. A, HE staining of beta-cell grafts (c, capsule of the kidney; g: grafts; k, kidney tubule); B,C, PB staining of beta-cell grafts in PEG-SPIO-incubated slices. C is the amplification of square area (graft region) in B. DeF, PB staining in (PEG)SPIO-exendin4-incubated slices, D and F are amplification of square area in E. D, renal tubule region; F, graft region.
B. Zhang et al. / Biomaterials 34 (2013) 5843e5852
5849
Fig. 9. Representative MR images of the insulinoma-bearing mice 24 h after intravenous administration of saline, non-targeted PEG-SPIO or (PEG)SPIO-exendin4.
with different contrast agent. Iron was proved only in the targeting tumor sections but not in non-targeting sections (Fig. 10). 3.5. Biodistribution and biocompatibility of PEG-SPIO and (PEG) SPIO-exendin4 in vivo We next investigated the biodistribution of PEG-SPIO and (PEG) SPIO-exendin4 in ICR mice strain. Nanoparticles were injected into mice via tail vein (iron content 3 mg/kg). At 24 h post-injection, mice were sacrificed. As shown in Fig. 11, iron was accumulated in liver, spleen, lung and kidney in non-targeted agent injection group, whereas it was absent in pancreatic (both exocrine and endocrine) tissues. In mice injected with (PEG)SPIO-exendin4, nanoparticles were unavoidably captured and accumulated in liver and spleen (Fig. 12). But notably, we observed the presence of iron in pancreatic tissue. In some areas, scattered iron deposited around
or in islets (Fig. 12). These findings manifest the beta-cell-targeted properties of exendin-4-conjugated nanoparticles in vivo. In addition, we performed histological analysis on various organs (liver, spleen, lung, and kidney) of mice injected with (PEG) SPIO-exendin4 to observe the probable acute toxicity reaction. There were no pathological changes in sections of above organs (data not show). Furthermore, there was no significant difference on blood cell panels and ALT/AST levels between (PEG)SPIO-exendin4-injected mice and untreated control (Table 1). 4. Discussion GLP-1 is an incretin hormone, which regulates glucose levels through binding to a G protein-coupled receptor (GLP-1R) and stimulating glucose dependent insulin secretion or biosynthesis, inducing glucagon secretion suppress, gastric emptying delay and
Fig. 10. PB staining of tumor sections from mice injected with PEG-SPIO or (PEG)SPIO-exendin4. A,B, PEG-SPIO-injected group; C,D, (PEG)SPIO-exendin4-injected group. B and D are the amplification of the region of interest (ROI) in A and C, and are counter-stained with nuclear fast red. Images confirm the deposits of oxide iron nanoparticles in targeted tumor tissues.
5850
B. Zhang et al. / Biomaterials 34 (2013) 5843e5852
Fig. 11. Histological analysis of biodistribution of PEG-SPIO. Organs were collected at 24 h post-injection. PB staining displayed that iron deposited in liver, kidney, spleen and lung, but not in pancreas (both exocrine and endocrine tissues).
Fig. 12. Biodistribution studies of (PEG)SPIO-exendin4. Nanoparticles were inevitably accumulated in liver and spleen. The presence of iron in pancreatic tissue and around/within islet were observed. Islets were circled and the deposited iron was marked by arrow.
B. Zhang et al. / Biomaterials 34 (2013) 5843e5852 Table 1 Serum liver enzyme levels and blood cell panels for mice administered with (PEG) SPIO-exendin4 and for mice receiving no injection. SPIO-PEG-exendin4 group ALT (U/l) AST (U/l) WBC (103/ml) RBC (106/ml) PLT (103/ml) Hb (g/dl)
27.67 72.67 3.47 8.31 1129.67 14.5
5.51 11.59 2.11 0.16 55.77 0.17
Untreated group 28.33 82 2.4 6.97 1144.33 12.7
4.51 6.24 0.52 1.17 219.81 1.42
promoting satiety [19]. GLP-1R is highly overexpressed in human insulinomas, gastrinomas and islets [20]. In islets, GLP-1R immunoreactivity is reported to be restricted to endocrine beta cells. Therefore, the ligand of GLP-1R could be an ideal probe for pancreatic beta-cell imaging. However, due to rapid degradation by the enzyme dipeptidyl peptidase-4 in the circulation, the half life of GLP-1 is very short, which makes it not suitable for serving as a conjugated substrate [21]. Exendin-4 is a GLP-1 analog, which shares 53% homology, displays similar biological properties and is much more stable for targeting imaging than GLP-1 metabolically. To date, there have been several reports of using radionuclidelabeled exendin-4 or exendin-3 for GLP-1R targeted imaging. Researchers developed 111In, 123I, 125I, 99mTc-exendin-4 conjugate for targeting GLP-1 receptor-positive tumors imaging and biodistribution studies by SPECT system, or using 18F, 68Ga-labeled exendin-4 for PET imaging. Besides of exendin-4, some researchers use 68Ga to label exendin-3 for PET imaging of GLP-1R-expressing insulinomas [22]. Interestingly, Mukai et al. [23] observed that 125 I-conjugated exendin (9e39), a truncated version of exendin with high affinity for the recpetor, could also achieve GLP-1R targeted imaging with the pancreas being the second organ possessing highest level of radioactivity. In the present study, we found that exendin-4-conjugated SPIO, compared with PEG-SPIO, could label both living and fixed beta cells more specifically, which was probably attributed to the ligandereceptor binding. With the ideal labeling efficiency, (PEG) SPIO-exendin4 might be serving as a good beta cell/islet-labeling contrast agent that lower iron concentration achieve superior detective sensitivity. To confirm the specificity properties of exendin4-based nanoparticles to beta cells, we prepared exendin4functionlized quantum dots to label INS-1 cells. Similarly, QDexendin4 displayed higher labeling efficiency in cells, either alive or fixed. These results suggest that efficient internalization of the ligand/receptor complex increased the concentration of the nanoparticles in beta cells, which possibly enhanced the effect of following MR/optical imaging. The ultimate goal of targeting beta-cell imaging is tracking the beta-cell grafts and visualizing pancreatic islets in situ by imaging technique after system administration of special contrast agent. Very recently, Wu et al. [24] used a PET probe-18F-labeled exendin4 to non-invasively evaluate islet graft mass in rodent intrahepatic transplantation model. In our study, it is interesting to note that our (PEG)SPIO-exendin4 could specifically bind to beta-cell grafts and islet in renal (containing grafts) and pancreatic slices. The differential binding to beta cells and the surrounding tissues by (PEG) SPIO-exendin4 highlights its potential role in tracking and imaging beta-cell grafts/islet in vivo by MRI. Another application of GLP-1R-targeted probe is to detect (GLP1R-positive) insulinoma. In clinical settings, the majority of insulinomas are presented as intrapancreatic, scattered and multiple small-sized (<2 cm) tumors. Especially, about 10e27% insulinomas are occult [25]. These make great challenges for accurate tumor localization (both primary and metastatic lesions) based on current
5851
imaging techniques if without specific targeting substrates. Receptor-targeted molecular imaging provides the possibility of disease diagnosis in above conditions and subsequent targeting therapy. Sowa-Staszczak et al. [8], for the first time, utilized 99mTClabeled exendin-4 to localize small insulinoma tumors in 11 patients, highlighting its promising application in clinical practice. In our insulinoma model, (PEG)SPIO-exendin4 also displayed tumortargeted properties evidenced by MR imaging and PB staining. It is worth further exploration that whether (PEG)SPIO-exendin4 could be serving as a GLP-1R-positive insulinoma specific diagnostic agent for clinical application. In our experiments, despite that (PEG)SPIO-exendin4 was unavoidably captured by reticulo-endothelial system (RES) in liver and spleen during biodistribution study, we are delighted to find that iron particles started to appear in and around islets in vivo. The improvement and modification of SPIO to avoid excessive consumption by RES in liver and spleen might provide opportunity for pancreatic islets to uptake more iron nanoparticles and to generate better signal. 5. Conclusion In summary, we here present the evidence that GLP-1 analog functionalized iron oxide nanoparticle probe is capable of targeting GLP-1 receptor-expressed beta cells and insulinoma in vitro and enable to achieve receptor-targeted MR imaging of the insulinoma in vivo. These results manifest that (PEG)SPIO-exendin4 might be served as a promising diagnostic probe for molecular targeting MR imaging of insulinoma in clinics. In addition, higher labeling efficiency of (PEG)SPIO-exendin4 to beta cells suggested that it might be an ideal beta-cell labeling agent. The properties of targeting to in situ beta cells also indicated its potential application in direct visualization of native islets in the near future. Acknowledgment This study was supported by grants from the National Natural Science Foundation of China (nos. 81100549, 81172158, 81001094, and 81272332) and the Ministry of Science and Technology of People’s Republic of China (no. 2007AA02Z476). Appendix ASupplementary data Supplementary data related to this article can be found at http:// dx.doi.org/10.1016/j.biomaterials.2013.04.021. References [1] Malaisse WJ, Louchami K, Sener A. Noninvasive imaging of pancreatic beta cells. Nat Rev Endocrinol 2009;5:394e400. [2] Oh CS, Kohanim S, Kong FL, Song HC, Huynh N, Mendez R, et al. Sulfonylurea receptor as a target for molecular imaging of pancreas beta cells with (99m) Tc-DTPA-glipizide. Ann Nucl Med 2012;26:253e61. [3] Clark PB, Gage HD, Brown-Proctor C, Buchheimer N, Calles-Escandon J, Mach RH, et al. Neurofunctional imaging of the pancreas utilizing the cholinergic PET radioligand [18F]4-fluorobenzyltrozamicol. Eur J Nucl Med Mol Imaging 2004;31:258e60. [4] Otonkoski T, Nanto-Salonen K, Seppanen M, Veijola R, Huopio H, Hussain K, et al. Noninvasive diagnosis of focal hyperinsulinism of infancy with [18F]DOPA positron emission tomography. Diabetes 2006;55:13e8. [5] Goland R, Freeby M, Parsey R, Saisho Y, Kumar D, Simpson N, et al. 11Cdihydrotetrabenazine PET of the pancreas in subjects with long-standing type 1 diabetes and in healthy controls. J Nucl Med 2009;50:382e9. [6] Ueberberg S, Meier JJ, Waengler C, Schechinger W, Dietrich JW, Tannapfel A, et al. Generation of novel single-chain antibodies by phage-display technology to direct imaging agents highly selective to pancreatic beta- or alpha-cells in vivo. Diabetes 2009;58:2324e34. [7] Zhang Y, Chen W. Radiolabeled glucagon-like peptide-1 analogues: a new pancreatic beta-cell imaging agent. Nucl Med Commun 2012;33:223e7.
5852
B. Zhang et al. / Biomaterials 34 (2013) 5843e5852
[8] Sowa-Staszczak A, Pach D, Mikolajczak R, Macke H, Jabrocka-Hybel A, Stefanska A, et al. Glucagon-like peptide-1 receptor imaging with [Lys(40)(Ahx-HYNIC- (99m)Tc/EDDA)NH 2]-exendin-4 for the detection of insulinoma. Eur J Nucl Med Mol Imaging 2013;40:524e31. [9] Kiesewetter DO, Gao H, Ma Y, Niu G, Quan Q, Guo N, et al. 18F-radiolabeled analogs of exendin-4 for PET imaging of GLP-1 in insulinoma. Eur J Nucl Med Mol Imaging 2012;39:463e73. [10] Wild D, Behe M, Wicki A, Storch D, Waser B, Gotthardt M, et al. [Lys40(AhxDTPA-111In)NH2]exendin-4, a very promising ligand for glucagon-like peptide-1 (GLP-1) receptor targeting. J Nucl Med 2006;47:2025e33. [11] Christ E, Wild D, Forrer F, Brandle M, Sahli R, Clerici T, et al. Glucagon-like peptide-1 receptor imaging for localization of insulinomas. J Clin Endocrinol Metab 2009;94:4398e405. [12] Pattou F, Kerr-Conte J, Wild D. GLP-1-receptor scanning for imaging of human beta cells transplanted in muscle. N Engl J Med 2010;363:1289e90. [13] Rumenapp C, Gleich B, Haase A. Magnetic nanoparticles in magnetic resonance imaging and diagnostics. Pharm Res 2012;29:1165e79. [14] Rosen JE, Chan L, Shieh DB, Gu FX. Iron oxide nanoparticles for targeted cancer imaging and diagnostics. Nanomedicine 2012;8:275e90. [15] Sun C, Du K, Fang C, Bhattarai N, Veiseh O, Kievit F, et al. PEG-mediated synthesis of highly dispersive multifunctional superparamagnetic nanoparticles: their physicochemical properties and function in vivo. ACS Nano 2010;4:2402e10. [16] Huang H, Xie Q, Kang M, Zhang B, Zhang H, Chen J, et al. Labeling transplanted mice islet with polyvinylpyrrolidone coated superparamagnetic iron oxide nanoparticles for in vivo detection by magnetic resonance imaging. Nanotechnology 2009;20:365101.
[17] Zhang B, Jiang B, Chen Y, Huang H, Xie Q, Kang M, et al. Detection of viability of transplanted beta cells labeled with a novel contrast agent polyvinylpyrrolidone-coated superparamagnetic iron oxide nanoparticles by magnetic resonance imaging. Contrast Media Mol Imaging 2012;7: 35e44. [18] Tai JH, Foster P, Rosales A, Feng B, Hasilo C, Martinez V, et al. Imaging islets labeled with magnetic nanoparticles at 1.5 Tesla. Diabetes 2006;55:2931e8. [19] Holst JJ, Deacon CF, Vilsboll T, Krarup T, Madsbad S. Glucagon-like peptide-1, glucose homeostasis and diabetes. Trends Mol Med 2008;14:161e8. [20] Kiesewetter DO, Guo N, Guo J, Gao H, Zhu L, Ma Y, et al. Evaluation of an [(18) F]AlF-NOTA Analog of Exendin-4 for imaging of GLP-1 receptor in insulinoma. Theranostics 2012;2:999e1009. [21] Deacon CF, Johnsen AH, Holst JJ. Degradation of glucagon-like peptide-1 by human plasma in vitro yields an N-terminally truncated peptide that is a major endogenous metabolite in vivo. J Clin Endocrinol Metab 1995;80: 952e7. [22] Brom M, Oyen WJ, Joosten L, Gotthardt M, Boerman OC. 68Ga-labelled exendin-3, a new agent for the detection of insulinomas with PET. Eur J Nucl Med Mol Imaging 2010;37:1345e55. [23] Mukai E, Toyoda K, Kimura H, Kawashima H, Fujimoto H, Ueda M, et al. GLP-1 receptor antagonist as a potential probe for pancreatic beta-cell imaging. Biochem Biophys Res Commun 2009;389:523e6. [24] Wu Z, Liu S, Hassink M, Nair I, Park R, Li L, et al. Development and evaluation of 18F-TTCO-Cys40-Exendin-4: a PET probe for imaging transplanted islets. J Nucl Med 2013;54:244e51. [25] Abboud B, Boujaoude J. Occult sporadic insulinoma: localization and surgical strategy. World J Gastroenterol 2008;14:657e65.