Article pubs.acs.org/ac
Aptamer-Conjugated Polymeric Nanoparticles for the Detection of Cancer Cells through “Turn-On” Retro-Self-Quenched Fluorescence Lin-Chen Ho,† Wei-Cheng Wu,‡ Chang-Yu Chang,§,∥ Hao-Hsuan Hsieh,† Ching-Hsiao Lee,§ and Huan-Tsung Chang*,† †
Department of Chemistry, National Taiwan University, 1, Section 4, Roosevelt Road, Taipei, Taiwan Nano Science and Technology Program, Taiwan International Graduate Program, Academia Sinica, Taipei, Taiwan and National Tsing-Hua University, Hsinchu, Taiwan § Jen-Teh Junior College of Medicine, Nursing and Management, Miaoli County, Taiwan ∥ Public Health Center of Zhunan Township, Miaoli County, Taiwan ‡
S Supporting Information *
ABSTRACT: We have developed a simple, sensitive, and rapid fluorescence assay for the detection of cancer cells, based on “turn-on” retro-self-quenched fluorescence inside the cells. 1,3-Phenylenediamine resin (DAR) nanoparticles (NPs) containing rhodamine 6G (R6G) are conjugated with aptamer (apt) sgc8c to prepare sgc8c-R6GDAR NPs, while that containing rhodamine 101 (R101) are conjugated with TD05 for the preparation of TD05-R101DAR NPs. The sgc8c-R6GDAR and TD05-R101DAR NPs separately recognize CCRF-CEM and Ramos cells. The fluorescence intensities of the two apt-DAR NPs are both weak due to self-quenching, but they increase inside the cells as a result of release of the fluorophores from the apt-DAR NPs. The apt-DAR NPs’ structure becomes less compact at low pH value, leading to the release of the fluorophores. The sgc8cR6GDAR and TD05-R101DAR NPs allow detection of as low as 44 CCRFCEM cells and 79 Ramos cells mL−1, respectively, using a commercial reader within 10 min. Practicality of the two probes have been validated by the quantitation and identification of CCRF-CEM and Ramos cells spiked in blood samples through conventional fluorescence and flow cytometry analysis, with advantages of sensitivity, selectivity, and rapidity.
E
accurate cell detection, leading to increased complexity and costs, in addition to longer analysis time. Aptamers are single-stranded DNA or RNA fragments that fold into various shapes to specifically bind targets such as proteins, metals, and small organic molecules.13,14 The sequence and conformation of an aptamer are important to determine its affinity toward a specific target. Relative to antibodies, they are more stable, inexpensive, and can be synthesized more easily.15 Different aptamer-based assays, particularly in combination with nanomaterials, have been developed to enhance the sensitivity for the detection of cancer cells.16−18 Aptamer-functionalized nanomaterials provide multivalence effects, leading to stronger interaction of aptamers with targeted cells and thus enhanced sensitivity.19−21 In addition, multiple aptamer−receptor interactions allow cellular internalization like endocytosis, facilitating translocation of nanomaterials through the cell membrane.22,23 Because it is difficult for nuclease to access the surfaces of nanomaterials, aptamers on
arly detection of cancer cells commonly done through noninvasive and sensitive conventional molecular imaging approaches greatly increases the chance for successful treatment;1−3 high-cost reagents and time-consumed processes are however usually required. Recently, it has shown important in early cancer diagnosis through the detection of circulating tumor cells (CTCs) that are cancer cells disseminating from tumors and disperse into the bloodstream acting as an origin point at the cellular level to cause fatal metastasis.4−6 Owing to its extremely low abundance (a few to hundreds per mL) among a high number (109 cells mL−1) of hematologic cells, detection of CTC is not an easy task.7,8 To detect them, a number of collecting methods such as magnetic activated cell sorting9,10 and microchip based extraction,11,12 followed by immunophenotypic analyses have been developed. However, target antigens are required for cellular recognition, which may not express on all cancerous cells, resulting in a difficulty of detection and possible false negative results. Additionally, antibody integrity is perennially contested due to the lifetime of the sources, leading to a difficulty of obtaining reliable antibodies sometimes. Thus, immunophenotypic analyses usually require multiple antibody recognition elements for © XXXX American Chemical Society
Received: February 11, 2015 Accepted: April 8, 2015
A
DOI: 10.1021/acs.analchem.5b00569 Anal. Chem. XXXX, XXX, XXX−XXX
Article
Analytical Chemistry
(R6GDAR or R101DAR NPs) was washed sequentially with ethanol (3 × 8 mL) and ultrapure water (3 × 8 mL) for 3 times to remove the unreacted chemicals. The R6GDAR and R101DAR NPs were then dried by employing an EYELA FDU-2100 lyophilizer (Tokyo, Japan) at −80 °C and 15 Pa prior to use. Preparation of Aptamer-Conjugated Fluorescent Nanoparticles. Aptamer sgc8c (5′-COOH-TTT TTT TTT TTT TTT TTT TTT TAT CTA ACT GCT GCG CCG CCG GGA AAA TAC TGT ACG GTT AGA-3′) and TD05 (5′COOH-TTT TTT TTT TTT TTT AAC ACC GGG AGG ATA GTT CGG TGG CTG TTC AGG GTC TCC TCC CGG TGA-3′) were used for specific targeting of CCRF-CEM and Ramos cells, respectively.28 The poly-T linker in each aptamer allows the recognition element to extend out from the DAR surface to form a right conformation for specific binding with the target. Both aptamers were conjugated onto NPs through the reaction of the carboxyl group on their 5′terminates with the amine groups on the DARs via the EDCcatalyzed reaction. 5′-Carboxylate aptamer (1 nmol) was mixed with EDC (15 nmol) and NHS (15 nmol) sequentially in 0.1 mL of 10 mM MES buffer (pH 5.5) before being subjected to vortex. After 15 min, the reactant mixtures were added separately to 50 μg of DARs, which were adjusted to pH 7.4 and then vortexed for another 3 h to perform conjugation. Each of the solutions was subjected to centrifugation at 10 000g for 10 min, and then the pellet (sgc8c-R6GDAR or TD05R101DAR NPs) was washed sequentially with Tris−HCl buffer (pH 8.0, 5.0 mM; 3 × 0.1 mL) for 3 times to remove the unreacted chemicals. The sgc8c-R6GDAR and TD05R101DAR NPs were then separately resuspended and stored at 4 °C in Tris−HCl buffer (pH 8.0, 5.0 mM, 100 μL). The concentration of aptamer molecules in the supernatant was determined by fluorescence measurement using a singlestranded DNA binding fluorophore (OliGreen; excitation and emission wavelengths at 480 and 524 nm, respectively). The coupling yields of sgc8c-R6GDAR and TD05-R101DAR to the NPs were determined to be 66.6% and 74.1%, respectively. Cell Culture. CCRF-CEM and Ramos cells were cultured in RPMI-1640 medium, supplemented with 10% fetal bovine serum and antibiotic-antimycotic (1%) in an environment equilibrated with 5% CO2 at 37 °C. NIH-3T3 cells were maintained in DMEM supplemented with FBS (10%), antibiotic-antimycotic (1%), L-glutamine (2 mM), and nonessential amino acids (1%) in an environment equilibrated with 5% CO2 at 37 °C. Stability of Apt-DAR NPs against Endonucleases. Aliquots (10 μL) of sgc8c-R6GDAR and TD05-R101DAR NPs were mixed with endonuclease DNase I (100 nM) in Tris−HCl buffer (pH 7.4, 50 mM) containing 150 mM NaCl, 5.0 mM KCl, 1.0 mM MgCl2, and 1.0 mM CaCl2 for 2 h. CCRF-CEM or Ramos cells (7500 counts) were added into the prepared mixture, which were then incubated for 5 min. The fluorescence intensities were recorded using a microplate reader. Cytotoxicity Assays. CCRF-CEM, Ramos, and NIH-3T3 cells (1.5 × 104 counts well−1) were maintained separately in culture media at 37 °C for 24 h. The cells were then incubated with various concentrations of sgc8c-R6GDAR or TD05R101DAR NPs (0−20 μg mL−1). After 24-h incubation, the cells were carefully washed with PBS (3 × 200 mL) three times, followed by treatment with the Alamar Blue reagent (1× , 500 μL well−1) for 4 h. The fluorescent product of 7-hydroxy-3H-
the surface of nanomaterials relative to their corresponding free aptamers are much more stable against nuclease digestion.24,25 Aptamer-based assays for simultaneous detection of multiple cancer cells are rare, leading to a demand for developing rapid, simple, and noninvasive sensors for simultaneous detection of multiple cancer cells, which is important for early cancer diagnosis.26 Here, we proposed a fluorescence-based method using aptamers functionalized 1,3-phenylenediamine resin (DAR) nanoparticles (NPs) for simultaneous detection of two cancer cells. To develop multiple probes for different cancer cells, each probe must (1) glow a distinct color for fluorescence analysis and (2) has a specific molecule for targeting the corresponding type of cancer cell. The proof-ofconcept experiment was tested by the preparation of two aptamers conjugated retro-self-quenching fluorescent DAR NPs; rhodamine 6G (R6G) or rhodamine 101 (R101) molecules were trapped inside the particles. The fluorescence intensities of aptamer sgc8c conjugated DAR NPs trapped with R6G (sgc8c-R6GDAR NPs) and aptamer TD05 conjugated DAR NPs trapped with R101 (TD05-R101DAR NPs) changed when they were located inside the cells, mainly due to pH induced release of the fluorophores from the NPs.27 We carefully evaluated the effect of ligand density as well as incubation time on the sensitivity and selectivity of the two probes for CCRF-CEM and Ramos cells. Practicality of the two probes was validated by rapid recognition and quantification of the two targeted cancer cells.
■
EXPERIMENTAL SECTION Materials. Formaldehyde and DA were purchased from Acros (Geel, Belgium). All carboxylate-modified oligonucleotides were purchased from Integrated DNA Technologies (Coralville, IA). 1-Ethyl-3-(3-(dimethylamino)propyl) carbodiimide (EDC), MES buffer, N-hydroxysuccinimide (NHS), R6G, and R101 were purchased from Sigma-Aldrich (St. Louis, MO). Phosphoric acid, and its monobasic, dibasic, and tribasic sodium salts were obtained from J.T. Baker (Phillipsburg, NJ). OliGreen was purchased from Molecular Probes (Eugene, OR). ACK lysing buffer, fetal bovine serum (FBS), and RPMI-1640 were purchased from Gibco BRL (Grand Island, NY). Antibiotic-antimycotic solution, L-glutamine, and MEM nonessential amino acid solution were obtained from Biowest (Lewes, U.K.). Human acute lymphoblastic leukemia CCRFCEM cells and human Burkitt’s lymphoma Ramos cells were obtained from the American Type Culture Association (Manassas, VA). Aliquots of phosphate-buffered saline solution (PBS, 1×, 1 L) containing NaCl (8 g), KCl (0.2 g), Na2HPO4 (1.44 g), and KH2PO4 (0.24 g) were adjusted to pH 4.0−8.0 by adding suitable amounts of phosphoric acid (200 mM) or sodium hydroxide (200 mM). A Milli-Q ultrapure water purifier from Millipore (Billerica, MA) was used to produce ultrapure water (18.2 MΩ cm). Synthesis of pH-Sensitive Fluorescent Nanoparticles. R6GDAR and R101DAR NPs were both prepared according to an extended Stöber method. Ammonia aqueous solution (25 wt %, 0.025 mL) and saturated formaldehyde solution (37%, 0.070 mL) were first mixed with ultrapure water (7 mL). R6G or R101 (1.0 mg) was then added to the solution, which was stirred at ambient temperature (25 °C) for 24 h. Freshly prepared DA solution (460 mM, 0.5 mL) was added to each of the reactant mixtures, which was stirred for another 24 h to form R6GDAR or R101DAR NPs. The solution was subjected to centrifugation at 10,000 g for 10 min, and then the pellet B
DOI: 10.1021/acs.analchem.5b00569 Anal. Chem. XXXX, XXX, XXX−XXX
Analytical Chemistry
■
Article
RESULTS AND DISCUSSION Characterization of DAR NPs and Apt-DAR NPs. R6G and R101 are ideal reporters for retro-self-quenching because of their pH-insensitive nature and high fluorescence. Figure S1 in the Supporting Information displays that R6GDAR NPs like R6G and R101DAR NPs like R101 have emission bands centered at 550 and 600 nm, respectively, when separately excited at 500 and 560 nm. At pH values higher than 7.0, their fluorescence intensities were both weak because the fluorophores at a high concentration were tightly packed on the hydrophobic patches of the NPs, leading to self-quenching. The fluorescence intensities of R6GDAR and R101DAR increased upon decreasing pH values from 8.0 to 4.0. Upon decreasing pH values, the pendant amino groups on DAR backbones underwent protonation, resulting in the electrostatic repulsion between the RNH3+ groups in the NPs.30 As a result, the polymer structure became expanded and some of the fluorophores released into solution, leading to increased fluorescence intensity.27 We point out that the structure of the DAR NPs is salt and viscosity insensitive. Two DNA-aptamers, sgc8c and TD05, were used to separately functionalize R6GDAR and R101DAR NPs, because they exhibit high affinity toward PTK7 on CCRF-CEM (Kd = 0.8 nM)31,32 and mIgm on Ramos cells (Kd = 74 nM), respectively.33,34 The sensing mechanisms of sgc8c-R6GDAR NPs and TD05-R101DAR NPs for CCRF-CEM and Ramos cells, respectively, are depicted in Scheme 1. Since the amino
phenoxazin-3-one (resorufin) reduced by live cells was measured at 590 nm when excited at 545 nm using a Synergy 4 Multi-Mode Microplate Reader (Biotek Instruments). Fluorescence Assays. The fluorescence emission spectra of R6GDAR, R101DAR, sgc8c-R6GDAR, and TD05R101DAR NPs at various pH values (4.0−8.0) were recorded separately to investigate their pH-dependent properties. PB solutions (5 mM, 1 mL) at various pH values were mixed separately with the NPs (final concentration 50 μg), which were then shaken at 100 rpm for 30 min before recording their fluorescence spectra by using a Cary Eclipse PL spectrophotometer (Varian, CA). The excitation wavelengths of R6GDAR/sgc8c-R6GDAR and R101DAR/TD05-R101DAR NPs were set at 500 and 560 nm, respectively. A series of CCRF-CEM and/or Ramos cells (0−30,000 counts) samples were mixed with sgc8c-R6GDAR (20 μg), TD05-R101DAR (20 μg), or their mixture (each 20 μg) in PBS, which were then equilibrated at ambient temperature for 5 min. The fluorescence spectrum of each of the mixtures was recorded using the Synergy 4 Multi-Mode Microplate Reader. The apt-DARs were further used for the detection of cancer cells spiked in human blood samples collected from five healthy volunteers (22−35 years old). The blood samples were collected through venipuncture into tubes containing ethylenediaminetetraacetic acid (EDTA) and stored at 4 °C until required for analysis. Aliquots of CCRF-CEM/Ramos cells (0− 7500 counts) were separately spiked into whole blood samples (1 mL). To minimize the interference from hemoglobin,29 these samples were subjected to red blood cell lysis by mixing with 10 mL of ACK lysing buffer for 10 min and the white blood cells were collected by centrifugation at 300g for 5 min at ambient temperature. The pellets were washed using cold PBS (2 × 1 mL) at 4 °C for 2 times to remove the unwound component. The collected cells were finally resuspended in PBS (1 mL) and incubated with the two probes (each for 20 μg) for 5 min before their fluorescence intensities were recorded. Cell Imaging. Fluorescence imaging was conducted using an Olympus IX-71 inverted microscope. The cellular images were taken with a 20× 0.70 NA objective. An excitation filter (470/10) and an emission filter (525/25) were used when using the sgc8c-R6GDAR NPs. For TD05-R101DA NPs, an excitation filter (560/15) and a 610 nm long-pass emission filter were used. Flow Cytometry Analysis. Flow cytometry analysis was also conducted to validate the two probes for detecting their corresponding targeted cells. Aliquots of CCRF-CEM/Ramos cells (0−2500 counts) were spiked into blood samples (200 μL) obtained from three healthy volunteers (22−35 years old) separately, which were then incubated with the two probes for 5 min. These samples were then subjected to red blood cell lysis by using FACS lysing solution (2 mL) before flow cytometry analysis using a FACScan cytometer from Becton− Dickinson Bioscience (Franklin Lakes, NJ). The fluorescence histogram was acquired for 105 events of each experiment, and only live cells were analyzed based on side and forward scattered lights. The threshold was set as the maximum fluorescence observed for the control set, which was not spiked with CCRF-CEM/Ramos cells. Cells with fluorescence higher than the threshold were considered to be recognized and detected by the probes.
Scheme 1. Schematic Diagram of Cell Detection/ Identification Using Sgc8c-R6GDAR and TD05-R101DAR NPs Based on “Turn-On” Retro-Self-Quenched Fluorescencea
a
The Sgc8c-R6GDAR and TD05-R101DAR NPs specifically bind to PTK7 in the CCRF-CEM cells and mIgm in the Ramos cells, respectively. The fluorescence of the two probes “light-up” inside the target cells as a result of pH induced fluorescence enhancement.
groups on sgc8c-R6GDAR and TD05-R101DAR enhanced the cellular uptake through clathrin-mediated endocytosis,35−37 an incubation time of 5 min was enough for this study. Their fluorescence increased inside the cells, mainly due to release of the fluorophores from the NPs. Inside the cells (lower pH environment relative to the PBS), the NPs become less compacted, leading to release of the fluorophores from the NPs. Because each NP possessed many aptamer molecules, improved binding and specificity due to multivalent avidity were expected.19−21 To facilitate the conformational folding of the aptamer, a poly-T linker was introduced between the aptamer and the DAR surface. The poly-T linker is essential to reduce the nonspecific and/or steric interactions between C
DOI: 10.1021/acs.analchem.5b00569 Anal. Chem. XXXX, XXX, XXX−XXX
Article
Analytical Chemistry
Figure 1. Emission spectra of (a) sgc8c-R6GDAR NPs and (b) TD05-R101DAR NPs in PBS solutions at various pH values; (c) sgc8c-R6GDAR NPs in the presence of various numbers of CCRF-CEM cells; and (d) TD05-R101DAR NPs in the presence of various numbers of Ramos cells. Insets in parts c and d show the fluorescence intensities of sgc8c-R6GDAR NPs (excitation/emission wavelengths = 500/550 nm) and TD05R101DAR NPs (excitation/emission wavelengths = 560/600 nm) versus CCRF-CEM (y = 1.0246x + 120.15, R2 = 0.98) and Ramos (y = 1.0372x + 641.39, R2 = 0.99) cell numbers, respectively. Fluorescence is plotted in an arbitrary unit (a.u.).
intensities of the sgc8c-R6GDAR and TD05-R101DAR upon increasing the number of CCRF-CEM and Ramos cells, respectively, revealing that the apt-DAR NPs were encapsulated in the target cells. Linear relationships of the fluorescence intensities against number of cells over the ranges of 1 500− 30 000 were both obtained for CCRF-CEM (R2 = 0.98) and Ramos cells (R2 = 0.99). The two probes provided detection limits at a signal-to-noise ratio of 3 for CCRF-CEM and Ramos cells of 44 and 79 cells mL−1, respectively. They are comparable to the reported data; for example, 250 cells for the two cells when using aptamer-conjugated fluorescent NPs41 and 50 cells mL−1 for CCRF-CEM17 and Ramos cells18 provided by applying a quantum-dot sensor. Although an impedimetric approach allowed detection of 10 to 105 mL−1 HeLa cells,42 a longer analysis time (50 min) was required. Our approach, on the other hand, could be done within 10 min, reducing the loss of cell viability and minimizing possible contamination from other types of cells.43,44 As a result, false-positive results were expected to be minimized. We then investigated the effect of aptamer density of each DAR NP on the sensitivity of cell detection. Figure S4 in the Supporting Information shows the sensitivity of the apt-DAR NPs increased upon increasing the aptamer density. The aptamer at a higher density on the surface of each NP resulted in higher sensitivity, mainly due to more binding sites and multivalent avidity for cell recognition.38 No significantly improved sensitivity was observed using the Apt-DAR NPs when the aptamer at a concentration higher than 1.0 nmol was employed for conjugation. Therefore, with respect to sensitivity, the Apt-DAR NPs conjugated with 1.0 nmol of apatamer were used for all further experiments. To apply sgc8c-R6GDAR and TD05-R101DAR NPs for identification and detection of CCRF-CEM and Ramos cells, the two probes at various ratios were evaluated at pH 5.0.
molecules on the cell surfaces and the aptamer molecules of the NPs.20,38 The sizes of the DAR and apt-DAR NPs were similar, both with ∼300 nm in diameter (not shown). Since the retro-selfquenching mechanism of DARs is highly dependent on the protonation of pendant amino groups,30 the fluorescence intensities of sgc8c-R6GDAR and TD05-R101DAR NPs at various pH values were recorded to ensure that the pHdependent fluorescence enhancement still existed after conjugation with the aptamers. As shown in Figure 1a,b, the fluorescence intensities of the two apt-DAR NPs over the pH range 4.0−8.0 increased upon decreasing pH values, which are similar to that found in the R6GDAR and R101DAR NPs. The results reveal that introduction of the aptamer molecules to the two DAR NPs did not alter their fluorescence retro-selfquenching characteristic significantly. Detection of Cancer Cells. The cytotoxicity of sgc8cR6GDAR and TD05-R101DAR NPs was evaluated separately on CCRF-CEM, Ramos, and NIH-3T3 cells as shown in Figure S2 in the Supporting Information. Only slight loss in the viabilities of these normal and cancer cells were observed after a 24-h treatment, showing their great biocompatibility. The sgc8c-R6GDAR and TD05-R101DAR NPs were then used to detect CCRF-CEM and Ramos cells in PBS by monitoring their fluorescence changes at 550 and 600 nm, respectively. Although upon increasing incubation time the sensitivity increased as a result of more NPs located inside the cells, their selectivity toward the corresponding cells decreased (Figure S3 in the Supporting Information). The chance for the NPs entered the “target” and “non-target” cells through various channels such as macropinocytosis increased with increasing incubation time.39,40 To optimize sensitivity and selectivity, 5 min was chosen as the optimal incubation time in this study. Figure 1c,d displays increased fluorescence D
DOI: 10.1021/acs.analchem.5b00569 Anal. Chem. XXXX, XXX, XXX−XXX
Article
Analytical Chemistry
Figure 2. (a) Fluorescence intensities of sgc8c-R6GDAR and TD05-R101DAR NPs in a solution containing 7500 CCRF-CEM or Ramos cells. (b) Fluorescence images of CCRF-CEM and Ramos cells. Fluorescence intensities of (c) sgc8c-R6GDAR and (d) TD05-R101DAR NPs versus numbers of CCRF-CEM (y = 1.3091x + 561.67, R2 = 0.98) and Ramos (y = 1.3288x + 653.73, R2 = 0.98) cells, respectively. The cells were separately incubated with sgc8c-R6GDAR (20 μg mL−1) and TD05-R101DAR (20 μg mL−1) NPs in PBS at 25 °C for 5 min. Other conditions are the same as in Figure 1
intensity against a number of cells over the ranges of 1500− 7500 were obtained for CCRF-CEM (R2 = 0.98) and Ramos (R2 = 0.98) cells. The mixed probes provided detection limits at a signal-to-noise ratio of 3 for CCRF-CEM and Ramos cells of 64 and 103 cells mL−1, respectively. The sensitivity and linearity of mixed probe are comparable to that obtained when using the individual probes. Our results revealed that like an aptamerconjugated fluorescence resonance energy transfer NP assay,45 our approach is selective and sensitive for detection of the two targeted cells, showing their potential for screening cancer types in the blood samples from leukemia patients. Practicality of the Apt-DAR NPs. The stability of nucleic acids toward enzymes and nucleases is an important aspect affecting their diagnostic applications. The fluorescence intensities of sgc8c-R6GDAR or TD05-R101DAR were similar in the absence and presence of DNase I as shown in Figure S6 in the Supporting Information, mainly because the salt and steric effects minimized the access of DNase I toward the surfaces of NPs and thus the aptamer molecules on the surfaces retained their activity.24,25 Additionally, the highly negative charged surfaces of Apt-DAR NPs were responsible for their increased stability.20,38,46 Figure 3a displays the emission intensity at 550 nm is proportional to the number (1 500− 7 500 counts) of spiked CCRF-CEM cells, with reproducibility (standard deviation values
