Surgery remains the primary treatment option for most solid malignant tumors [1
]. Precise and complete resection of the whole tumor without unnecessary removal of the neighboring healthy tissue is a prerequisite for a successful outcome [2
]. Unfortunately, visual distinction between the malignant and healthy tissue using only the surgeon’s naked eye is often almost impossible. Therefore, development of a tumor-specific marker that would visualize the tumor margins is highly desirable.
Radical resection with adequate negative margins is one of the most important factors influencing the prognoses of patients with head and neck squamous cell carcinoma (HNSCC). Traditionally, the safe margin is defined as greater than 5.0 mm, but similar outcomes are possible with a surgical margin of more than 2.2–3.0 mm [3
Optical imaging could be used for precise identification of the tumor margins. Indeed, narrow-band imaging (NBI) is already used in clinical practice. This horizontal imaging method is based on the contrast between pathological and healthy microvascular architecture [5
]. For evaluation of the deeper tissue layers, vertical diagnostic methods, such as confocal or high resolution endomicroscopy [6
], could be used.
Recently, a novel strategy to highlight tumor tissue, mainly tumor margins, using actively and passively targeted fluorescent nanoprobes, was described [7
]. Intraoperatively, fluorescence intensity could guide the surgeon to resect tumor tissue together with a safe margin. Epidermal growth factor receptor (EGFR) is known to be overexpressed on the cell surface of various types of malignant tumors, including breast and lung adenocarcinomas (40% of cases) [8
], anal cancers [10
], glioblastoma (50%) [11
] and epithelial tumors of the head and neck (80–100%) [12
]. Recently, we reported N
-(2-hydroxypropyl)methacrylamide (HPMA)-based polymer probes for fluorescence-guided surgery targeted to EGFR with the oligopeptide YHWYGYTPQNVI (GE-11) [13
]. It was demonstrated that these EGFR-targeted polymer nanoprobes accumulated to a higher extent in EGFR-positive cells in vitro when compared to the non-targeted control polymer. Similarly, EGFR-targeted nanoprobes were significantly accumulated in EGFR-positive tumors in vivo, thus showing the future potential of these polymer nanoprobes within fluorescence-navigated surgery. Moreover, the EGFR-targeted polymer nanoprobe demonstrated a significant signal in the margin of the tumors, thus clearly showing the benefit of the nanoprobe for the precise navigation of surgeons during solid-tumor removal.
The present study compared several HPMA-based fluorescent probes for EGFR-specific labeling of hypopharyngeal carcinoma cells (FaDu) and breast adenocarcinoma cells (MDA-MB-231) [14
]. Besides the already mentioned GE-11 targeting oligopeptide designed, prepared and evaluated, polymer probes were also targeted with human epidermal growth factor (EGF), which is the natural ligand for EGFR, and with a monoclonal antibody cetuximab, which is a clinically approved EGFR inhibitor distributed under the trade name Erbitux. EGFR-specific cell binding of all the targeted polymer probes in vitro was evaluated using flow cytometry; fluorescent visualization of EGFR-positive tumors in vivo was performed using an in vivo imaging system in-Vivo Xtreme (Bruker).
In parallel with the actively targeted polymer probes, we also investigated high molecular weight fluorescent polymers without the targeting ligands described above. These polymers passively accumulate within the tumor tissue due to the so-called enhanced permeability and retention (EPR) effect [16
], which is based on a leaky neovasculature and impaired or missing lymphatic drainage in the tumor.
The potential of the actively and passively targeted polymer probes for the fluorescence-guided surgery of malignant tumors was compared and discussed. Targeted polymeric fluorescent nanoprobes have the potential to increase the safety of oncological surgery in the upper aerodigestive tract and to improve the overall prognosis of patients.
2. Materials and Methods
1-Aminopropan-2-ol, 2,2′-azobis-(isobutyronitrile) (AIBN), N,N′-diisopropylcarbodiimide (DIC), N,N-dimethylacetamide (DMA), N,N-dimethylformamide (DMF), ethyldiisopropylamine (DIPEA), dithiothreitol (DTT), 1-hydroxybenzotriazole (HOBt), methacryloyl chloride, piperidine, trifluoroacetic acid (TFA), triisopropylsilane (TIPS), and all other reagents and solvents were purchased from Sigma-Aldrich (Prague, Czech Republic). TentaGel Rink amide resin, ethyl cyano(hydroxyimino)acetate (Oxyma), (benzotriazol-1-yloxy)-trispyrrolidinophosphoniumhexafluorophosphate (PyBOP), 9-fluorenylmethoxycarbonyl (Fmoc)-amino acid derivatives and 1-(9-fluorenylmethyloxycarbonyl)amino-3,6,9,12,15,18,21,24,27,30,33,36-dodecaoxanonatriacontan-39-oic acid (Fmoc-Peg12-COOH) were purchased from Iris Biotech GmbH (Marktredwitz, Germany). 5-Azidopentanoic acid was obtained from Bachem (Bubendorf, Schwitzerland) and amino-1-(11,12-didehydrodibenzo[b,f]azocin-5(6H)-yl)propan-1-one (Dbco-NH2) was purchased from Click Chemistry Tools (Scottsdale, AZ, USA). Amino and NHS-ester derivatives of fluorescent dyes Cyanine 7 (Cy7-NH2, Cy7-NHS) and Dyomics-633 (Dy-633-NH2, Dy-633-NHS) were obtained from Lumiprobe GmbH (Hannover, Germany) and Dyomics GmbH (Jena, Germany). All amino acids were L-configuration. Human EGF (hEGF) was purchased from Biovision Inc. (Milpitas, CA, USA). Methacryloyl chloride, 1-aminopropan-2-ol, and dichloromethane were distilled immediately before use. All chemicals and solvents were of analytical grade. Solvents were purified and dried using standard procedures.
2.2. Physico-Chemical Characterization
Monitoring of the peptide purity and conjugation of the peptide to the reactive copolymer were performed by high-performance liquid chromatography (HPLC) using a Chromolith Performance RP-18e column (100 × 4.6 mm, Merck, Gernsheim, Germany), and a linear gradient of water–acetonitrile, 0%−100% acetonitrile, in the presence of 0.1% TFA with a UV-vis diode array detector (Shimadzu, Kyoto, Japan). Determination of the molecular weight and dispersity of the copolymers was performed by size exclusion chromatography (SEC) on a HPLC system (Shimadzu, Kyoto, Japan) equipped with refractive index, UV, and multiangle light scattering DAWN 8 EOS (Wyatt Technology Corp., Santa Barbara, CA, USA) detector using a TSK 3000 SWXL column (Tosoh Bioscience, Kyoto, Japan) and 80% methanol, 20% 0.3 M acetate buffer pH 6.5 at a flow rate of 0.5 mL/min. The calculation of molecular weights from the light scattering detector was based on the known injected mass assuming 100% mass recovery. The molecular weight and dispersity values of the fluorescently labeled polymers were calculated from SEC chromatograms based on pHPMA calibration, using RI detector data. The content of thiazolidine-2-thione (TT) groups was determined spectrophotometrically on a Helios Alpha UV/vis spectrophotometer (Thermo Fisher Scientific, Waltham, MA, USA) using the molar absorption coefficient for TT in methanol, ε305 = 10,280 L mol−1 cm−1. Determination of dyes Dy-633 or Cy7 was also determined spectrophotometrically using the molar absorption coefficient for Dy-633 (Cy7) in methanol, ε637 = 159,000 L mol−1 cm−1 (ε750 = 199,000 L mol−1 cm−1).
2.3. Synthesis of Monomers and Polymer Precursors
Monomers and amino-reactive copolymer precursor poly(HPMA-co
-Ma-β-Ala-TT) (P-TT) were prepared as described previously [13
]. Reactive polymer with dye Dy-633-NH2 or with Cy7-NH2 and control polymers with the dye and without targeting peptides (P-Dy-633 and P-Cy7) were prepared as described previously [13
2.4. Synthesis of Peptide-Targeted Nanoprobes
The peptide-targeted nanoprobes were prepared as described earlier [13
] to form the polymer conjugates P-GE11-Dy-633 or P-GE11-Cy7. Similarly, the scrambled control was prepared to form the conjugate P-scrGE11-Cy7 [13
]. The amino acid analysis was used for the oligopeptide content determination.
2.5. Synthesis of hEGF-Targeted Nanoprobes
HEGF (2 mg) was added to 0.5 mL of 0.03 M phosphate buffer pH 8 to form a suspension and 4 mg of the polymer precursor P-TT-Dy-633 or P-TT-Cy7 was dissolved in 80 µL of DMA. The solution was added dropwise to the suspension of hEGF and vortexed overnight. The next day, the solution was clear, and the completeness of the reaction was verified by HPLC. The reaction mixture was chromatographed on Sephadex G 25 resin in water (PD 10 column, GE Healthcare). The polymeric fraction was freeze-dried. The molecular characteristics of conjugates P-EGF-Dy-633 and P-EGF-Cy7 are shown in Table 1
. The amino acid analysis was used for the peptide content determination.
2.6. Synthesis of Monoclonal Antibody (mAb)-Targeted Nanoprobe
Radical copolymerization in dimethylsolfoxide (DMSO) was applied for the synthesis of semitelechelic polymer precursor (polymer sP-TT, Table 1
) containing Boc-protected hydrazide groups. Functionalized initiator 3,3’-[azobis(4-cyano-4-methyl-1-oxobutane-4,1-diyl)] bis(thiazolidine-2-thione) was employed as described earlier [18
]. The end-chain reactive maleimide (MI) group was introduced into copolymer sP-TT by the reaction of TT group with N-(2-aminoethyl)maleimide, as described previously [19
DY-633 and Cy7 containing polymers, sP-DY-633-MI and sP-Cy7-MI, were prepared from sP-MI by the reactions of the hydrazide groups of the polymeric precursor with NHS-esters of the corresponding dyes. Briefly, the polymer precursor sP-MI (50 mg) and DY-633-NHS ester (2 mg, 2 µmol) were dissolved in 0.5 mL of DMA and the solutions were mixed together. The reaction was monitored by TLC (eluent: methanol:ethyl acetate:acetic acid 10:4:0.5): RfDY-633-NHS ester = 0.84, Rfpolymer = 0. After 2 h at room temperature, TLC showed ca 85% yield and the low-molecular weight impurities were removed by gel filtration (Sephadex LH-20, solvent methanol). The purified polymer probe was isolated by precipitation in ethyl acetate. The mAb targeted nanoprobe was prepared in two consecutive steps as described elsewhere [19
]. First, the mAb, cetuximab, was mildly reduced by DTT to introduce sulfanyl groups. Then, the reduced mAb was reacted with semitelechelic copolymers containing MI groups, P-DY-633-MI or P-Cy7-MI, to form the star-like mAb-polymer-dye conjugate. The product was characterized using SEC, UV-VIS spectrophotometry and electrophoresis. The amino acid analysis was used for the mAb content determination.
2.7. Synthesis of Star Polymers
The star polymer precursors St-P-1 and St-P-2 were synthesized by grafting the thiazolidine-2-thione (TT)-terminated semitelechelic HPMA copolymer sP-TT onto the 2nd or 4th generation polyamidoamine (PAMAM) dendrimers containing terminal amino groups, as described in [20
]. Briefly, semitelechelic copolymer precursor sP-TT with chain-terminated TT groups (32 mg; 0.003 mmol TT groups) was dissolved in 1 mL of methanol and added into a stirring solution of 0.8 mg of D-NH2 (G2, diaminobutane core, 16 amino groups; 0.002 mmol of dendrimer), or 0.4 mg of D-NH2 (G4, diaminobutane core, 64 amino groups; 0.001 mmol of dendrimer), respectively, in 0.4 mL of methanol. After 2 h, the reaction was terminated by adding 5 μL of 1-aminopropan-2-ol and the remaining free amino groups of the dendrimer were end-capped by reaction with acetic anhydride. The star precursor was freed of low molecular weight impurities by gel filtration (Sephadex LH-20, solvent methanol) and subsequently isolated by precipitation in ethyl acetate. TFA was used for the deprotection of Boc-protected hydrazide groups.
Star polymer conjugates St-P-1-Cy7 and St-P-2-Cy7 (Table 1
) bearing fluorescent dye Cy7, attached via a hydrazide bond, were prepared by the reaction of star polymer precursors with Cy7-NHS ester, as described above. Reference polymer sP-Cy7 was prepared from the sP-TT precursor after the removal of TT groups with 1-aminopropan-2-ol, deprotection and attachment of Cy7 by the procedure described above.
2.8. Cell Culture
The cell lines breast carcinoma MDA-MB-231 and HNSC FaDu were purchased from American Type Culture Collection (ATCC, Manassas, VA, USA). MDA-MB-231 was cultured in a mixture of Dulbecco’s Modified Eagle Medium (DMEM, Gibco, Gaithersburg, MD, USA) and Roswell Park Memorial Institute (RPMI) 1640 medium (1:1, Gibco), supplemented with 10% fetal bovine serum (FBS, Sigma-Aldrich, Prague, Czech Republic) and 1% penicillin–streptomycin (Gibco, Waltham, MA USA). FaDu cells were grown in Minimum Essential Medium Eagle (Sigma-Aldrich, Prague, Czech Republic), enriched with 2 mM L-glutamine (Gibco, Waltham, MA USA), 1% (v/v) non-essential amino acids (NEAA) solution (Gibco, Waltham, MA USA), 10% FBS (Sigma-Aldrich) and 1% penicillin-streptomycin (Gibco, Waltham, MA USA). Cell lines were maintained at 37 °C in a 5% CO2 humidified atmosphere. FaDu cell line, expressing Red Fluorescent Protein (RFP), was a kind gift from Dr. Trzil (BIOPHARM, Research Institute of Biopharmacy and Veterinary Drugs, Liběchov, Czech Republic). All experiments were conducted on cells in passage less than five.
2.9. Flow Cytometry
Cells were harvested in 100 mM HEPES buffer with 20 mM NaCl, 10 mM EDTA and 0.5% bovine serum albumin (BSA, Sigma), pH 7.4 (MDA-MB-231) or 0.05% trypsin-EDTA (Gibco; FaDu). Collected cells were centrifuged (1500 rpm, 3 min) and washed once with PBS containing 0.5% BSA (PBS-BSA). After washing, 5 × 104 cells/vial were incubated with the appropriate concentration of tested conjugate or phycoerythrin (PE) conjugated anti-EGFR monoclonal antibody (1:10, Exbio, Prague, Czech Republic) for 1 h at 4 °C in the dark. Afterwards, cells were washed once from unbounded conjugate with PBS-BSA, centrifuged and mixed with Sytox Blue Dead Cell Stain (Thermo Fisher Scientific, Waltham, MA, USA) to distinguish between live and dead cells. Data acquisition was performed using a FACSVerse (Becton Dickinson, Franklin Lakes, NJ, USA) with subsequent analysis using FlowJo software version 10 (Tree Star Inc., Ashland, OR, USA).
2.10. Statistical Analysis
All data are presented as mean ± SEM. The values were determined from at least two experiments performed in duplicate. Saturation binding data were analyzed by GraphPad Prism 5 software, ver. 5.1 (GraphPad Software Inc., San Diego, CA, USA) using non-linear regression curve fit (one-site binding model). Kd values were calculated after subtraction of the control conjugate (non-specific binding), as the targeting conjugate concentration occupies 50% of the receptors at equilibrium.
For in vivo experiments, one-way ANOVA and a post hoc procedure consisting of multiple two-sided t-tests with the Holm method to control the family-wise error rate were performed using statistical language R. P values lower than 0.05 were considered statistically significant.
2.11. Immunofluorescence Analysis of Cells
Cells in microscopic chambers were fixed with buffered 4% formaldehyde for 15 min and washed in PBS three times. Subsequently, cells were permeabilized by 0.3% Triton X-100 in PBS for 15 min and blocked by 5% BSA in PBS for 30 min. Samples were incubated with primary antibody for 60 min at ambient temperature, then 60 min with secondary antibody at ambient temperature. Finally, 25 µL of Vectashield® antifade medium with DAPI (Vector Laboratories, Inc., 30 Ingold Road, Burlingame, CA 94010, USA) was added to each cell chamber for nuclear staining.
2.12. Immunofluorescence Analysis of Tissues
Tumor tissues obtained from the nude mice or human subjects were fixed with buffered 4% formaldehyde for 24 h, dehydrated, embedded in paraffin blocks and cut into 5 µm slides. Subsequently, the tissue slides were deparaffinized, and rehydrated. Antigen retrieval was performed by boiling in sodium citrate buffer (pH 6.0) in a microwave oven for 15 min. Samples were then incubated with primary antibody for 60 min at ambient temperature and with secondary antibody for 60 min. A droplet of Vectashield® antifade medium with DAPI (Vector Laboratories, Inc., 30 Ingold Road, Burlingame, CA, USA) was added per slide for nuclear staining before mounting, as per manufacturer’s protocol.
Primary antibodies anti-EGFR rabbit polyclonal antibody (PA11110, Thermo Fisher Scientific, Pierce Biotechnology, Rockford, IL, USA), dilution 1:200, and anti-human cytokeratin mouse monoclonal antibody AE1/AE3 (M3515, DAKO Denmark A/S, Glostrup, Denmark), dilution 1:200. Secondary antibodies donkey anti-mouse IgG (H+L) highly cross-adsorbed secondary antibody, Alexa Fluor 680 (A10038, Invitrogen, Paisley, UK), dilution 1:1000 and donkey anti-rabbit IgG (H+L) highly cross-adsorbed secondary antibody, Alexa Fluor 488 (A-21206, Invitrogen), dilution 1:1000.
2.14. Fluorescence Microscopy
The confocal Leica TCS SP5 microscope (Leica Microsystems, Wetzlar, Germany) was used for imaging. Alexa Fluor 488 was excited using 488 nm argon laser line, Alexa Fluor 680 was excited using 633 nm HeNe laser line. DAPI was excited using 405 nm laser. Open source FIJI ImageJ was utilized for the processing of obtained images [21
2.15. Intravital Tumor Accumulation Assessment
All animal experiments were performed in accordance with the Act on Experimental Work with Animals (Public Notice of the Ministry of Agriculture of the Czech Republic No. 246/1992, No. 311/1997, No. 207/2004; Decree of the Ministry of the Environment of the Czech Republic No. 117/1987; and Act of the Czech National Assembly No. 149/2004) of the Czech Republic, which is fully compatible with the corresponding European Union directives. Animal study protocol, covering all hereafter described animal experiments, was approved by the Czech Ministry of Education Youth and Sports (Approval No. MSMT-12181/2016-4, date of approval: 15 May 2016).
All patients signed informed consent before entering the study and the study protocol was approved by the Central Ethics Committee of Hospital Motol, Prague, Czech Republic (Approval No. AZV16-28594A, date of approval: 24 June 2015). In addition, all data were analyzed with respect to patient privacy.
FaDu-RFP cells, where RFP means red fluorescent protein, were trypsinized, spun and re-suspended in an FBS-free medium at a concentration of 2 × 107 cells per mL. BD Matrigel™ (I.T.A.-Intertact, Ltd., Prague, Czech Republic) was added (1/2 of the cell suspension volume) and 0.15 mL of the mixture was subcutaneously administered (2 × 106 cells per mouse) into the abdominal right flank of athymic nude mice (Velaz, Ltd. and Charles River Laboratories International, Inc., Prague, Czech Republic). When the tumors had reached a minimum size of 6 mm in diameter, mice were divided into groups of eight and randomly allocated for individual intravenous conjugate administration. In case of conjugates sP-Cy7, St-P-1-Cy7 and St-P-2-Cy7, 0.25 mg of probes per mouse were administered, dissolved in 0.1 mL. For the experiment with actively targeted conjugates mAb-P-Cy7, P-scrGE11-Cy7, P-GE11-Cy7 and P-EGF-Cy7, the amount of conjugates was unified based on the Cy7 dye content, as these conjugates had uneven Cy7 content. Tumor accumulation and the biodistribution of polymeric carriers conjugated to Cy7 were determined using the Xtreme In Vivo Imaging System (Bruker BioSpin, Ettlingen, Germany) by intravital imaging 15 min, 4 and 24 h after administration, as described in [13
]. Images were taken separately for both RFP and Cy7 fluorescent channel (exposition time of 5 s for both channels), and the reflectance images were acquired. The light source was Xenon Arc Lamp with 400 W power. Band filters were used as excitation emission filters to pass a single wavelength of + - about 5 nm. The camera has a CCD sensor and is cooled to −65 °C. Open source software FIJI was utilized for the adjustment of obtained images and quantification of ROIs [21