Preparation and Biofunctionalization of Multicolor Conjugated Polymer Nanoparticles for Imaging and Detection of Tumor Cells

Abstract

Multicolor conjugated polymer nano­particles (CPNs) are designed and biofunctionalized with antibody for targeted cell imaging and cell detection. In comparison to single-antibody recognition mode to tumor cells, enhanced specificity for targeted imaging and detection of tumor cells is achieved by binding two antibody functionalized CPNs to one tumor cell. This strategy provides a new platform for designing CPNs useful for targeted imaging, cells detection and other biological applications. Conjugated polymer nanoparticles (CPNs) combine the properties of conjugated polymers (CPs) and nanoparticles and define a new class of promising fluorescence materials that are being integrated in the interdisciplinary fields of materials chemistry, biology and medicine.1-5 Compared with conventional organic dyes and inorganic semiconducting quantum dots, CPNs display several distinguishing features, such as high brightness, excellent photostability, low cytotoxicity, large absorption cross sections, and versatile surface modification.6-11 Although water-soluble CPs have been widely studied in biological applications, obstacles remain for modification with ligands (such as antibodies, DNA, biotin, folic acid, and drugs) due to challenging conjugation processes and difficult separation/purification steps. CPNs can more easily be modified with biomolecules according to different requirements and have thus been recently applied in fluorescence imaging, drug/gene deliver, anticancer and anti-microorganism activity.12-22 For the targeted imaging and detection of tumor cells, the specificity for the reported CPN systems remains unsatisfied due to the single antibody recognition mode to tumor cells. For example, single antibody-labeled CPNs can target various tumor cells expressing the same antigen. Developing multiplexed imaging has become important in cell counting, cell sorting, and clinical diagnostics. Fluorescence resonance energy transfer (FRET) cascade techniques can be used to tune emission signatures under a single excitation wavelength.23-25 One way to achieve multicolor imaging is to prepare CPNs with three different colors (blue, green and red). The resulting emission characteristics will then be influenced by changing the choice, amount and ratio within each CPN.7, 26 Additionally, as single multicolor CPNs can be tuned to match the different excitation sources of commercial optical instruments, such as fluorescence spectrometers, fluorescence microscopes, and in flow cytometry. In this work, we describe the design and synthesis of four conjugated polymers with blue, green, yellow and red emissions, and used them to prepare carboxyl functionalized CPNs by a co-precipitation method based on hydrophobic interactions between the conjugated polymers and poly(styrene-co-maleic anhydride) (PSMA).6, 27 The carboxyl functionalized CPNs can also be prepared by co-precipitation of four conjugated polymers with PSMA, which show multicolor emissions by under single excitation wavelength. CPNs were modified with primary antibody to obtain CPNs-antibody conjugates. In comparison to single antibody recognition, higher specificity for tumor cells detection was achieved by binding two CPNs labeled with different antibodies to a single tumor cell. Finally, we show that the multicolor CPNs can match different excitation sources of fluorescence microscopy and flow cytometry to achieve straightforward cell imaging and detection. Four π-conjugated polymers (P1 ∼ P4) were obtained by Suzuki cross-coupling polymerization of monomers 1–4 with monomer 517, 28 in yields of 15∼48% (Scheme 1a). Absorption and fluorescence characteristics were determined in CHCl3 and show that each polymer spans a different region of the visible spectrum (Table S1). Compared with the corresponding monomers, the absorption spectra of P1∼P4 (Figure S1) not only have characteristic absorptions of monomers but also exhibit new combination absorptions with the increase of aromatic heterocyclic units. In these polymers, the absorptions in shorter wavelength region originate from the fluorene units, and the longer absorptions lie in the signal aromatic heterocyclic unit or cooperative actions of different monomers. The emission spectra show maxima for P1, P2, P3 and P4 are 422, 500, 540 and 670 nm, respectively. Emission quantum yields (QY) vary depending on the electron acceptor unit in the polymer backbones (3 ∼ 78%). Upon excitation at 360 nm, the shorter wavelength-emissive polymers were anticipated to act as the donor for longer wavelength-emissive ones (acceptors), through interchain multi-step FRET. Thus, by varying the mixing ratio of the polymers, multicolor emission can be regulated through FRET among the four polymers under one excitation wavelength. In order to achieve bioconjugation, CPNs are required to provide surface functional groups (such as –COOH, –NH2, –N3, –C≡CH, etc.) for subsequent modification. Herein, such CPNs with carboxyl groups on the surface were prepared by a modified co-precipitation method6, 27 based on hydrophobic interactions of conjugated polymers and PSMA. The surface carboxyl groups can then be modified with antibodies (Scheme 1b-c). For these CPNs, conjugated polymers P1 ∼ P4 are the light-emitting vectors with tunable emission colors from blue to red; and PSMA acts as a coating and a source of carboxyl functionalities. The photophysical properties of these CPNs were determined in water (Figure S2 and Table S1). The absorption and emission spectra of CPNs are similar with corresponding polymers for P1∼ P3. It is noted that, in comparison to P4, blue-shifts of 9 nm and 20 nm take place for absorption and emission maxima of P4/PSMA, respectively, which attributes to a relaxed and larger π-conjugated extension of P4 in organic solvent (CHCl3), relative to the nanoparticles P4/PSMA in water. The size and morphology of functionalized CPNs were characterized by dynamic light scattering (DLS) (Figure 1a) and scanning electron microscopy (SEM) (Figure 1b). As an example, P3/PSMA CPNs exhibit a spherical morphology with an average diameter of ∼35 nm. Photostability measurement shows that the fluorescence of P3/PSMA CPNs retains more than 85% of the original intensity after continuously irradiating at 455 nm for 120 s (Figure 1c). Furthermore, the cytotoxicity of P3/PSMA CPNs was assayed in the dark and under white light (Figure 1d) by using the MTT method, in which the cell viability was depended on the conversion of 3-(4′,5′-dimethylthiazol-2′-yl)-2,5-diphenyl-2H-tetrazolium hydrobromide (MTT) to formazan.29 More than 80% and 75% cell viabilities were obtained in the presence of CPNs in the dark and under white light (10 mW•cm−2, 30 min), respectively. This level of cytotoxicity and the good photostability of P3/PSMA CPNs are desirable characteristics for cell imaging and other biological applications, particularly in view of the cytotoxicity of other imaging reagents such as quantum dots and organic dyes.8, 18 Because CPNs are modified with carboxyl on the surface after co-precipitation with PSMA, the reaction between the amine groups on antibody and the carboxyl groups can afford the desired CPNs-antibody conjugates. In order to improve bioconjugation efficiency, CPNs were initially linked with NHS-SO3Na to activate carboxyl groups. Subsequently, an antibody was added to the mixture and the conjugation reaction was performed. In the reaction system, 0.1–0.3 wt% polyethylene glycol (PEG) was added to avoid nonspecific adsorption between CPNs and antibodies. The successful covalent conjugation of CPNs with three antibodies (anti-EpCAM, anti-ErbB 2 and IgG, respectively) was verified by electrophoresis with a 0.7% agarose gel (Figure S3) and zeta potential (ζ) measurements (Table S2). Zeta potentials of the bioconjugated CPNs have lower negative values than unmodified CPNs, which suggest that carboxyl moieties on the surface of nanoparticles were successfully linked to the amino groups of the antibodies. Furthermore, due to more negative zeta potential, the unmodified CPNs display faster mobility in the gel medium than those of bioconjugated CPNs. Also the average diameters of CPNs increase from about 30 nm to 50 nm upon conjugation of antibody. Taken together, these results indicate that the antibodies are covalently connected to CPNs via amide bond formations. Previous studies have indicated that bare CPNs can be used for cell imaging and that they can locate in the cytoplasm region through endocytosis when CPNs are adhered to the surface of cells by nonspecific interactions (electrostatic or hydrophobic). In order to improve the specificity of targeted cell imaging, we took advantage of the well-known interactions between a primary antibody and a receptor on a cells surface (antigen). An epithelial cell surface receptor (EpCAM) on the surface of live human breast cancer cell (MCF-7) was studied as a cellular target. First, the corresponding antibody (anti-EpCAM) was covalently linked to P3/PSMA CPNs (P3/PSMA/anti-EpCAM). Second, live MCF-7 cells were cultured with P3/PSMA/anti-EpCAM at 37 °C for 30 min. Targeted imaging of the CPNs for MCF-7 was investigated by confocal laser scanning microscopy. As shown in Figure S4a-d, P3/PSMA/anti-EpCAM specifically binds on the surface of MCF-7 cells. As a control experiment, the live MCF-7 cells were also cultured with bare CPNs under the same conditions, and there is no corresponding fluorescent stain with bare CPNs (Figure S4a,b). Besides fluorescence imaging, flow cytometry is another powerful tool for analyzing cells stained with dyes. In the experiment, P3/PSMA/anti-EpCAM was demonstrated for cell analysis using flow cytometry (Figure S4e,f). Much higher intensity counts are observed for MCF-7 cells stained with P3/PSMA/anti-EpCAM (Figure S4g) in comparison to those of MCF-7 cells themselves in control experiments (Figure S4e). Thus the enhanced signal readings must come from efficient staining of cells with P3/PSMA/anti-EpCAM. Based on the flow cytometry data, the targeted efficiency of P3/PSMA/anti-EpCAM to MCF-7 cells reaches to 95% or more. The results of flow cytometry are consistent with fluorescence imaging results. To amplify the imaging signal for tumor cells, a second antibody (IgG) strategy was used by modifying CPNs with IgG. First, MCF-7 cells were cultured with P3/PSMA/anti-EpCAM CPNs at 37 °C for 30 min, followed by culturing with P3/PSMA/IgG for 30 min at 37 °C. Compared with P3/PSMA/anti-EpCAM CPNs only, the imaging signal for MCF-7 cells is significantly enhanced through the second antibody strategy (Figure S5a-b). This amplification effect can be verified by fluorescence spectra taken from images of confocal laser scanning microscopy (Figure S5c). Based on FRET mechanism, multicolor P1–4/PSMA CPNs with entire visible region emission were obtained through co-precipitating P1, P2, P3, P4 and PSMA (see Scheme 1c). In these experiments, the homogeneous solution of P1, P2, P3, P4 and PSMA (2.0 μg/mL of P1, 7.0 μg/mL of P2, 4.0 μg/mL of P3, 12.0 μg/mL of P4, and 20.0 μg/mL of PSMA) in THF was rapidly added to Milli-Q water. The THF in the solution was removed by bubbling nitrogen at room temperature and the nanoparticle dispersion was obtained by concentrating the solution and filtering through a 0.22 micron filter. The photophysical properties of P1–4/PSMA CPNs were determined in water. Their absorption and emission can cover the entire range of visible light (Figure 2a–b). Measurements of DLS and SEM show that P1–4/PSMA CPNs exhibit a spherical morphology with an average diameter of ∼30 nm (Figure S6). P1–4/PSMA CPNs were modified by the antibody anti-EpCAM and were applied to MCF-7 cell imaging. Before the cell imaging experiment, MCF-7 cells were cultured with P1–4/PSMA-EpCAM CPNs at 37 °C for 30 min. As shown in Figure 2c, upon excitation at 405 nm, 488 nm and 559 nm, multicolor fluorescence images (blue, green and red) of P1–4/PSMA CPNs are observed. Thus, one kind of P1–4/PSMA CPNs can be excited by different excitation wavelengths that match well to those available in commercial fluorescence instruments used for bioassays and imaging. Since the detection specificity of single antibody recognition mode-based CPNs for tumor cells is unsatisfied, we designed and constructed a detection ensemble for labeling different antibodies to one tumor cell that is comprised of two CPNs modified with two different antibodies for one tumor cell. In these experiments, P3/PSMA CPNs were separately connected with anti-EpCAM and anti-ErbB2, and then three tumor cells (SK-BR-3, MCF-7 and HeLa) were cultured with the ensemble at 37 °C for 30 min. For anti-EpCAM, its expresses positively on SK-BR-3 and MCF-7 cells and negatively on HeLa cells; while for anti-ErbB2, the expression for SK-BR-3 cells is positive and that of MCF-7 and HeLa cells is negative. As shown in Figure 3, SK-BR-3 cells can be stained by both CPNs, MCF-7 cells can only be stained by P3/PSMA/anti-EpCAM CPNs, and HeLa cells cannot be stained by any either the anti-EpCAM or anti-ErbB2 modified P3/PSMA CPNs. The data are consistent with the immunofluorescence labeling results (Figure S7). Based on the experimental results, we can easily discriminate different tumor cells by different cell imaging. More importantly, the SK-BR-3 cells can be specifically detected with an ensemble with two CPNs, even though SK-BR-3 and MCF-7 cells belong to the same breast cancer cell lines. In comparison to single antibody recognition mode to tumor cells, much better specificity for tumor cell detection was therefore achieved by the double-antibody recognition approach. In summary, the present work introduces a method for targeted cell imaging and specific cell detection based on a new class of detection systems, namely multicolor CPNs bioconjugated with antibodies. Our study shows that the modification strategy through amide coupling of antibodies with carboxyl functionalized CPNs is feasible and efficient, and may be suitable for bioconjugation with proteins, DNA, drug candidates or other molecules. CPNs exhibit good photostability, low cytotoxicity and phototoxicity and can be obtained with an average diameter of ∼30 nm. Compared with primary antibody-modified CPNs only, amplified imaging signal to target tumor cells is realized through the second antibody strategy. The multicolor CPNs with the entire visible emission range were also prepared by simultaneous co-precipitation of four conjugated polymers (P1 ∼ P4) and PSMA, which show multicolor emissions under single excitation wavelength. One kind of P1–4/PSMA CPNs can be excited at multiple wavelengths, which match well to most of the fluorescence instruments (e.g. fluorescence microscopy and flow cytometry) used for cell imaging. In comparison to single-antibody recognition mode to tumor cells, much better specificity for targeted imaging and detection was achieved by binding each tumor cell to two CPNs labeled with different antibodies. This new strategy has the potential to enable application of multicolor CPNs in targeting imaging, cells detection and other biological applications. Preparation of conjugated polymer nanoparticles: Carboxyl group-functionalized CPNs (P1/PSMA ∼ P4/PSMA and P1–4/PSMA) were obtained by a co-precipitation method. For a typical preparation, there are main four steps. First, 10 mL of THF solution with 50 μg/mL CPs and 20 μg/mL PSMA was prepared by stock solutions of 1.0 mg/mL CPs and 2.0 mg/mL PSMA in THF, respectively. Second, the above solution was mixed sufficiently in order to form a homogeneous solution. Then, under sonication, the solution was rapidly added to 20 mL of milli-Q water in an ice bath and the mixture was sequentially sonicated for 5∼10 min. Third, the THF in the solution was removed by bubbling nitrogen at room temperature. Finally, the nanoparticle dispersion was obtained by concentrating the mixture to 10 mL on a 90 °C oil bath followed by filtration through a 0.22 micron filter. Cell imaging in vitro: For cells imaging, there are two methods according to the type of functionalized CPNs with primary antibodies or second antibody (IgG). These CPN bioconjugates were prepared by the above described bioconjugated reaction. Similarly, the cells were cultured until the density reaches about 70% on a confocal dish (Coverglass Bottom Dish). Then the cells were cultured with 2.5 μg/mL CPNs-anti-EpCAM or CPNs-anti-ErbB2 for 30 min at 37 °C, respectively. After washing the cells two times by PBS buffer, the cells were fixed with 4% paraformaldehyde at room temperature for 20 min, followed by another two-washing step. The stained cells were imaged in PBS buffer with fluorescence confocal microscope. For the CPNs-IgG imaging, the whole procedure was the same with the immunofluorescent labeling except that CPNs-IgG was used instead of dye-IgG. The authors are grateful to the National Natural Science Foundation of China (Nos. 21033010, 21371110 and TRR61), the Major Research Plan of China (No. 2011CB932302) and the China Postdoctoral Science Foundation (2013T60177). Work at UCSB is supported through the NSF (DMR 1005546). We also thank Zachary Henson for assistance during the preparation of the manuscript.