Phosphate-Modified TiO2 Nanoparticles for Selective Detection of

Jun 12, 2007 - A nano-sized Au electrode fabricated using lithographic technology for electrochemical detection of dopamine. Yue Zhao , Shu-Hong Li , ...
0 downloads 0 Views 124KB Size
Download PDF
7880

Langmuir 2007, 23, 7880-7885

Phosphate-Modified TiO2 Nanoparticles for Selective Detection of Dopamine, Levodopa, Adrenaline, and Catechol Based on Fluorescence Quenching Hsin-Pin Wu,† Tian-Lu Cheng,‡,§ and Wei-Lung Tseng*,†,§ Department of Chemistry, National Sun Yat-sen UniVersity, Taiwan, Faculty of Biomedical Science and EnVironmental Biology, Kaohsiung Medical UniVersity, Kaohsiung, Taiwan, and National Sun Yat-sen UniVersity-Kaohsiung Medical UniVersity Joint Research Center, Kaohsiung, Taiwan ReceiVed February 26, 2007. In Final Form: May 3, 2007 For the first time, an aqueous solution, comprising 6-nm phosphate-modified titanium dioxide (P-TiO2) nanoparticles (NPs) and fluorescein, has been used for sensing dopamine (DA), levodopa (L-DOPA), adrenaline, and catechol. The complexes obtained by means of chelation of surface Ti(IV) ions with an enediol group exhibit strong absorption at 428 nm; thus, they can be designed as efficient quenchers for fluorescein. The fluorescence of a fluorescein solution containing 1.4 mM P-TiO2 NPs at pH 8.0 decreases if the solution comprises DA, L-DOPA, adrenaline, and catechol, but not noradrenaline, ascorbic acid, and salicylic acid. We consider that P-TiO2 NPs have a number of advantages over bare TiO2 NPs, such as ease of preparation, high selectivity, and high stability. By measuring fluorescence quenching, the limits of detection at a signal-to-noise ratio of 3 are calculated as 33.5, 81.8, 20.3, and 92.1 nM for DA, L-DOPA, adrenaline, and catechol, respectively. In contrast, UV-vis absorption reveals the relatively poor sensitivity of these compounds. We have validated the applicability of our method by means of analyses of DA in urine samples. High-performance liquid chromatography in combination with an electrochemical cell has been used to further confirm our results. We believe that this approach has great potential for diagnostic purposes.

Introduction Nanoparticles (NPs) are becoming increasingly attractive materials for use in biosensors because they not only exhibit nanoscale dimensions comparable to the dimensions of biomacromolecules but also possess size-dependent optical and electronic properties.1-4 Gold (Au) NPs are one of the most popular materials and provide high-affinity binding for biomolecules containing thiol (SH) or amino (NH2) groups.5 For example, thiolated oligonucleotides can be adsorbed spontaneously onto Au surfaces to generate well-organized self-assembled monolayers.6 Oligonucleotide-functionalized Au NPs aggregate in the presence of a complementary target DNA, leading to a change in solution color. In addition, crown-5-functionalized Au NPs,7 antigen-modified Au NPs,8 and dextran-coated Au NPs9 have been used for analyzing metal ion, antibodies, and concanavalin A, respectively. Methods based on molecular recognition may also be used with NPs of noble metals other than Au and * Corresponding author. Address: Department of Chemistry, National Sun Yat-sen University, 70, Lien-hai Road, Kaohsiung, Taiwan 804. E-mail: [email protected]. Fax: 011-886-7-3684046. † National Sun Yat-sen University. ‡ Kaohsiung Medical University. § National Sun Yat-sen University-Kaohsiung Medical University Joint Research Center. (1) Bruchez, M., Jr.; Moronne, M.; Gin, P.; Weiss, S.; Alivisatos, A. P. Science 1998, 281, 2013-2016. (2) Chan, W. C. W.; Nie, S. Science 1998, 281, 2016-2018. (3) You, C.-C.; Verma, A.; Rotello, V. M. Soft Matter 2006, 2, 190-204. (4) Veiseh, M.; Zareie, M. H.; Zhang, M. Langmuir 2002, 18, 6671-6678. (5) (a) Bhattacharjee, R. R.; Chakraborty, M.; Mandal, T. K. J. Phys. Chem. B 2006, 110, 6768-6775. (b) Niemeyer, C. M. Angew. Chem., Int. Ed. 2001, 40, 4128-4158. (6) Rosi, N. L.; Mirkin, C. A. Chem. ReV. 2005, 105, 1547-1562. (7) (a) Lin, S.-Y.; Liu, S.-W.; Lin, C.-M.; Chen, C.-h. Anal. Chem. 2002, 74, 330-335. (b) Lin, S.-Y.; Chen, C.-h.; Lin, M.-C.; Hsu, H.-F. Anal. Chem. 2005, 77, 4821-4828. (8) Thanh, N. T. K.; Rosenzweig, Z. Anal. Chem. 2002, 74, 1624-1628. (9) (a) Lee, S.; Pe´rez-Luna, V. H. Anal. Chem. 2005, 77, 7204-7211. (b) Aslan, K.; Zhang, J.; Lakowicz, J. R.; Geddes, C. D. J. Fluoresc. 2004, 14, 391-400.

semiconductor colloidal NPs.10 Platinum (Pt) NPs modified with thiolated nucleic acid have been used for detecting target DNA by means of catalytic reactions; these NPs generate chemiluminescence in the presence of H2O2 and luminol.11 On the basis of fluorescence resonance energy transfer, immunocomplexes between antigens and antibodies have been investigated by using two types of quantum dots.12 The tunability of the localized surface plasmon resonance of Ag nanotriangles has been shown to facilitate the understanding of biomolecular interactions.13 Titanium dioxide (TiO2) NPs are promising materials for applications such as photocatalysts,14 photovoltaic cells,15 air filters,16 and optical filters,17 although very little research has been conducted to understand their impact on biosensors. Interestingly, biomolecules having enediol,18 phosphate,19 and carboxylic20 groups have been known to bind to the surface of TiO2 NPs. The surface modification of TiO2 NPs with enediol ligands such as catechol and dopamine (DA) provide strong absorption bands in the visible region.21 According to these results, (10) (a) Costa-Ferna´ndez, J. M.; Pereiro, R.; Sanz-Medel, A. TrAC, Trends Anal. Chem. 2006, 25, 207-218. (b) Willner, I.; Baron, R.; Willner, B. Biosens. Bioelectron., in press. (11) (a) Polsky, R.; Gill, R.; Kaganovsky, L.; Willner, I. Anal. Chem. 2006, 78, 2268-2271. (b) Gill, R.; Polsky, R.; Willner, I. Small 2006, 2, 1037-1041. (12) Wang, S.; Mamedova, N.; Kotov, N. A.; Chen, W.; Studer, J. Nano Lett. 2002, 2, 817-822. (13) (a) Haes, A. J.; Stuart, D. A.; Nie, S.; Duyne, R. P. V. J. Fluoresc. 2004, 14, 355-367. (b) Haes, A. J.; Hall, W. P.; Chang, L.; Klein, W. L.; Van Duyne, R. P. Nano Lett. 2004, 4, 1029-1034. (14) Hoffmann, M. R.; Martin, S. T.; Choi, W.; Bahnemann, D. W. Chem. ReV. 1995, 95, 69-96. (15) Uchida, S.; Chiba, R.; Tomiha, M.; Masaki, N.; Shirai, M. Electrochemistry 2002, 70, 418-420. (16) Inami, Y.; Takai, H.; Shiratori, S. Trans. Mater. Res. Soc. Jpn. 2002, 27, 403-406. (17) Lin, Y.; Wang, A.; Claus, R. J. Phys. Chem. B 1997, 101, 1385-1388. (18) Tae, E. L.; Lee, S. H.; Lee, J. K.; Yoo, S. S.; Kang, E. J.; Yoon, K. B. J. Phys. Chem. B 2005, 109, 22513-22522. (19) Chen, C.-T.; Chen, Y.-C. Anal. Chem. 2005, 77, 5912-5919. (20) Granot, E.; Patolsky, F.; Willner, I. J. Phys. Chem. B 2004, 108, 58755881.

10.1021/la700555y CCC: $37.00 © 2007 American Chemical Society Published on Web 06/12/2007

Using P-TiO2 NPs to Detect Catecholamines

Langmuir, Vol. 23, No. 14, 2007 7881

the selective analysis of catecholamines in biological samples and clinical applications. Experimental Section

Figure 1. The chemical structure of fluorescein.

it appears that TiO2 NPs can be used to detect catecholamine on the basis of the formation of a Ti-O bond. Catecholamines, such as DA, levodopa (L-DOPA), adrenaline, and noradrenaline, are an important class of neurotransmitters and are involved in a variety of central nervous system functions.22 High catecholamine levels are known to be cardiotoxic, leading to rapid heart rate, high blood pressure, and possible death of the heart muscles.23 On the contrary, a loss of DA-containing neurons may result in some serious diseases such as Parkinson’s disease.24 The detection of catecholamines is most often accomplished by the measurement of redox potential or intrinsic (native) fluorescence.25 For example, the monitoring of DA, adrenaline, and noradrenaline in urine has been achieved by capillary electrophoresis with laser-induced native fluorescence; thus, these compounds can be studied at the subnanomolar level.26 However, these methods are more complicated, expensive, and time-consuming. Recently, Au NPs have been generated by using catecholamines as active reducing agents, thereby enabling the quantitative colorimetric detection of the catecholamines.27 However, poor selectivity was observed since the catalytic growth of Au NPs can be performed by other reducing agents such as phenolic acids and hydrogen peroxide.27b Another sensitive approach for the detection of catecholamines makes use of fluorescence quenching of cadmium-selenium (CdSe) nanocrystals.28 Unfortunately, the same problem still occurs because the fluorescence of the CdSe nanocrystals can also be quenched by lactic acid and ascorbic acid. In contrast to these studies, we report a simple approach for the selective detection of DA, L-DOPA, and adrenaline by phosphate-modified TiO2 (P-TiO2) NPs in the presence of fluorescein (Figure 1). After the binding of DA, the P-TiO2 NPs become neutral and even positively charged. The adsorption of fluorescein on the particles results in the quenching of fluorescein by TiO2-DA complexes, which have strong absorption at 428 nm. Compared with the maximum absorption of fluorescein (489 nm), a 1.9-fold decrease in fluorescence intensity is observed by exciting at 428 nm. However, the fluorescence quenching of fluorescein does not occur if the excitation wavelength is set to 489 nm. By monitoring the decreases in fluorescence at 520 nm for fluorescein, we calculated the limits of detection (LODs) for DA, L-DOPA, and adrenaline at a signal-to-noise (S/N) ratio of 3, which were 33.5, 81.8, and 20.3 nM, respectively. The results imply that the proposed methods have great potential for use in (21) Rajh, T.; Chen, L. X.; Lukas, K.; Liu, T.; Thurnauer, M. C.; Tiede, D. M. J. Phys. Chem. B 2002, 106, 10543-10552. (22) Kopin, I. J. Pharmacol. ReV. 1985, 37, 333-364. (23) So´tonyi, P.; Merkely, B.; Hubay, M.; Ja´ray, J.; Zima, E.; Soo´s, P.; Kova´cs, A.; Szentma´riay, I. Toxicol. Sci. 2004, 77, 368-374. (24) Olanow, C. W. Neurology 1990, 40, 32-37. (25) (a) Peaston, R. T.; Weinkove, C. Ann. Clin. Biochem. 2004, 41, 17-38. (b) Nikolajsen, R. P. H.; Hansen, A. M. Anal. Chim. Acta 2001, 449, 1-15. (26) Park, Y. H.; Zhang, X.; Rubakhin, S. S.; Sweedler, J. V. Anal. Chem. 1999, 71, 4997-5002. (27) (a) Baron, R.; Zayats, M.; Willner, I. Anal. Chem. 2005, 77, 1566-1571. (b) Willner, I.; Baron, R.; Willner, B. AdV. Mater. 2006, 18, 1109-1120. (28) Ma, Y.; Yang, C.; Li, N.; Yang, X. Talanta 2005, 67, 979-983.

Chemicals. DA, L-DOPA, adrenaline, noradrenaline, catechol, ascorbic acid, salicylic acid, and fluorescein were obtained from Sigma (St. Louis, MO). Sodium hydroxide, NaH2PO4, and Na2HPO4 were purchased from Aldrich (Milwaukee, WI). Anatase P-TiO2 NPs (6 nm nominal diameter) were obtained from Go Yen Chemical Industry (Kaohsiung, Taiwan); the concentration of the original solution was 2.8 mM (1.73 × 1018 particles/mL). Buffer solutions were 10 mM NaH2PO4 (10 mM Na2HPO4 was used to adjust the pH range from 4 to 9). Without pretreatment, urine samples collected from a healthy male were measured using the solution comprising fluorescein and P-TiO2 NPs. Milli-Q ultrapure water was used in all of the experiments. Apparatus. A double-beam UV-visible spectrophotometer (Cintra 10e, GBC Scientific Equipment Pty, Ltd., Dandenong, Victoria, Australia) was used to measure the absorbance of the complex between P-TiO2 NPs and catecholamines. Their average size was determined by dynamic light scattering using an N5 Submicron Particle Size Analyzer (Beckman Coulter, Inc., Fullerton, CA). An Aminoco-Bowman fluorometer (ThermoSpectronic, Pittsford, NY) was used to measure the fluorescence emission spectra of fluorescein while the excitation wavelength was set at 428 nm. Determination of Catecholamines. A stock solution of fluorescein (1 mM) was prepared in methanol and diluted with phosphate buffer if necessary. Aliquots of fluorescein (2-100 µL) were added separately to the 0.28-1.96 mM solution of P-TiO2 NPs (prepared in 0.02 M phosphate buffer at different values of pH) such that the final volume of the mixture was 2 mL and the final concentration of fluorescein ranged from 1 to 50 µM. The catecholamines and related compounds were added separately to the as-prepared solution and equilibrated for 5 min. To investigate the selectivity of P-TiO2 NPs for catecholamines, a 0.1 mM solution of ascorbic acid, salicylic acid, and other analytes were also added separately to the as-prepared solution. After 5 min, both the fluorescence and UV-vis absorptions of the solution were recorded. High-Performance Liquid Chromatography (HPLC). Detection of catecholamines in urine samples was performed using an isocratic HPLC system (Chromsystems, Munich, Germany), including a pump (CLC-300), an autosampler (CLC-200), and an electrochemical detector (CLC-300). A CLC-200 autosampler injector with a 100 µL loop was used coupled with a CLC-300 electrochemical detector equipped with a gold electrode as the working electrode and Ag/AgCl as the reference electrode. The potential of the working electrode was set at +810 mV versus Ag/ AgCl. The same HPLC system was used for the Chromsystems kit with a 15 cm column (diameter and packing material not listed). Reagent kits for analysis of catecholamine in urine are also obtained from Chromsystems Chemicals (cat. no. 6000).

Results and Discussion Fluorescence Quenching of Fluorescein by TiO2-DA Complex. The surface charges of bare TiO2 NPs depend on their isoelectric point (pI ∼ 5.5).29 Above this pI, the surface is negatively charged because the surface OH groups undergo basic reactions. In contrast, below this pI, the surface is positively charged because of acidic reactions involving the surface OH groups. For the oxygen monocoordinated titanium, pK1 for > Ti-OH2+ f >Ti-OH + H+ is 2.60, and pK2 for > Ti-OH f TiO- + H+ is 9.00.30 Thus, it is our opinion that the bare TiO2 NPs aggregate when the pH is near the pI.31 To overcome this problem, phosphate ions, which are known to be adsorbed (29) Castellana, E. T.; Kataoka, S.; Albertorio, F.; Cremer, P. S. Anal. Chem. 2006, 78, 107-112. (30) Hsieh, Y.-L.; Chen, T.-H.; Liu, C.-P.; Liu, C.-Y. Electrophoresis 2005, 26, 4089-4097. (31) O’Shea, K. E.; Pernas, E.; Saiers, J. Langmuir 1999, 15, 2071-2076.

7882 Langmuir, Vol. 23, No. 14, 2007

Wu et al.

Figure 2. Effect of pH on the stability of 0.84 mM P-TiO2 NPs. The size distribution of P-TiO2 NPs was measured by dynamic light scattering. The concentration of phosphate buffer was 20 mM.

spontaneously on TiO2 NPs, were used to modify the surface charges of metal oxides.32 Dynamic light scattering spectra show no shift in the particles size distribution at pH 3.0-10.0 (Figure 2). We consider that P-TiO2 NPs are very stable over a wide range of pH values, mainly due to Coulombic repulsion between the negatively charged TiO2 NPs. It is also important to note that the P-TiO2 NPs are reasonably stable when high concentrations of a phosphate buffer are used. To further explore the selective reaction of the P-TiO2 NPs, a series of catecholamines and related compounds was tested by using 0.84 mM P-TiO2 NPs. Except for noradrenaline and ascorbic acid, the color of the surfaces of the P-TiO2 NPs changed due to the attachment of electron-donating enediol ligands on the surface (Supporting Information, Figure S1). In contrast to the bare TiO2 NPs, the selective reactivity toward the enediol ligands was improved by using the P-TiO2 NPs. It is worth mentioning that no coordination was observed between ascorbic acid and the P-TiO2 NPs; however, this was not true in the case of the bare TiO2 NPs.21 It appears that the P-TiO2 NPs could be used to distinguish between DA and ascorbic acid in real biological matrixes.33 Similar phenomena were observed for salicylic acid and noradrenaline. It is considered that the phosphate ions play an important role in determining the selectivity of the TiO2 NPs. A series of catechol derivatives, including DA, L-DOPA, and adrenaline, can displace the phosphate ions absorbed on TiO2 by means of the chelation of surface Ti(IV) ions with the enediol groups. The results were further confirmed by measuring the particle size distribution of the P-TiO2 NPs. The size of the P-TiO2 NPs increased from 8.2 ( 0.9 nm in the absence of DA to 9.4 ( 1.4 nm in the presence of DA (Supporting Information, Figure S2). Next, Figure 3A shows the absorption spectra of the P-TiO2 NPs in the absence and presence of DA and ascorbic acid. In comparison with ascorbic acid, a new strong absorption band for DA was found at 428 nm, indicating the formation of TiO2-DA complexes. The extinction coefficient of these complexes is ∼3500 cm-1 mol-1 L in the visible regions. It is considered that the strong absorption bands in the visible region arise from ligandto-metal charge-transfer interaction between the ligand and surface metal atoms. This phenomenon causes the excitation of electrons from the chelating ligand directly into the conduction band of the TiO2 NPs.18 Additionally, prior to the binding of DA, the (32) (a) Abdullah, M.; Low, G. K.-C.; Matthews, R. W. J. Phys. Chem. 1990, 94, 6820-6825. (b) Nelson, B. P.; Candal, R.; Corn, R. M.; Anderson, M. A. Langmuir 2000, 16, 6094-6101. (33) Zhang, P.; Wu, F.-H.; Zhao, G.-C.; Wei, X.-W. Bioelectrochemistry 2005, 67, 109-114.

Figure 3. (A) UV-vis absorption spectra of solutions of 0.84 mM P-TiO2 NPs containing (a) no analyte, (b) 100 µM DA, and (c) 100 µM ascorbic acid. (B) Fluorescence spectra of 10 µM fluorescein solutions containing (a) 0.84 mM P-TiO2 NPs, (b) 0.84 mM P-TiO2 NPs and 100 µM DA, and (c) 0.84 mM P-TiO2 NPs and 100 µM ascorbic acid. The samples were incubated for 5 min at room temperature. The excitation wavelength was set to 428 nm.

P-TiO2 NPs are negatively charged, and these repel the negatively charged fluorescein (pKa 6.5) at high pH range. Upon the binding of DA, the P-TiO2 NPs become neutral and even positively charged. As a result, the adsorption of fluorescein on the particles results in the quenching of fluorescein by P-TiO2 NPs. The local high concentration of fluorescein on the particles also occurs when the P-TiO2 NPs interact with DA. This phenomenon will cause a fluorescence self-quenching of fluorescein. To test our hypothesis, fluorescein (10 µM) with a high quantum yield was added to a solution of 0.84 mM P-TiO2 NPs. In comparison with spectrum a (the emission spectrum of fluorescein) shown in Figure 3B, spectrum b clearly indicates that the fluorescence intensity of fluorescein at 520 nm decreases drastically in the presence of 0.1 mM DA. Once the DA molecules are adsorbed onto the surface of the P-TiO2 NPs by means of Ti-O bonding, fluorescence quenching occurs between fluorescein and the TiO2DA complexes. However, a decrease in the fluorescence intensity of fluorescein is not observed when the excitation wavelength is set to 489 nm (the maximum absorption of fluorescein). We suggested that fluorescence quenching is not due to the selfquenching of fluorescein. It is not surprising that a similar spectrum c, as shown in Figure 3B, was obtained after adding ascorbic acid to the solution of 0.84 mM P-TiO2 NPs. We note that the bands appearing near 428 nm arise due to Rayleigh scattering. The depiction in Figure 4 demonstrates the mechanisms of the assays presented in this report. Effects of Concentration and pH. To optimize the sensitivity of detection, we explored the factors that influence the fluorescence quenching, including the concentration of the

Using P-TiO2 NPs to Detect Catecholamines

Langmuir, Vol. 23, No. 14, 2007 7883

Figure 4. Schematic depiction of the mechanisms of the assays using a solution of P-TiO2 NPs and fluorescein. The fluorescence quenching of fluorescein occurs in the presence of DA.

Figure 5. Effect of (A) concentration of P-TiO2 NPs and (B) solution pH on the fluorescence quenching efficiency. F0 and F denote the fluorescence intensity of fluorescein (10 µM) in the absence and presence of 100 µM DA, respectively. The error bars represent standard deviations based on three independent measurements. The excitation wavelength was set to 428 nm.

P-TiO2 NPs and the pH of the solution. It is essential to control the concentration of the P-TiO2 NPs used to equilibrate DA. Figure 5A shows the effects of the concentrations of the P-TiO2 NPs (0.28-1.96 mM) on the fluorescence quenching efficiency. The fluorescence differences between the absence and presence of 0.1 mM DA were maximized when 1.4 mM P-TiO2 NPs were used. It is considered that free DA molecules were found in the bulk solution at concentrations below 1.4 mM. However,

the fluorescence differences tended to decrease when an excess of P-TiO2 NPs (>1.4 mM) was used. This finding indicates that the amount of DA adsorbed on a single particle decreased with an increase in the concentration of the added P-TiO2 NPs. The results also substantiate the concept that only those molecules adsorbed on the particle surface can be used to quench the fluorescence of fluorescein. On the basis of these results, we chose to use 1.4 mM P-TiO2 NPs for this study. Additionally, the pH of the solution might play a role in determining the sensitivity of our analysis because it affects the quantum yield of fluorescein and the displacement capacity of DA. Figure 5B shows that the fluorescence difference between the absence and presence of 0.1 mM DA is sensitive to pH. It is well-known that the quantum yield of fluorescein is significantly enhanced by increasing the pH of the solution.34 It is also considered that the efficiency of the displacement of a phosphate ion from the P-TiO2 surface by DA increased by increasing the pH of the phosphate solution from 4.0 to 9.0. We note that the degree of Coulombic repulsion between the phosphate ion and the TiO2 NPs increased with the pH, resulting in a large decrease in the amount of phosphate ions absorbed on the surface of the P-TiO2 NPs.35 Thus, the displacement capability of DA was enhanced at higher pH. We believe that it is preferable to conduct such experiments at pH 8.0. Quantitative Analyses of Catecholamines and Catechol. The results presented above indicate that it is possible to determine the concentrations of DA, L-DOPA, adrenaline, and catechol by using a solution of 1.4 mM P-TiO2 NPs and 10 µM fluorescein in a 20 mM phosphate buffer at pH 8.0. The fluorescence quenching of fluorescein at different concentrations of DA is shown in Figure 6A. As expected, the fluorescence intensity decreased with an increase in the concentration of DA. A calibration curve was constructed by a linear regression of the fluorescence differences (λex ) 428 nm, λem ) 520 nm) against the DA concentrations. The plot presented in Figure 6B exhibits good linearity (R2 ) 0.9987) over the range of 0.5-100 µM. The LOD for DA at an S/N ratio of 3 is 33.5 nM. Moreover, Table 1 shows that, when the other compounds, including L-DOPA, adrenaline, and catechol, are detected by a solution containing 1.4 mM P-TiO2 NPs and 10 µM fluorescein, the fluorescence difference between the absence and presence of analytes changes linearly with their concentrations. Therefore, the plot of each catecholamine and catechol exhibits good linearity (R2 > 0.98). (34) Kurian, A.; George, S. D.; Bindhu, C. V.; Nampoori, V. P.; Vallabhan, C. P. Spectrochim. Acta A: Mol. Biomol. Spectrosc., in press. (35) Connor, P. A.; McQuillan, A. J. Langmuir 1999, 15, 2916-2921.

7884 Langmuir, Vol. 23, No. 14, 2007

Wu et al.

Table 1. Comparison of the Dynamic Range and LOD for Catechol and Catecholamines Using 1.4 mM P-TiO2 NPs by Two Different Detection Modesa fluorescenceb R2

sample

dynamic range

DA L-DOPA epinephrine catechol

5× -1 × 10 5 × 10-7-1 × 10-4 1 × 10-6-5 × 10-5 1 × 10-6-5 × 10-4 10-7

-4

absorption LOD (nM)

dynamic range

33.5 81.8 20.3 92.1

5× -1 × 10 5 × 10-5-1 × 10-3 NDc 5 × 10-5-5 × 10-4

0.9987 0.9802 0.9944 0.9948

10-5

-3

R2

LOD (nM)

0.9998 0.9998 NDc 0.9999

440 480 NDc 440

a Conditions as presented in Figure 6. b The excitation wavelength was set to 428 nm. c Not determined because a high concentration of epinephrine (>100 µM) cannot dissolve in water.

Figure 6. (A) Effect of DA on the fluorescence spectra of solutions containing 10 µM fluorescein and 1.4 mM P-TiO2 NPs in 20 mM phosphate buffer (pH 8.0). The concentration of DA was increased, from top to bottom, 0, 5, 10, 50, 100, and 500 µM. (B) Calibration curve for the quantitative analysis of DA. The error bars represent standard deviations based on three independent measurements. The other conditions are the same as those in Figure 5.

Figure 7. (A) Detection of DA in urine samples based on fluorescence quenching. The urine samples were spiked (a) without and (b) with 25 µM of DA. The other conditions are the same as those in Figure 6. (B) Calibration curve for the detection of DA in human urine. The error bars represent standard deviations based on three independent measurements.

The LODs for L-DOPA, adrenaline, and catechol at an S/N ratio of 3 are 81.8, 20.3, and 92.1 nM, respectively. In comparison with the measurement of DA by absorption at 428 nm (Supporting Information, Figure S3), fluorescence quenching provided an approximately 10-fold improvement in the LODs (Table 1). It is interesting to note that the LOD for the L-DOPA is relatively higher than that for DA. The lower sensitivity may be attributed to the carboxylic group of L-DOPA, thereby resulting in a relatively large repulsion between the L-DOPA molecules and the P-TiO2 NPs. Detection of DA in Urine Samples. We tested the application of our proposed method by using it for the practical analyses of DA in urine samples. The mean concentrations of DA, adrenaline, and noradrenaline in the urine samples are 333 ( 144, 6.5 ( 1.1, and 34.3 ( 5.2 µg/L, respectively.36 Although the concentration of noradrenaline is close to that of DA, the results shown above indicate that the P-TiO2 NPs do not react with noradrenaline.

In addition, the concentration of DA is 100 times higher than that of adrenaline. Thus, we expect that the detection of DA in urine samples can be achieved by our proposed method. An apparent fluorescence quenching of fluorescein was obtained by the formation of TiO2-DA complexes after 10 µM DA was spiked into the urine samples (Figure 7A). By using a standard addition method, we estimated that the concentration of DA in the urine samples was 392.1 ( 22.9 µg/L (2.56 ( 0.15 µM, n ) 3), which is in good agreement with the normal value of between 75.3 and 452 µg/L (Figure 7B).37 The result was further confirmed by HPLC used in combination with an electrochemical cell. Consequently, we determined the concentration of DA in the urine samples to be 202.8 ( 15.3 µg/L (1.32 ( 0.10 µM, n ) 3) by using a standard addition method (Supporting Information, Figure S4). Although other catecholamines such as adrenaline also react with the P-TiO2 NPs, their interference is not significant because of their considerably lower concentrations in the urine samples.

(36) Anderson, G. M.; Young, J. G.; Jatlow, P. I.; Cohen, D. J. Clin. Chem. 1981, 27, 2060-2063.

(37) Panholzer, T. J.; Beyer, J.; Lichtwald, K. Clin. Chem. 1999, 45, 262268.

Using P-TiO2 NPs to Detect Catecholamines

Conclusions We describe simple assays based on fluorescence quenching for the selective detection of catechol and its derivatives by using a solution of TiO2 NPs and fluorescein. The fluorescence of fluorescein at 520 nm is suppressed because of fluorescence quenching between the complexes (P-TiO2 NPs and enediol ligands) and fluorescein. In comparison with bare TiO2 NPs, the P-TiO2 NPs provide high stability over a wide range of pH and high selectivity toward DA, L-DOPA, adrenaline, and catechol. To evaluate the factors that affect the sensitivity of detection, the concentration of the P-TiO2 NPs and the pH of the solution have been optimized. In addition, the concentration of DA (392.1 ( 22.9 µg/L) in the urine samples is estimated by our proposed method. To confirm the results, we have also analyzed DA in the urine samples by applying HPLC in combination with an electrochemical cell; as a result, the average DA concentration is determined to be 202.8 ( 15.3 µg/L. Our future goals include

Langmuir, Vol. 23, No. 14, 2007 7885

using P-TiO2 NPs for sensing catecholamines in biological samples that are separated by capillary electrophoresis and are determined by surface-assisted laser desorption/ionization mass spectrometry. Acknowledgment. We would like to thank the National Science Council (NSC 95-2113-M-110-020-) and the National Sun Yat-sen University-Kaohsiung Medical University Joint Research Center for the financial support of this study. Supporting Information Available: Depiction of the selective detection of DA, L-DOPA, catechol, and adrenaline; DLS size distributions of P-TiO2 NPs; UV-vis absorption spectra illustrating the effect of DA concentrations on P-TiO2 NPs; and the separation of catecholamines in urine samples using HPLC with electrochemical detection This material is available free of charge via the Internet at http://pubs.acs.org. LA700555Y