Two-Photon Ratiometric Fluorescent Sensor Based on Specific

Nov 21, 2013 - features two independent emission peaks located at 515 nm for. ATD and 650 nm for QDs, respectively, under two-photon excitation at 800...
1 downloads 0 Views 2MB Size
Download PDF
Article pubs.acs.org/ac

Two-Photon Ratiometric Fluorescent Sensor Based on Specific Biomolecular Recognition for Selective and Sensitive Detection of Copper Ions in Live Cells Yan Fu,† Changqin Ding,† Anwei Zhu,† Zifeng Deng,† Yang Tian,*,† and Ming Jin*,‡ †

Department of Chemistry, Tongji University, Siping Road 1239, Shanghai 200092, P. R. China School of Materials Science and Engineering, Tongji University, Caoan Road 4800, Shanghai 201804, P. R. China

‡

S Supporting Information *

ABSTRACT: In this work, we develop a ratiometric two-photon fluorescent probe, ATD@QD-E2Zn2SOD (ATD = amino triphenylamine dendron, QD = CdSe/ZnSe quantum dot, E2Zn2SOD = Cu-free derivative of bovine liver copper−zinc superoxide dismutase), for imaging and sensing the changes of intracellular Cu2+ level with clear red-to-yellow color change based on specific biomolecular recognition of E2Zn2SOD for Cu2+ ion. The inorganic−organic nanohybrided fluorescent probe features two independent emission peaks located at 515 nm for ATD and 650 nm for QDs, respectively, under two-photon excitation at 800 nm. Upon addition of Cu2+ ions, the red fluorescence of QDs drastically quenches, while the green emission from ATD stays constant and serves as a reference signal, thus resulting in the ratiometric detection of Cu2+ with high accuracy by two-photon microscopy (TPM). The present probe shows high sensivity, broad linear range (10−7−10−3 M), low detection limit down to ∼10 nM, and excellent selectivity over other metal ions, amino acids, and other biological species. Meanwhile, a QD-based inorganic-organic probe demonstrates long-term photostability, good cell-permeability, and low cytotoxicity. As a result, the present probe can visualize Cu2+ changes in live cells by TPM. To the best of our knowledge, this is the first report for the development of a QD-based two-photon ratiometric fluorescence probe suitable for detection of Cu2+ in live cells.

I

excitation, has become attractive.4e,f TPM offers several advantages such as increased penetration depth (>500 μm), localized excitation, and low phototoxicity associated with the use of near-infrared excitation. Recently, semiconductor quantum dot (QD) has been considered as a mile stone of fluorescent materials with high quantum yield, bright and photostable fluorescence, a broad excitation spectrum but a narrow Gaussian emission at wavelengths.4e However, use of QDs for two-photon imaging and sensing is rarely reported. In this paper, we developed a two-photon fluorescent probe for ratiometric biosensing and imaging of Cu2+ ions in live cells. Two new strategies were developed in this work. First of all, we designed and synthesized two-photon fluorescent moleculeamino triphenylamine dendron (ATD), which shows an emission peak at 515 nm upon two-photon excitation at 800 nm. ATD was then hybrided onto CdSe/ZnSe QDs to form a two-photon ratiometric fluorescent probe ATD@QD. The probe features two independent emission peaks at 515 nm (green) and 650 nm (red), respectively, under two-photon

ntracellular copper ion (Cu2+) plays a critical role in physiological and pathological events,1 as a catalytic cofactor including superoxide dismutase, cytochrome c oxidase, and tyrosinase.1c Disruption of copper homeostasis is associated with various diseases including Menkes2 and Wilson’s diseases,3 familial amyotropic lateral sclerosis,4 Alzheimer’s disease,5 and prion diseases.6 Thus, development of analytical methods for detection of Cu2+ in biological systems is the key bottleneck to help understanding the complex contributions to healthy and disease states of Cu2+.7−10 Recently, several efficient strategies have been reported for Cu2+ detection with high selectivity and sensitivity.6b,d,e Fluorescent sensors have attracted more attention because they are available for bioimaging and biosensing directly in live cells, tissues, and animals.6a,b We have also developed a ratiometric fluorescence probe for intracellular sensing and imaging of Cu2+, with high sensitivity and accuracy.6a,7a However, most of these reported fluorescent probes require one-photon excitation using short-wavelength UV−vis light (350−500 nm). The utilization of UV−vis excitation2e−g may cause general problem of background fluorescence3a,b and scattering light,3c,d photodamage to biological samples, and photobleaching of fluorophores. In the past decades, two-photon microscopy (TPM), a new technique that utilizes two photons of lower energy for © 2013 American Chemical Society

Received: September 6, 2013 Accepted: November 21, 2013 Published: November 21, 2013 11936

dx.doi.org/10.1021/ac403527c | Anal. Chem. 2013, 85, 11936−11943

Analytical Chemistry

Article

Synthesis of ATD. As shown in Scheme S1 in the Supporting Information, 4-(bis(4-(4-(diphenylamino)styryl)phenyl)amino)benzaldehyde (TD) was synthesized according to the reported procedure.12c TD (0.5 mmol) was first added to a cold stirred mixture of poly(ethylene glycol) bis(amine) (PEG, 5 mmol), anhydrous methanol (4 mL), and anhydrous sodium sulfate (0.5 g) for 4 h. After completion of the reaction, sodium borohydride (1 mmol) was added in portions with stirring at room temperature for 10 h. The sodium sulfate was removed by filtration, and the resulting solution was concentrated and then extracted with dichloromethane and aqueous ammonium chloride (wt n%) (2:1 v/v). The organic phase was evaporated to dryness and purified by column chromatography on silica.12d Yield, 80%. 1H NMR (CDCl3, 400 MHz) δ(ppm): 7.40−7.37 (m, 9H, Ar), 7.26−7.29 (m, 12H, Ar), 7.09−7.14 (m, 12H, Ar), 6.97−7.06 (m, 8H, Ar), 6.95 (s, 2H), 5.37 (m, 2H), 4.68 (s, 2H), 4.24 (br, 3H), 3.65 (s, 200H), 3.17−3.04 (m, 4H), 2.91 (s, 2H). 13C NMR (CDCl3, 100 MHz) δ(ppm): 147.586, 129.250, 127.154, 124.420, 124.041, 123.703, 122.951, 77.342, 77.024, 76.707, 70.564, 53.406. Preparation of E2Zn2SOD. The Cu-free derivative E2Zn2SOD was prepared according to the method described by Cocco.12d Briefly, DDC was added to 0.1 mM SOD solutions, buffered with 0.1 M potassium phosphate at pH 7.4, at a final concentration of 0.5 mM. The mixture was incubated at 310 K for about 2 h until no further increase in absorbance at 450 nm was observed. Then, the yellow solution was centrifuged at 39 000g for 30 min, and the colorless supernatant was exhaustively dialyzed against doubly distilled water. Finally, the E2Zn2SOD solution was lyophilized and kept in the refrigerator for further use. Conjugation of ATD@QD and ATD@QD-E2Zn2SOD. In a small glass vial with a small stir-bar, 5 μL of 8 μM stock solution of Qdot carboxyl quantum dots were diluted to 2 mL using 10 mM in PBS buffer (pH = 7.4) under stirring. After 2 min, 100 μL of 50 μM ATD and 10 μL of 300 nM EDC were added to the Qdot carboxyl quantum dots reagent for 0.5 h to form ATD@QD. Next, 200 μL of 5 × 10−4 M NTA-Ni was added under vigorous stirring for 4 h to obtain ATD@QDNTA. Then, ATD@QD-NTA was centrifuged with a Millpore ultrafiltration centrifuge tube at 10 000g for 5 min to remove any excess unbound NTA-Ni and ATD. Afterward, residual ATD@QD-NTA was redispersed in PBS buffer. Then, 200 μL of Cu-free derivative E2Zn2SOD (10−4 M) was added to the above ATD@QD-NTA for 12 h to form ATD@QDE2Zn2SOD. Instruments and Measurements. NMR spectra were collected in CDCl3 at 25 °C on a Bruker AV-400 spectrometer. All chemical shifts were recorded in the standard δ notation of parts per million with chemical shifts reported in ppm. Optical absorption spectrum was noted on an Agilent 8453 UV−vis spectrometer using a quartz cuvette having a 1 cm path length. Infrared spectroscopic data were collected by a NEXUS470 infrared spectrometer (Nicolet). One-photon fluorescence spectra were obtained on a Hitachi F-2700 spectrofluorimeter equipped with a 450 W xenon lamp. Samples for absorption and emission measurements were conducted in 1 cm × 1 cm quartz cuvettes. The two-photon excited fluorescence spectra were measured at 800 nm excitation wavelength on a spectrometer (Horiba model iHR 550) and the pump laser beam came from a mode-locked Ti:sapphire laser (Coherent Mira 900) with a pulse duration of 80 fs and a repetition rate of 76 MHz. Two-photon fluorescence images of probe-labeled

excitation at the wavelength of 800 nm. Next, Cu-free derivative of bovine liver copper−zinc superoxide dismutase (SOD); E2Zn2SOD (E designates an empty site) was designed and prepared as the unique receptor specific for Cu2+ because E2Zn2SOD can catch Cu2+ with high specificity to reconstitute SOD.11 The affinity (association constant) of E2Zn2SOD toward Cu2+ is K = 4 × 1015 M−1.11e Then, the specific E2Zn2SOD was conjugated with the ATD@QD to generate ATD@QD-E2Zn2SOD fluorescent probe. The functionalized probe demonstrates high selectivity for Cu2+ over other metal ions, amino acids, and other biological species, leading to red fluorescence of QDs quenching, whereas the green fluorescence of ATD stays constant, as shown in Scheme 1. ATD dye is inert Scheme 1. Schematic Illustration for the Working Principle of Two-Photon Ratiometric Imaging and Sensing of Cu2+

to Cu2+ and only serves as reference signal for providing built-in correction, thus avoiding environmental effects. As a consequence, variations of the two fluorescence intensities display clear color change from red to yellow upon addition of Cu2+, resulting in a two-photon ratiometric fluorescent sensor for Cu2+. Finally, the present sensor with high sensitivity and selectivity enables the imaging and sensing of Cu2+ in live cells by TPM.

■

EXPERIMENTAL SECTION Materials and Reagents. Qdot 655 ITK carboxyl-quantum dots (8 μM) and cell culture media were purchased from Invitrogen Corporation. Cell culture reagents were purchased from Gibco. The Annexin V-FITC Apoptosis Assay Detection Kit was purchased from KeyGEN Biotech. Methyl thiazolyl tetrazolium (MTT), amino acids (99%), 1-ethyl-3-(3dimethylaminopropyl)carbodiimide (EDC), N-hydroxysuccinimide (NHS), poly(ethylene glycol) bis(amine) (PEG), superoxide dismutase (SOD) from bovine liver, and nitrilotriacetic acid disodium salt (NTA) were obtained from Sigma. Metal salts, ethanol, dichloromethane, methanol, and silica gel (200− 300 meshes) were purchased from Sinopharm Chemical Regent Co. All chemicals were commercial ly available and of analytical grade. Solutions of metal ions were all prepared from their chloride salts. Ultrapure water was used from a Millipore water purification system. 11937

dx.doi.org/10.1021/ac403527c | Anal. Chem. 2013, 85, 11936−11943

Analytical Chemistry

Article

Figure 1. (A) TEM images of CdSe/ZnSe QDs. Inset shows lattice spacing of CdSe/ZnSe QDs. (B) FT-IR spectra of (I) CdSe/ZnS QDs, (II) ATD@QD, (III) ATD@QD-NTA, and (IV) ATD@QD-E2Zn2SOD. (C) One-photon (curve I, solid line, 400 nm excitation) and two-photon (curve II, black dot, 800 nm excitation) fluorescence spectra of the synthesized ATD. Inset: the quadratic relationship of the observed two-photon luminescent intensity of ATD with excitation laser wavelength at 800 nm. (D) One-photon (curve I, solid line, 400 nm excitation) and two-photon (curve II, short dot line, 800 nm excitation) fluorescence spectra of ATD@QD-E2Zn2SOD QD.

crystals. Absorbance was measured at 490 nm in a Multiskan MK3. The number of viable cells was determined by an MTT assay.4f,g Flow Cytometry. HeLa cells were incubated with ATD@ QD-E2Zn2SOD (0, 5, 10, 15, 20 nM) for 48 h. Cells floating in the cell medium were collected by centrifuge while adherent cells were collected by treating with trypsin-EDTA. After washing with PBS, cells were then stained with FITC-annexin V (Molecular Probe) and Propidium Iodide (PI, Aldrich) following the standard protocol. A Becton-Dickinson flow cytometer was used for the flow cytometry (FACS) measurements. Confocal Imaging in Live Cells. HeLa cells were cultured in Dulbecco’s Modified Eagle’s medium (DMEM) supplemented with 10% fetal bovine serum, 100 units mL−1 penicillin and 100 μg mL−1 streptomycin. One day before imaging studies in vivo, the cultured cells were passaged and plated on a Petri dish. The culture media were exchanged for DMEM containing ∼10 nM nanoprobe and incubated for 1 h at 37 °C, after which the nanoprobe-loaded media was replaced with 1× PBS and the cells were washed with 1× PBS (pH = 7.4) three times to remove the extracellular remaining probe and mounted on the microscope stage, confocal fluorescence imaging including an XY-scan; a spectrum-scan was performed on a Leica TCS SP8, and two-photon images were excited with an excitation wavelength at 800 nm and emissions were then collected

cells were obtained with spectral confocal and multiphoton microscopes (Leica TCS SP8) with ×10 dry and ×63 oil objectives, numerical aperture (NA) = 0.4 and 1.4 and a DMI 6000 microscope (Leica) by exciting the probes with a modelocked titanium-sapphire laser source (Mai Tai DeepSee, 80 MHz, 90 fs) at wavelength 800 nm and output power 2920 mW. In order to observe the morphology of CdSe/ZnS QDs, transmission electron microscopy (HR-TEM, JEOL 2100, Japan) were employed. All the experiments were performed at room temperature. Cell Cultures and MTT Assay. HeLa cells were cultured in Dulbecco’s Modified Eagle’s medium (DMEM) containing high glucose supplemented with 10% fetal bovine serum, 100 units mL−1 penicillin and 100 μg mL−1 streptomycin. When in the proliferative period, HeLa cells (∼3 × 105 cell/mL) were dispersed within replicate 96-well microliter plates to a total volume of 100 μL/well and maintained at 310 K in a 5% CO2/ 95% air incubator for 24 h. Then, the culture media was removed and the cells were incubated in culture medium containing the as-prepared ATD@QD-E2Zn2SOD with different concentrations (5, 10, 15, 20, 25 nM) for 24 and 48 h and washed with the culture medium. An amount of 100 μL of the new culture medium containing MTT (10 μL, 5 mg mL−1) was then added, followed by incubating for 4 h to allow the formation of formazan dye. After removing the medium, 150 μL of DMSO was added to each well to dissolve the formazan 11938

dx.doi.org/10.1021/ac403527c | Anal. Chem. 2013, 85, 11936−11943

Analytical Chemistry

Article

Figure 2. (A) Two-photon fluorescence spectra of ATD@QD-E2Zn2SOD nanoprobe (20 nM) with the addition of 10−8, 10−7, 10−6, 10−5, 10−4, 10−3, and 5 × 10−3 M Cu2+ ions under laser excitation of 800 nm. (B) The relationship between the integrated fluorescence intensity Fred/Fgreen (Fgreen = F480−580 nm, Fred = F600−700 nm) and Cu2+ concentration.

QD was then conjugated with Ni-NTA, which was evident by FTIR and XPS data. The peak at ∼1470 cm−1 in curve III (Figure 1B) corresponds to the −NH streching mode, implying that Ni-NTA was conjugated onto ATD@QD. Furthermore, the presence of Ni2p3 and Ni2p1 peaks located at ∼855.8 and ∼874.3 eV in XPS data (Supporting Information, Figure S4) suggested that Ni-NTA was successfully modified onto ATD@ QD to form ATD@QD-NTA. On the other hand, the Cu-free derivative E2Zn2SOD was prepared according to the method described by Cocco and Calabrese16b and finally conjugated with Ni- NTA due to metal-chelate affinity to form [email protected],16c,d The peaks observed at ∼1645 cm−1 (Amide I CO) and ∼1539 cm−1 (Amide II −COO−) in the FTIR spectrum (curve IV, Figure 1B) demonstrates the successful attachment of E2Zn2SOD onto the surface of ATD@QD-NTA. In addition, the restoration of the S2p peak (Supporting Information, Figure S4) of the thiol group or disulfide at ∼163.0 eV was evidence of the exact modification on the ATD@QD-NTA surface, because this peak greatly decreased after the modification of ATD and Ni-NTA onto QDs surface. Upon both one-photon and two-photon excitation, the synthesized ATD exhibits a fluorescence maximum at ∼515 nm (Figure 1C). The fluorescence quantum yield (ΦF) was determined by comparing the integrated emission with that of rhodamine 6G ethanol solutions with equal optical density at the excitation wavelength. The quantum yield values were corrected for the differences in refractive index between ethanol and water.18d,19f,g The fluorescence quantum yield (ΦF) of ATD was calculated to be 24.2 ± 1.0%. As shown in the inset of Figure 1C, the quadratic relationship between the excitation laser power and the luminescence intensity is obvious, thus confirming that the excitation with two near-infrared photons was indeed responsible for the observed visible luminescence of ATD. On the other hand, the aqueous solution of QDs was found to be strongly emissive and shows an emission maximum at ∼650 nm (Figure S5A in the Supporting Information). Using the same method, the fluorescence quantum yield (ΦF) of CdSe/ZnSe QDs was determined to be 32.3 ± 1.0%.12a Thus, both the obtained ATD@QD (Figure S5B in the Supporting Information) and ATD@QD-E2Zn2SOD (Figure 1D) show well-resolved dual emission bands centered at ∼515 and ∼650 nm, respectively. Using a femtosecond (fs) fluorescence measurement technique, the two-photon action cross section (σ2P) of QDs and ATD at 800 nm were estimated to be 16 000

simultaneously for the two channel scan units in the 480−580 nm range and 600−700 nm range. Then, the exogenous Cu2+ ion source, 50 μM CuCl2, and 100 μM PDTC were added directly to the Petri dish and incubated for 40 min, after that 50 μM was added again on the microscope stage to visualize changes in the fluorescence intensity ratios in cells treated with the exogenous Cu2+ ion source, finally treated with 100 μM EDTA for 0.5 h

■

RESULTS AND DISCUSSION A typical TEM image of CdSe/ZnSe QDs is shown in Figure 1A. The CdSe/ZnSe QDs demonstrate ahomogeneous rod shape, with ∼2 nm in width and ∼10 nm in length. A further close observation (inset in Figure 1A) reveals that CdSe/ZnSe QDs in the micrographs show well resolved lattice fringes with a measured lattice spacing of 0.4 nm.12a The infrared (IR) spectrum of the CdSe/ZnSe QDs is given in Figure 1B. As shown in curve I, the band observed at ∼1653 cm−1 indicates the existence of carbonyl (CO) groups,13 while the peak located at ∼3430 cm−1 corresponds to the OH stretching mode. These data demonstrate that the CdSe/ZnSe QDs are surrounded by −COOH and/or −OH groups. Aminocontaining ATD was prepared via the reduction of triphenylamine dendro (TD) using sodium borohydride (NaBH4) as a reducing agent (see the Supporting Information).14 The structure of this ATD was confirmed by NMR (see the Supporting Information, Figures S1 and S2),15 and IR (Figure S3 in the Supporting Information). Then, the synthesized ATD was stably conjugated onto CdSe/ZnSe QDs by using 1-ethyl3-(3-dimethylaminopropyl) carbodiimide hydrochloride (EDC) and N-hydroxysuccinimide (NHS) as coupling reagents to form [email protected] The hybrid of ATD on the surface of CdSe/ZnSe QDs was confirmed by FT-IR and X-ray photoelectron spectroscopy (XPS). As shown in Figure 1B (curve II), the peak observed at ∼1674 cm−1 is attributed to the CO group, while that located at ∼1469 cm−1 indicates the existence of the −NH group. Moreover, the peak observed at ∼3416 cm−1 corresponds to the stretching vibration of the −NH group. These data confirmed the successful attachment of ATD on the CdSe/ZnSe QDs surface, which was also supported by XPS data (Supporting Information, Figure S4). An obvious peak was observed at ∼400.1 eV for N1s after ATD was conjugated onto QDs,17c which was not obtained at bare QDs surface. This observation suggests that ATD attaches onto QDs well to form ATD@QD. For further preparation, ATD@ 11939

dx.doi.org/10.1021/ac403527c | Anal. Chem. 2013, 85, 11936−11943

Analytical Chemistry

Article

Figure 3. (A) Fluorescence responses of 20 nM ATD@QD-E2Zn2SOD toward various metal ions. Black bars represent the addition of an excess of metal ions (1 mM for Na+, K+, Ca2+, and Mg2+; 10 μM for other cations) to a 20 nM solution of ATD@QD-E2Zn2SOD. White bars represent the subsequent addition of 2 μM Cu2+ to the solution. (B) Fluorescence responses of 20 nM ATD@QD-E2Zn2SOD toward various amino acids and other biological species. Black bars represent the addition of 10 μM amino acids and 10 μM DA, AA, glucose, H2O2, hemoglobin (HB), myoglobin (MB), and cytochrome c to a 20 nM solution of ATD@QD-E2Zn2SOD (pH = 7.4). White bars represent the subsequent addition of 2 μM Cu2+ to the solution. Excitation was provided at 800 nm. (C) Effect of pH value on fluorescent responses of ATD@QD-E2Zn2SOD (20 nM). The data shown as mean (±S.D., n = 3) ranging from 5.4 to 9.3. (D) MTT assay of HeLa cells in the presence of different concentrations of ATD@QDE2Zn2SOD (5, 10, 15, 20, 25 nM) for 24 h (black) and 48 h (gray) at 37 °C, respectively.

± 2000.0 GM17a and 1200 ± 100.0 GM17b (Goeppert-Mayer unit, with 1 GM = 10−50 cm 4 s/photon), respectively.18 The response of the dual-emission fluorescent probe toward Cu2+ ions (CuCl2) was then investigated to prove the working principle, as demonstrated in Figure 2. Upon addition of Cu2+ ions, the intensity of red emission (from 600 to 700 nm) from the CdSe/ZnS QDs shows continuous quenching, whereas that of green emission (480−580 nm) from ATD remains constant. The integrated fluorescence intensity ratio between the two emission channels Fred/Fgreen (Fred = F600−700 nm and Fgreen = F480−580 nm) gradually decreases with increasing concentration of Cu2+ ions. As shown in Figure 2B, the signal ratio shows good linearity with Cu2+ concentration in the range of 10−7− 10−3 M. The detection limit was calculated to be ∼10 nM (based on a signal-to-noise ratio of S/N = 3). Compared with those of previously reported, the probe exhibited higher sensitivity, broader linear range, and lower detection limit for Cu2+ detection.5b−d Therefore, the integrated fluorescence intensity ratio (Fred/Fgreen) can be used to quantify Cu2+ through ratiometric measurement, which is superior to “turnoff” and “turn-on” methods because it provided a built-in correction to avoid environmental effects and high sensitivity. The interaction between E2Zn2SOD and Cu2+ was further studied by XPS (Supporting Information, Figure S4). Two clear peaks ascribed to Cu2p were observed at ∼934.2 and ∼954.5 eV in XPS only after Cu2+ was reconstituted to the ATD@QDE2Zn2SOD surface. In addition, the time-resolved fluorescence (TRF) signals of ATD@QD-E2Zn2SOD were probed in the

absence and presence of Cu2+ at 650 nm with an excitation at 400 nm (Figure S6A in the Supporting Information).18e,f In the absence of Cu2+, the fluorescence decays single exponentially by a 11.6 ns time constant, which is the lifetime of QDs in the nanohybrids. However, the lifetime decreased to 11.0 and 9.9 ns after 2 μM and 4 μM Cu2+ were, respectively, added into ATD@QD-E2Zn2SOD probes. Compared to ATD@QDE2Zn2SOD, the lifetime of bare QDs was 24.5 ns, as demonstrated in Figure S6B in the Supporting Information. However, the lifetime slightly decreased to 6.0 and 5.5 ns after the addition of 2 μM and 4 μM Cu2+, respectively. These results imply the different quenching mechanism of bare QDs and ATD@QD-E2Zn2SOD with the addition of Cu2+. For further understanding, with the quenching mechanism of ATD@QD-E2 Zn 2 SOD with the addition of Cu2+, the fluorescence quenching data were analyzed by the Stern− Volmer equation. In our ATD@QD-E2Zn2SOD system, the plot of F0/F versus Cu2+ concentration did not fit a conventional linear Stern−Volmer equation (Figure S7A in the Supporting Information). A steep upward curvature may indicate that both dynamic and static quenching seem to act together, suggesting a more complex quenching model.18g,h A good linear relationship (R = 0.989) was observed up to a Cu2+ concentration ranging from 0 to 4 × 10−4 M (Figure S7B in the Supporting Information) when using a modified Stern−Volmer plot for mechanisms where both dynamic and static quenching act together.18g,i log(F0/F ) = KSV[Q] + C 11940

(1)

dx.doi.org/10.1021/ac403527c | Anal. Chem. 2013, 85, 11936−11943

Analytical Chemistry

Article

Figure 4. (A) Two-photon confocal fluorescence image, (B) bright field image, and (C) overlapped fluorescence and bright field image of HeLa cells incubated with 10 nM ATD@QD-E2Zn2SOD for 2 h; (D) normalized fluorescence emission from ATD@QD-E2Zn2SOD nanohybrid in HeLa cells. (E,F) Two-photon ratiometric images of HeLa cells after treatment with (E) 50 μM CuCl2 and (F) 100 mM CuCl2 for 1 h; (G) two-photon ratiometric confocal fluorescence image of HeLa cells then treated with 100 mM EDTA for 0.5 h; (H) integrated fluorescence intensity from 600 to 700 nm over that from 480 to 580 nm in HeLa cells (a, blank, b, 50 μM CuCl2; c, 100 μM CuCl2; d, 100 mM EDTA).

where F0 and F are the fluorescence intensities of ATD@QDE2Zn2SOD in the absence and presence of Cu2+, [Q] is denoted as the concentration of Cu2+, and C is the constant. From the slope of the calibration plot shown in Figure S7B in the Supporting Information, KSV was evaluated to be 1.24 × 103 M−1. Thus, the quenching mechanism of the ATD@QDE2Zn2SOD with Cu2+ is possibly explained in terms of facilitating nonradiative e−/h+ recombination anniliation on the surface of QDs through a effective electron transfer process between surface functional E2Zn2SOD and Cu2+ based on the strong affinity. In addition, the absorption of the reconstituted SOD (addition of Cu2+ into the E2Zn2SOD solution) observed in the range of 600−750 nm was overlapped with the emission of QDs from 600 to 700 nm (Figure S8 in the Supporting Information), thus the quenching mechanim of QDs after the addition of Cu2+ might be explained by the excitation energy transfer19b−e from the QDs to the copper d-orbital. Because no fluorescence emission was observed for the reconsituted SOD solution, the direct observation of fluorescence resonance energy transfer was not obtained. Accordingly, we can conclude that the quenching mechanism of ATD@QD-E2Zn2SOD probe toward Cu2+ may be ascribed to energy transfer and/or electron transfer, although the clear mechanism is still unknown at the present stage. To study the selectivity of the ATD@QD-E2Zn2SOD fluorescent sensor for Cu2+ in a complex intracellular environment, the interferences from metal ions, amino acids, and other biological species were examined.20 First, the selectivity experiments were carried out by monitoring the intensity ratio (Fred/Fgreen) of the probe in the presence of millimolar concentrations of Na+, K+, Ca2+, and Mg2+, 10 μM concentrations of Mn2+, Fe2+, Co2+, Ni2+, Zn2+, and Cu+, which may coexist in the living system. As shown in Figure 3A, no obvious signals were observed for the other metal ions, compared with that obtained for Cu2+. Furthermore, these potential metal ion interferences showed negligible effects on the signal for Cu2+ sensing. Taking into account that amino

acids in the biological system are capable of interacting with a lot of metal cations, several typical amino acids were also examined. Little effect on the intensity ratio of the probes was observed (Figure 3B) after their exposure to 10 μM concentrations of amino acids. In addition, negligible responses and effects were observed for the two-photon ATD@QDE2Zn2SOD fluorescence probe toward other biological species such as glucose, dopamine (DA), ascorbic acid (AA), proteins, and so on (Figure 3B). All these data demonstrate the excellent selectivity of the developed two-photon ratiometric probe for determination of Cu2+ against other metal ions, amino acids, and potential biological species which may exist in an intracellular system, due to the specific interaction of the designed molecule E2Zn2SOD with Cu2+ to reconstitute SOD. In addition, photostability of the fluorescent responses was investigated at 515 and 650 nm, respectively, under 800 nm laser excitation up to around 2 h in 10 mM phosphate buffered solution (PBS). Negligible changes (