Phospholipid-Modified Upconversion Nanoprobe ... - ACS Publications

Jun 18, 2014 - Phospholipase D (PLD) is a critical component of intracellular signal transduction and has been implicated in many important biological...
0 downloads 0 Views 3MB Size
Download PDF
Article pubs.acs.org/ac

Phospholipid-Modified Upconversion Nanoprobe for Ratiometric Fluorescence Detection and Imaging of Phospholipase D in Cell Lysate and in Living Cells Yao Cen, Yan-Mei Wu, Xiang-Juan Kong, Shuang Wu, Ru-Qin Yu, and Xia Chu* State Key Laboratory of Chemo/Bio-Sensing and Chemometrics, College of Chemistry and Chemical Engineering, Hunan University, Changsha 410082, People’s Republic of China S Supporting Information *

ABSTRACT: Phospholipase D (PLD) is a critical component of intracellular signal transduction and has been implicated in many important biological processes. It has been observed that there are abnormalities in PLD expression in many human cancers, and PLD is thus recognized as a potential diagnostic biomarker as well as a target for drug discovery. We report for the first time a phospholipid-modified nanoprobe for ratiometric upconversion fluorescence (UCF) sensing and bioimaging of PLD activity. The nanoprobe can be synthesized by a facile one-step self-assembly of a phospholipid monolayer composed of poly(ethylene glycol) (PEG)ylated phospholipid and rhodamine B-labeled phospholipid on the surface of upconversion nanoparticles (UCNPs) NaYF4: 20%Yb, 2%Er. The fluorescence resonance energy transfer (FRET) process from the UCF emission at 540 nm of the UCNPs to the absorbance of the rhodamine B occurs in the nanoprobe. The PLD-mediated hydrolysis of the phosphodiester bond makes rhodamine B apart from the UCNP surface, leading to the inhibition of FRET. Using the unaffected UCF emission at 655 nm as an internal standard, the nanoprobe can be used for ratiometric UCF detection of PLD activity with high sensitivity and selectivity. The PLD activity in cell lysates is also determined by the nanoprobe, confirming that PLD activity in a breast cancer cell is at least 7-fold higher than in normal cell. Moreover, the nanoprobe has been successfully applied to monitoring PLD activity in living cells by UCF bioimaging. The results reveal that the nanoprobe provides a simple, sensitive, and robust platform for point-of-care diagnostics and drug screening in biomedical applications.

P

functionalized nanomaterials have been reported for the detection of phospholipase.10−12 Although these methods are quite powerful, technologies that can circumvent the interferences from the complex biological matrixes are highly desirable for clinical diagnosis owning to the high complexity of the biological samples. Furthermore, all of the methods could not image PLD activity in living cells, which limits the application of the methods to in situ diagnostics for cancer. Lanthanide-doped upconversion nanoparticles (UCNPs) have attracted increasing attention because of their intriguing properties, such as tunable multicolor emission, no autofluorescence from biological samples, a remarkable light penetration depth, and exceptional photostability.13−25 These properties make UCNPs an ideal choice for bioassay in complex biological samples and living cell imaging. To date, several groups have developed upconversion fluorescence biosensors to detect special ions,26 small molecules,27,28 DNA,29,30 protein,31−33 mycotoxins,34 and adenoviruses35 based on fluorescence resonance energy transfer (FRET) systems. Moreover, a few UCNPs-based probes with both sensing and imaging function

hospholipase D (PLD) is a critical component of intracellular signal transduction that catalyzes the hydrolysis of phosphatidylcholine to generate the second messenger phosphatidic acid (PA) and choline.1 This hydrolysis reaction has been implicated in many important biological processes, including signal transduction, membrane vesicle trafficking, cytoskeletal reorganization, and cell migration.2,3 It has now been observed that there are abnormalities in PLD expression and activity in many human cancers.4 PLD is thus recognized as a potential diagnostic biomarker as well as a target for drug discovery in cancers. This has fueled the need to develop highly sensitive, simple, and robust detection methods for PLD. Currently, the widely used assay for PLD is the radiolabeled phosphatidylcholine-based methods.5 However, the requirement of radioactive labels as well as the time-consuming lipid extraction process restricts their widespread applications. Choline oxidase-based enzyme-coupled assay is an effective method for measuring PLD activity;6 however, to obtain the optimal activity measurements, the detailed optimization of assay parameters is usually necessary. Moreover, a number of alternative methods have been developed for the detection of PLD activity, including chromatography,7 mass spectrometry,8 and infrared spectroscopy,9 but they are still somewhat restrained by the requirement of special instruments. Recently, fluorometric and colorimetric assays based on phospholipid© 2014 American Chemical Society

Received: May 6, 2014 Accepted: June 18, 2014 Published: June 18, 2014 7119

dx.doi.org/10.1021/ac5016694 | Anal. Chem. 2014, 86, 7119−7127

Analytical Chemistry

Article

Germany). Phospholipase A2 (PLA2) from bovine pancreas was purchased from Sigma. 5-Fluoro-2-indolyl des-chlorohalopemide (FIPI) was purchased from Cayman Chemical (Michigan, USA). 2-Amino-2-(hydroxymethyl)-1,3-propanediol (Tris), avidin, and cytochrome C (Cyto-C) were obtained from Sangon Biotechnology Co., Ltd. (Shanghai, China). Human serum albumin (HSA), immunoglobulin G (IgG), transferrin, trypsin, and lysozyme were obtained from Dingguo Biotechnology Co., Ltd. (Beijing, China). Cyclic-AMP dependent protein kinase A (PKA) was supplied by New England Biolabs (Ipswich, England). O-Tricyclo[5.2.1.02,6]dec-9-yl dithiocarbonate potassium salt (D609) (95%) was purchased from Sigma-Aldrich. All chemicals were of analytical grade. All aqueous solutions were prepared using ultrapure water, which was obtained through a Millipore Milli-Q water purification system (Billerica, USA) and had an electric resistance >18.2 MΩ. Instruments. The transmission electron microscopy (TEM) images were collected on a field emission high resolution 2100F transmission electron microscope (JEOL, Japan) operating at an acceleration voltage of 200 kV. The crystal phases of UCNPs were identified with X-ray diffraction (XRD) on a Bruker D8 Advance X-ray diffractometer with a 2θ range from 10° to 70° at a scanning rate of 4° per minute. Dynamic light scattering (DLS) experiments were carried out on a Malvern Zetasizer (Nano-ZS, USA). The UV−vis absorption spectrum of PE-Rhod was recorded on a UV-2450 UV−vis spectrometer (Shimadzu, Japan). A 980 nm diode CW laser (Changchun New Industries Optoelectronics Tech. Co., Ltd.) was used as the excitation source with the power being set at 3 W. The upconversion fluorescence spectra were measured using a FluoroMax-4 Spectrofluorometer (HORIBA Jobin Yvon, Inc., NJ, USA) equipped with an external 980 nm laser instead of the internal excitation source. Synthesis of Oleic Acid-Capped NaYF4: Yb, Er UCNPs (OA-UCNPs). Oleic acid-capped NaYF4: 20%Yb, 2%Er UCNPs was synthesized by a modified solvothermal process according to the reported method.47 YCl3·6H2O (0.78 mmol), YbCl3· 6H2O (0.20 mmol), and ErCl3·6H2O (0.02 mmol) were added to a 50 mL three-necked flask containing 6 mL of OA and 15 mL of ODE. The mixture was heated to 160 °C for 30 min to form a homogeneous solution and then cooled down to room temperature. Then, 10 mL of methanol solution containing NaOH (2.5 mmol) and NH4F (4 mmol) was slowly added into the flask, and the mixture was stirred for 30 min. Subsequently, the solution was slowly heated and degassed at 100 °C for 10 min to remove methanol and then the solution was heated to 300 °C and maintained for 1 h under argon atmosphere. After the solution was cooled naturally, nanoparticles were precipitated from the solution with ethanol and washed with ethanol for three times. The resulting nanoparticles were dried under a vacuum for further experiments. Synthesis of Phospholipid-Modified UCNPs (LipoUCNPs). Lipo-UCNPs were synthesized according to the literature method with minor modifications.46,48 Briefly, the oleic acid-capped NaYF4: 20%Yb, 2%Er UCNPs (1.0 mg) were added into a chloroform solution (2 mL) containing 10.0 mg of DSPE-PEG phospholipid in a round-bottom flask (10 mL) and the mixture was sonicated for 10 min. Then the mixture was dried in a rotary evaporator under reduced pressure at room temperature to form a lipid film on the inside wall of the flask. The lipid film was hydrated with ultrapure water (4 mL), and the UCNPs became soluble after vigorously sonication. The

have also been reported to monitor special analytes in living cell level.36−38 In the development of the UCNPs-based biosensor, a major challenge is to make water-dispersible, biocompatible, and functionalizable UCNPs, because they are normally prepared in organic solvents and capped with hydrophobic ligands.39 Silica coating deposited on UCNP surface has been developed to render the UCNPs dispersible in water,40−42 but attenuated fluorescence intensity and the requirement of further surface modification make this approach undesirable for bioassay. In addition, direct synthesis of water-dispersible UCNPs or replacement of hydrophobic ligands with small hydrophilic ligands could also make UCNPs both water-dispersible and functionalizable,43−45 but there is still a need for a simple and general method for producing biocompatible UCNPs. Recently, Lu’s group has reported a novel approach to engineer the UCNP surface coating with a monolayer of functional phospholipids, allowing facile synthesis of UCNPs with flexible chemical surface properties for biomolecular conjugation and targeted cell imaging.46 However, to our knowledge, the direct use of phospholipid-modified UCNPs as biosensors for the detection and imaging of biomolecules remains largely unexplored. In the present study, we developed for the first time a phospholipid-modified upconversion nanoprobe for ratiometric fluorescence sensing and imaging of PLD activity in cell lysate and in living cells. The nanoprobe could be synthesized by a facile one-step self-assembly of a phospholipid monolayer composed of poly(ethylene glycol) (PEG)-modified phospholipid and rhodamine B-labeled phospholipid onto the hydrophobic UCNP NaYF4: 20%Yb, 2%Er surface. The PEGylated phospholipid rendered the nanoprobe water-dispersible and biocompatible, whereas the rhodamine B-labeled phospholipid afforded a sensing function. The FRET from the UCNPs to the chromophore (rhodamine B) occurred in the nanosystem. The cleavage of the phosphodiester bond in rhodamine B-labeled phospholipid induced by PLD made the rhodamine B apart from UCNP surface, and the FRET process was thus blocked, resulting in a change in upconversion emission spectra. Using the ratiometric upconversion emission at 540 to 655 nm as a detection signal, the PLD activity in cell lysates of two cells, normal mammary cell MCF-10A and human mammary carcinoma cell MDA-MB-231, has been successfully determined. Importantly, the upconversion nanoprobe could image PLD activity in these two cells. The results revealed that the developed nanoprobe allowed sensitive, simple, and robust detection of PLD activity in cell lysates as well as high-contrast imaging of PLD activity in living cells, implying that it provided a promising platform for point-of-care diagnostics and drug screening.

■

EXPERIMENTAL SECTION Materials. Rare earth chlorides YCl3·6H2O, YbCl3·6H2O, ErCl3·6H2O, oleic acid (OA), and 1-octadecene (ODE 90%) were purchased from Sigma-Aldrich. Phospholipids with different functional headgroups including 1,2-dioleoyl-snglycero-3-phosphoethanolamine-N-(lissamine rhodamine B sulfonyl) (ammonium salt) (PE-Rhod) and 1,2-distearoyl-snglycero-3-phosphoethanolamine-N-[methoxy(polyethylene glycol)-2000] (ammonium salt) (DSPE-PEG) were all purchased from Avanti Polar Lipids. Phospholipase D (PLD) from Streptomyces chromofuscus and phospholipase C (PLC) from Bacillus cereus were obtained from Merck (Darmstadt, 7120

dx.doi.org/10.1021/ac5016694 | Anal. Chem. 2014, 86, 7119−7127

Analytical Chemistry

Article

Scheme 1. Schematic Illustration of the Design of Rhod-Lipo-UCNP Nanoprobe and Its Sensing Principle to Phospholipase D with a Change in Upconversion Emission

streptomycin (100 μg mL−1) in a humidified atmosphere containing 5% CO2 at 37 °C. The cell lysates were prepared as described in the Supporting Information. Aliquots (5 μL) of the cell lysates (1 × 107 cells mL−1, 10-fold diluted by 10 mM Tris−HCl buffer, pH 8.0) were spiked with 10 μL of standard PLD solutions at concentrations of 0−50 or 0−200 U L−1. D609 (2.5 μL, 0.4 mM) was then added to the spiked samples and preincubated for 20 min at room temperature to eliminate the interference from PLC in the cell lysates. The mixture was then added to 32.5 μL of reaction system containing 20 μL of Rhod-Lipo-UCNP solution (4 mg mL−1), 5 μL of Tris−HCl buffer (pH 8.0, 100 mM Tris, 50 mM CaCl2), and 7.5 μL of ultrapure water, and incubated at 37 °C for 1 h. After that, the resulting solution was diluted to 100 μL with ultrapure water and then subjected to fluorescence measurement under the excitation of a 980 nm laser. Upconversion Fluorescence Imaging of PLD Activity in Living Cells. MCF-10A and MDA-MB-231 cells were plated on a 35 mm Petri dish with 10 mm bottom well in the culture medium for 24 h, respectively. Before the experiments, the cells were washed with PBS buffer for three times, and then the cells were incubated with the culture medium containing 0.5 mg mL−1 Rhod-Lipo-UCNPs and 20 μM D609 for 4 h at 37 °C. Cell imaging was then carried out after washing the cells with PBS. Upconversion fluorescence imaging was performed with an Olympus IX81 inverted microscope with an Olympus FV1000 confocal scanning system. A Spectra-Physics Mai Tai HP pulsed laser at 980 nm was used as the excitation source, and the fluorescence emission was collected in the range of 510−560 and 575−675 nm.

solution was transferred to a microtube and centrifuged slightly, the sediment was discarded to remove possible large aggregates. Excess lipids were removed from Lipo-UCNPs by ultracentrifugation (20000 rpm, 15 min) and washing. The obtained Lipo-UCNPs were finally suspended in 250 μL of ultrapure water and stored at 4 °C for further experiments. The concentration of Lipo-UCNPs was calculated as ∼4.0 mg mL−1. For the preparation of PE-Rhod coated Lipo-UCNPs (denoted as Rhod-Lipo-UCNPs), there were a few modifications of the procedure: the oleic acid-capped UCNPs (1.0 mg) were added into a chloroform solution (2 mL) containing DSPE-PEG and a series of different amounts of PE-Rhod varied from 1% to 5% mass ratio (the total amount of phospholipids was 10.0 mg). Determination of PLD Activity in Aqueous Solution. In a typical assay in aqueous solution, 20 μL of 5% PE-Rhod coated Lipo-UCNP aqueous solution (4.0 mg mL−1) was first added to 20 μL of reaction buffer containing 5 μL of Tris−HCl buffer (pH 8.0, 100 mM Tris, 50 mM CaCl2) and 15 μL of ultrapure water. Then 10 μL of PLD (dissolved in 10 mM Tris−HCl buffer, pH 8.0 with final concentrations ranging from 0 to 1500 U L−1) was added, and the mixture was incubated at 37 °C for 1 h. After that, the resulting mixture was diluted to a final volume of 100 μL with ultrapure water and then subjected to fluorescence measurement with the excitation of 980 nm laser. To examine the specificity of the Rhod-Lipo-UCNP nanoprobe toward PLD, some other biomolecules were added into the Rhod-Lipo-UCNP aqueous solution in place of PLD with the same experimental conditions and procedures. For the inhibitor analysis of FIPI, various concentrations of FIPI (10−61 mg mL−1) were preincubated with PLD (400 U L−1) in 10 μL of Tris−HCl solution (10 mM, pH 8.0) for 20 min at room temperature, and then the identical detection procedures were carried out. The upconversion fluorescence spectra were recorded under the excitation of 980 nm laser. Determination of PLD Activity in Cell Lysates. MCF10A and MDA-MB-231 cells were cultured in Dulbecco’s modified Eagle’s medium (DMEM) supplemented with 10% (v/v) fetal bovine serum, penicillin (100 units mL−1), and

■

RESULTS AND DISCUSSION Design and Principle of the Rhod-Lipo-UCNP Nanoprobe for PLD Activity Assay. The design strategy of the nanoprobe was based on PLD modulation of FRET process in Rhod-Lipo-UCNPs, with ratiometric fluorescence emission as the output signal (Scheme 1). To this end, NaYF4: 20%Yb, 2% Er nanocrystals with green and red upconversion emission were synthesized and used as an energy donor; meanwhile, two 7121

dx.doi.org/10.1021/ac5016694 | Anal. Chem. 2014, 86, 7119−7127

Analytical Chemistry

Article

Figure 1. TEM images of the (A) as-prepared OA-UCNPs and (B) Lipo-UCNPs coated with phospholipids DSPE-PEG. Inset shows the highresolution TEM images of the respective samples. (C) X-ray diffraction pattern of NaYF4: Yb, Er nanocrystals. The standard pattern of β-NaYF4 (JCPDS no. 28-1192) is also shown. (D) Upconversion fluorescence spectra of OA-UCNPs in cyclohexane (red line) and Lipo-UCNPs in water (blue line) under excitation at 980 nm. Inset: photographs of the water solution of Lipo-UCNPs without laser illumination (left) and the upconverted visible fluorescence under continuous-wave 980 nm laser illumination (right).

As the first demonstration of phospholipid-modified UCNPsbased fluorescence nanoprobe, this design has indeed several significant advantages. First, the nanoprobe can be easily prepared through a facile one-step assembly of a mixed phospholipid monolayer, and such a phospholipid coating offers the nanoprobe both excellent stability and sensing function. Second, owing to the excitation with near-infrared (NIR) light source, the nanoprobe can effectively eliminate the background autofluorescence interference from complex biological samples, thereby allowing sensitive detection with high signal-to-noise ratio. Third, the UCNPs-based design can afford ratiometric sensing because of the presence of two emission bands, one of which is strongly affected by the PLD activity and the other is not. Such ratiometric sensing permits signal rationing, thus providing built-in correction for environmental effects. Finally, the integrated nanoprobe design holds great potential for imaging of living cells, which is distinctly advantageous over the sensing system based on FRET from UCNPs to nanomaterials such as graphene oxide. Synthesis and Characterization of Lipo-UCNPs. Highly efficient upconverting NaYF4: 20%Yb, 2%Er nanoparticles were synthesized by the solvothermal method with oleic acid (OA) as the surface ligand.47 The transimission electron microscopy (TEM) image showed that these nanocrystals display a uniform hexagonal plate-like morphology with a mean size of approximately 34 nm (Figure 1A). The small particles in the TEM image may be the irregular NaYF4 nanocrystals, which were not eventually grown into uniform, large nanocrystals with regular shapes. The X-ray diffraction (XRD) analysis (Figure 1C) indicated that the peak positions and intensities of the nanocrystals agreed well with the calculated values of the pure

phospholipid components with different functions were chosen to construct the nanosystem, one was the rhodamine B-labeled phospholipid (PE-Rhod), acting as an energy acceptor, and the other was the PEGylated phospholipid (DSPE-PEG) to afford water solubility and biocompatibility. The nanoprobe could be easily prepared through a one-step assembly of a phospholipid monolayer driven by the hydrophobic van der Waals interactions between the hydrophobic tail of the phospholipids and the primary oleate ligands on the OA-UCNP surface. Such a process not only made these nanocrystals well dispersible in water with excellent stability, but also drew the rhodamine B energy acceptor molecules close to the surface of UCNPs. Additionally, in such a way, the surface of UCNPs could be modified with a large number of hydrophilic PEG polymers, which could form a hydrophilic coating on the external surface of the nanoprobe and prevent the cellular lipids from inserting the inner hydrophobic layer of the nanoprobe, providing excellent stability when the nanoprobe was incubated with cell lysate and living cells. Due to the significant spectral overlap between the green emission of UCNPs and the absorption of rhodamine B, FRET from UCNPs to dye molecules occurred. In the presence of PLD, the cleavage of the phosphodiester bond in rhodamine B-labeled phospholipid made the rhodamine B molecules apart from UCNP surface, and the FRET process was thus blocked, resulting in an increase in green emission peak of UCNPs at 540 nm. The red emission peak of UCNPs at 655 nm was not affected because rhodamine B molecules have no absorption at this wavelength. As a result, the red emission at 655 nm could act as an internal standard, and a ratio of upconversion fluorescence intensity at 540 and 655 nm (UCF540/UCF655) could be used as the output signal of the ratiometric fluorescence detection. 7122

dx.doi.org/10.1021/ac5016694 | Anal. Chem. 2014, 86, 7119−7127

Analytical Chemistry

Article

Figure 2. (A) Absorbance spectrum (red line) and fluorescence emission spectrum (blue line) of PE-Rhod in chloroform, and upconversion fluorescence spectrum (black line) of OA-UCNPs in cyclohexane. (B) Upconversion fluorescence spectra of Rhod-Lipo-UCNPs coated with varying amounts of PE-Rhod under excitation at 980 nm. Inset shows the photographs of the water solution of PE-Rhod-coated (5%) Lipo-UCNPs without laser illumination (left) and the upconverted visible fluorescence under 980 nm laser illumination (right). (C) Upconversion fluorescence quenching efficiency versus amount of PE-Rhod coated on the Lipo-UCNP surface (0, 1%, 2%, 3%, 4%, and 5% mass ratio, respectively).

energy to the organic chromophores (acceptor), resulting in changes in upconversion emission spectra. Herein, rhodamine B was selected as the organic chromophore because its absorbance peak significantly overlapped with the green emission peak at 540 nm of UCNPs, which was a prerequisite for FRET (Figure 2A). The rhodamine B-labeled phospholipid (PE-Rhod) was then mixed with the DSPE-PEG phospholipid and used for coating the OA-UCNPs to introduce the organic chromophore acceptor molecules on the surface of LipoUCNPs (denoted as Rhod-Lipo-UCNPs). A series of RhodLipo-UCNPs with varying amounts of PE-Rhod were synthesized by changing the mass ratio of PE-Rhod and DSPE-PEG in their lipid coating (from 1% to 5% PE-Rhod). Under continuous wave excitation at 980 nm, the intensity of the green characteristic emission peak of Rhod-Lipo-UCNPs at 540 nm deceased gradually with the increase in the amount of PE-Rhod, accompanied by the appearance of a new emission peak of rhodamine B at 590 nm (Figure 2B). This result suggested the occurrence of FRET from UCNPs to the organic chromophore acceptor molecules. Compared with the emission spectrum of DSPE-PEG-coated Lipo-UCNPs, significant quenching (∼95%) in the emission peak at 540 nm was observed for 5% PE-Rhod-coated Lipo-UCNPs. The red emission of Lipo-UCNPs at 655 nm was unaffected by the introduction of PE-Rhod because this organic dye had no absorption at this wavelength. As a result, when exposed to a 980 nm NIR laser, the fluorescence of the 5% PE-Rhod-coated Lipo-UCNPs in water appeared predominantly red (Figure 2B, inset). Figure 2C depicted the relationship between the quenching efficiency of the emission peak at 540 nm and the amount of PE-Rhod coated on the surface of Lipo-UCNPs. The quenching efficiency was calculated according to the formula (F0 − F)/F0, where F0 and F represent the fluorescence intensity at 540 nm in the absence and presence of PE-Rhod, respectively. The quenching efficiency increased gradually with an increasing amount of PE-Rhod and reached a maximum value of ∼95% for 5% PE-Rhod. In contrast, simple physical mixing of OA-UCNPs with the identical amount (5%) of PERhod in cyclohexane led to only a 10% quenching efficiency in the green emission at 540 nm (Figure S3 in the Supporting Information). These results indicated that the quenching effect of Rhod-Lipo-UCNPs was mainly ascribed to the FRET process rather than simply absorbed light process by PE-Rhod. Sensing of Rhod-Lipo-UCNP Nanoprobe for PLD Activity in Aqueous Solution. Having successfully con-

hexagonal-phase NaYF4: Yb, Er nanocrystals (JCPDS no. 281192). To transfer the hydrophobic OA-coated UCNPs into the aqueous solution, an amphiphilic PEGylated phospholipid, DSPE-PEG, was used to coat the OA-UCNPs. The TEM image of the resulting Lipo-UCNPs (Figure 1B) indicated that they remained monodisperse without obvious changes in size, shape, and crystallinity after modification with the phospholipid. Highresolution TEM investigation (Figure 1B, inset) confirmed the core/shell nanocrystals with a uniform, approximately 2 nm thick, hydrophobic oleic acid/lipid layer around the surface. Dynamic light scattering (DLS) measurements indicated that the Lipo-UCNPs were well-dispersed in water with a mean hydrodynamic diameter of approximately 60 nm (Figure S1 in the Supporting Information). In comparison with OA-UCNPs dispersed in cyclohexane (ca. 45 nm), this increase of approximately 15 nm in diameter was in agreement with a monolayer of the PEGylated phospholipids. In addition, the assembly of the phospholipid monolayer on the OA-UCNP surface was further confirmed by FT-IR (Figure S2 in the Supporting Information). The absorption bands around 1563 and 1448 cm−1 for the OA-UCNPs, attributed to the stretching vibration of carboxyl group in oleic acid, were shifted to 1662 cm−1 for the Lipo-UCNPs, corresponding to the stretching vibration of amide bond in DSPE-PEG phospholipid. Compared with the spectrum of OA-UCNPs, a new peak at 1099 cm−1 appearing on the Lipo-UCNPs was ascribed to the stretching vibrations of the C−O bond in PEG, indicating that the DSPE-PEG phospholipid was successfully assembled on the OA-UCNP surface by self-assembly. Furthermore, the upconversion fluorescence spectrum of Lipo-UCNPs in water was similar to that of the OA-UCNPs in cyclohexane with a slight decrease owing to the surface quenching effect of water molecules (Figure 1D). The Lipo-UCNPs showed excellent water solubility with long-term stability in water and resistance to aggregation over several months (Figure 1D, inset). Upon continuous excitation at 980 nm, the fluorescence of the LipoUCNPs in water appeared predominantly green (Figure 1D, inset). These results strongly indicated that the characteristic upconversion property of the nanoparticles was unaffected by the phospholipid coating. Construction of the FRET-based Rhod-Lipo-UCNP Nanoprobe. To achieve the sensing function, the LipoUCNPs should be combined with other chromophores through the FRET process, where the Lipo-UCNPs (donor) transfers 7123

dx.doi.org/10.1021/ac5016694 | Anal. Chem. 2014, 86, 7119−7127

Analytical Chemistry

Article

Figure 3. (A) Upconversion fluorescence spectra of 5% PE-Rhod-coated Lipo-UCNPs before (green line) and after incubation with 1500 U L−1 PLD (purple line), 50 mg mL−1 HSA (red line), and denatured PLD (blue line). (B) Time-dependent fluorescence recovery curves at 540 nm obtained for PLD of 0 (black line), 10 (blue line), 100 (red line), 300 (green line), and 1000 U L−1 (purple line) at 37 °C. (C) Upconversion fluorescence spectra of the Rhod-Lipo-UCNPs after incubation with PLD of varying concentrations. (D) The ratio of the upconversion fluorescence intensity at 540 and 655 nm (UCF540/UCF655) as a function of PLD concentration. Error bars are standard deviation of three repetitive experiments.

structed the FRET-based Rhod-Lipo-UCNP nanoprobe, we then explored the usage of the nanoprobe for sensing PLD activity. The Rhod-Lipo-UCNPs coated with 5% PE-Rhod were used because of their high quenching efficiency. Figure 3A depicts the typical fluorescence spectral responses of the nanoprobe in the assay of PLD. The 5% PE-Rhod-coated LipoUCNPs were observed to exhibit very weak fluorescence intensity at the emission peak of 540 nm. After incubation of the Rhod-Lipo-UCNPs with 1500 U L−1 PLD, the reaction mixture showed a very strong fluorescence signal at the emission peak of 540 nm and the fluorescence enhancement ratio was ∼12. This result indicated the recovery of the fluorescence signal of these Rhod-Lipo-UCNPs during the reaction with PLD, which resulted from the catalytic cleavage of phosphodiester bond in rhodamine B-labeled phospholipid by PLD, leading to the release of rhodamine B from UCNP surface and the resulting suppression of FRET. In contrast, when human serum albumin (HSA) was added to the reaction system to replace PLD or the PLD was denatured by heating at 90 °C for 10 min, no obvious fluorescence increase at the emission peak of 540 nm was observed, verifying that the fluorescence recovery was specifically mediated by active PLD. The cleavage reaction could also be validated by LTQ Orbitrap Velos mass spectrometry analysis (Figure S4 in the Supporting Information). Before the PLD-mediated catalytic reaction, the Rhod-Lipo-UCNPs gave a peak for PE-Rhod (m/z: 1282.66), whereas after the reaction, a new peak corresponding to the hydrolysis product phosphatidic acid (m/z: 699.49) appeared. These data gave immediate evidence for the cleavage reaction catalyzed by PLD. In addition, the microscopic image experiments of the Rhod-Lipo-UCNPs in the absence and

presence of PLD were also performed (Figure S5 in the Supporting Information). For the 5% PE-Rhod-coated LipoUCNPs, no obvious green upconversion fluorescence at 510− 560 nm was observed under the excitation of the 980 nm pulsed laser. After incubation with 1500 U L−1 PLD, the RhodLipo-UCNPs showed a bright green upconversion fluorescence emission, indicating the release of rhodamine B from UCNP surface and the resulting suppression of FRET. Furthermore, the time-dependent fluorescence recovery profiles at 540 nm were also recorded at various PLD concentrations to monitor in real time the hydrolysis reaction catalyzed by PLD (Figure 3B). The reaction rate increased gradually with the increase in PLD concentration, and a 60 min reaction time was selected in the subsequent experiments. It should be noted that 300 U L−1 PLD could not yield full recovery of luminescence at 540 nm. This might be due to the steric hindrance of the PEG large molecules on the surface of UCNPs, which hindered the PLD enzyme getting close to its substrate PE-Rhod, and so it could not completely cleave the phosphodiester bond. The ability of the Rhod-Lipo-UCNP nanoprobe for quantitative analysis of the activity of PLD was then investigated. Figure 3C displays typical upconversion fluorescence spectra of the Rhod-Lipo-UCNPs after incubation with PLD of varying concentrations. The intensity of the green upconversion emission at 540 nm increased linearly with the PLD concentration in the range 0−400 U L−1, and the reaction reached saturation upon addition of 800 U L−1 PLD (Figure 3D). When the upconversion fluorescence (UCF) at 655 nm was used as an internal standard, and the ratio of the fluorescence intensity at 540 and 655 nm (UCF540/UCF655) was chosen as the detection signal, the detection limit of the 7124

dx.doi.org/10.1021/ac5016694 | Anal. Chem. 2014, 86, 7119−7127

Analytical Chemistry

Article

determined the IC50 value of FIPI toward PLD (400 U L−1) to be 3.75 μM. This IC50 value of FIPI for PLD agreed with that (2.66 μM) determined by the commercial Amplex Red PLD assay (Figure S7 in the Supporting Information). These results suggest that the Rhod-Lipo-UCNP nanoprobe holds great potential for screening inhibitors of PLD. Determination of PLD Activity in Cell Lysates. To test the practicality of the biosensor, we performed analyses of PLD in the cell lysates of two cell lines, normal mammary cell MCF10A and human mammary carcinoma cell MDA-MB-231. D609 (20 μM) was added into the samples of the cell lysates to eliminate the interference from PLC. Using a standard addition method (Figure S8 in the Supporting Information), the RhodLipo-UCNP nanoprobe provided a linear response to PLD in the MCF-10A cell lysate at 0−50 U L−1 and in the MDA-MB231 cell lysate at 0−200 U L−1. Table 1 lists the activities of

Rhod-Lipo-UCNPs (in terms of the rule of 3 times deviation over the blank response) was estimated to be 4 U L−1. Such a low detection limit can be attributed to the ratiometric detection and a low fluorescence background for upconversion detection. The Rhod-Lipo-UCNPs provided a relatively low detection limit and a wider dynamic range (2 orders of magnitude) than those provided by existing optical methods for PLD analysis.49,50 To further test the specificity of this biosensor toward PLD, we incubated aliquots of Rhod-Lipo-UCNP suspension (0.8 mg mL−1) with other possible interfering species including cysteine, glycine, glutathione, transferrin, trypsin, lysozyme, immunoglobulin G, human serum albumin, avidin, cytochrome C, protein kinase A, phospholipase A 2 (PLA 2 ), and phospholipase C (PLC). As shown in Figure 4, except for

Table 1. Determination of the PLD Activity in Two Cell Lysates by the Rhod-Lipo-UCNPs and Commercial Amplex Red PLD Assay Kit

cell linea MCF10A MDAMB231

Figure 4. Relative fluorescence increase (R − R0)/R0 at 540 nm of Rhod-Lipo-UCNPs (0.8 mg mL−1) upon addition of other amino acids, proteins or enzymes. R0 and R represent the ratio UCF540/ UCF655 of the Rhod-Lipo-UCNPs in the absence and presence of interfering species or PLD, respectively. The concentrations of PLA2, PLC, and PLD were 1500 U L−1, the PKA concentration was 25 000 U L−1, and the concentrations of other interfering species were 100 nM. Error bars represent the standard deviations of three repetitive experiments.

Student’s ttest values between the two approachesb

spiked [PLD] (U/L)

Rhod-LipoUCNPs [mean ± SD (U/L, n = 5)]

commercial Amplex Red PLD assay kit [mean ± SD (U/L, n = 5)]

0−50

408.6 ± 28.7

431.7 ± 30.8

1.23

0−200

2921.4 ± 216.7

3124.3 ± 227.7

1.44

Concentration of cell lines is 1 × 107 cells/mL. bThe Student’s t-test value is 2.306 at the 95% confidence level. a

PLD in the cell lysate samples from the two cell lines determined by the present assay and the commercial Amplex Red PLD assay. According to the values from the Student’s ttest, the results from the present method showed no significant differences from those obtained by the Amplex Red PLD assay. The activities of PLD from the normal mammary MCF-10A cell and the mammary carcinoma MDA-MB-231 cell lysates (1 × 107 cells mL−1) were 408.6 ± 28.7 and 2921.4 ± 216.7 U L−1, respectively, which was in agreement with the previously reported ∼7-fold higher activity of PLD in MDA-MB-231 cell line compared to MCF-10A cell line.54 Upconversion Fluorescence Imaging of PLD Activity in Living Cells. To demonstrate the applicability of the RhodLipo-UCNPs in the bioimaging of intracellular PLD activity, upconversion fluorescence microscopy experiments were carried out. As shown in Figure 5, MCF-10A cells incubated with 0.5 mg mL−1 Rhod-Lipo-UCNPs and 20 μM D609 for 4 h at 37 °C showed only a week green UCF emission at 540 nm, whereas MDA-MB-231 cells displayed a strong enhancement in the green emission under the same incubation conditions. This was due to that the nanoprobe delivered in the cytosol reacted with PLD in the cytosol of MDA-MB-231 cells, resulting in the remove of rhodamine B, accompanied by the enhancement of green fluorescence. Bright-field measurements confirmed that the cells treated with Rhod-Lipo-UCNPs remained viable throughout the imaging experiments. Overlay of UCF imaging and bright-field images revealed that the UCF signals were localized in the intracellular region, indicating the high intracellular delivery efficiency of Rhod-Lipo-UCNPs. In addition, a further localization study revealed that the RhodLipo-UCNPs were not colocalized with lysosomes, confirming

PLC, these species did not interfere in the detection of PLD. Because the PLC could also catalyze the hydrolysis of the phosphodiester bond, the addition of PLC led to significant enhancement of the green upconversion emission at 540 nm of the UCNPs. However, the interference of PLC could be effectively eliminated by the addition of 20 μM D609, a selective inhibitor of PLC that did not inhibit the PLD activity.51,52 As can be seen from Figure 4, when incubated with the mixture of PLC and D609, the Rhod-Lipo-UCNPs gave a negligible signal enhancement, whereas the determination of PLD activity was not affected by the introduction of D609. Phospholipase PLA2 catalytically cleaved one fatty acyl chain and the other was still linked with the phospholipid, which would not result in the remove of rhodamine B from the phospholipid. As a result, PLA2 did not interfere in the determination of PLD. Therefore, using D609 as an effective inhibitor of PLC, the Rhod-Lipo-UCNPs can act as a highly selective and sensitive ratiometric upconversion nanoprobe for PLD activity. Further investigation of the biosensor for enzyme−inhibitor assay was performed. 5-Fluoro-2-indolyl des-chlorohalopemide (FIPI) is a well-known inhibitor of PLD.53 We incubated RhodLipo-UCNPs with 400 U L−1 PLD in the presence of FIPI of varying concentrations (10−6, 10−5, 10−4, 10−3, 10−2, 10−1, and 1 mg mL−1) to obtain the inhibition curve (Figure S6 in the Supporting Information). From the inhibition curve of the ratio UCF540/UCF655 against the log concentration of FIPI, we 7125

dx.doi.org/10.1021/ac5016694 | Anal. Chem. 2014, 86, 7119−7127

Analytical Chemistry

■

Article

ASSOCIATED CONTENT

S Supporting Information *

Experimental details and additional figures as noted in the text. This material is available free of charge via the Internet at http://pubs.acs.org.

■

AUTHOR INFORMATION

Corresponding Author

*X. Chu. E-mail: [email protected]. Tel.: 86-731-88821916. Fax: 86-731-88821916. Notes

The authors declare no competing financial interest.

Figure 5. Ratiometric UCF images in MCF-10A cells (A) and MDAMB-231 cells (B) incubated with 0.5 mg mL−1 Rhod-Lipo-UCNPs and 20 μM D609 for 4 h at 37 °C. Emission was collected by green channel at 510−560 nm (A1 and B1) and red channel at 575−675 nm (A2 and B2) under excitation at 980 nm. (A3 and B3) Overlay of brightfield, green, and red UCF images. (A4 and B4) Ratiometric UCF images with ratio of green to red channels. Scale bar: 20 μm.

■

ACKNOWLEDGMENTS This work was supported by the National Natural Science Foundation of China (No. 21275045), NCET-11-0121, and Hunan Provincial Natural Science Foundation of China (Grant 12JJ1004).

■

cytosolic delivery of the UCNPs (Figure S9 in the Supporting Information). Furthermore, in light of the significant enhancement of green UCF at 510−560 nm and the unchanged red UCF at 575−675 nm under excitation of UCNPs with 980 nm light, the ratiometric UCF imaging was investigated when UCF emissions were collected at green channel (510−560 nm) and red channel (575−675 nm). As shown in Figure 5, MCF-10A cells showed a ratio of the green channel to red one of