Ultrasensitive and Highly Selective Detection of Bioaccumulation of

Methylmercury (CH3Hg+), the common organic source of mercury, is well-known as one of the most toxic compounds that is more toxic than inorganic or ...
0 downloads 0 Views 2MB Size
Download PDF
Article pubs.acs.org/ac

Ultrasensitive and Highly Selective Detection of Bioaccumulation of Methyl-Mercury in Fish Samples via Ag0/Hg0 Amalgamation Li Deng,† Yan Li,‡,§ Xiuping Yan,§ Jun Xiao,⊥ Cheng Ma,† Jing Zheng,† Shaojun Liu,⊥ and Ronghua Yang*,† †

State Key Laboratory of Chemo/Biosensing and Chemometrics, College of Chemistry and Chemical Engineering, Collaborative Innovation Center for Chemistry and Molecular Medicine, Hunan University, Changsha, 410082, China ‡ College of Chemistry, Tianjin Normal University, Tianjin, 300387, China § State Key Laboratory of Medicinal Chemical Biology, College of Chemistry, Nankai University, Tianjin, 300071, China ⊥ Key Laboratory of Protein Chemistry and Fish Developmental Biology of the Ministry of Education of China, College of Life Sciences, Hunan Normal University, Changsha, 410081, China S Supporting Information *

ABSTRACT: Methylmercury (CH3Hg+), the common organic source of mercury, is well-known as one of the most toxic compounds that is more toxic than inorganic or elemental mercury. In seabeds, the deposited Hg2+ ions are converted into CH3Hg+ by bacteria, where they are subsequently consumed and bioaccumulated in the tissue of fish, and finally, to enter the human diet, causing severe health problems. Therefore, sensitive and selective detection of bioaccumulation of CH3Hg+ in fish samples is desirable. However, selective assay of CH3Hg+ in the mercury-containing samples has been seriously hampered by the difficulty to distinguish CH3Hg+ from ionic mercury. We report here that metal amalgamation, a natural phenomenon occurring between mercury atoms and certain metal atoms, combining with DNA-protected silver nanoparticles, can be used to detect CH3Hg+ with high sensitivity and superior selectivity over Hg2+ and other heavy metals. In our proposed approach, discrimination between CH3Hg+ and Hg2+ ions was realized by forming Ag/Hg amalgam with a CH3Hg+-specific scaffold. We have found that Ag/Hg amalgam can be formed on a CH3Hg+-specific DNA template between silver atoms and mercury atoms but cannot between silver atoms and CH3Hg+. With a dye-labeled DNA strand, the sensor can detect CH3Hg+ down to the picomolar level, which is >125-fold sensitive over Hg2+. Moreover, the presence of 50-fold Hg2+ and 106-fold other metal ions do not interfere with the CH3Hg+ detection. The results shown herein have important implications for the fast, easy, and selective detection and monitoring of CH3Hg+ in environmental and biological samples.

M

health problems, such as sensory, motor, and neurological damage,4,9 as seen in such unfortunate incidents as Minamata disease10 and CH3Hg+ poisoning in Iraq.11 Therefore, sensitive, reliable, and cost-effective detection of CH3Hg+ concentrations in biota and fish is important to avoid ecotoxicological risks and understand the biogeochemical cycling of mercury species in the environment.12−15 Because the natural concentration of mercury presented in aquatic systems is very low,16 current methods for detection and quantification of mercury species rely mainly on instrumental analytical techniques, such as atomic absorption spectrometry, inductively coupled plasma atomic emission spectrometry, and inductively coupled plasma mass spectrom-

ercury is a widespread pollutant with distinct toxicological profiles.1,2 Methylmercury (CH3Hg+), the most common organic source of mercury, is well-known as one of the most toxic compounds,3,4 which is more toxic than inorganic or elemental mercury. Mercury that is emitted through natural environments and human activities experiences long-range atmospheric transport and will be oxidized into Hg2+. In aqueous solution, the deposited Hg2+ ions are converted into CH3Hg+ by bacteria,5 which is consumed and bioaccumulated through the food chain,6 for example, in the tissue of fish, in which the CH3Hg+ concentrations are frequently found to exceed the maximum levels recommended by the U.S. Environmental Protection Agency (EPA) and the World Health Organization (WHO) for human consumption (Figure 1).7,8 The biomagnified CH3Hg+ finally enters the human diet by fishery of the fish or seafood and is ingested by humans via food consumption to cause several serious human © 2015 American Chemical Society

Received: November 27, 2014 Accepted: January 22, 2015 Published: January 22, 2015 2452

DOI: 10.1021/ac504538v Anal. Chem. 2015, 87, 2452−2458

Article

Analytical Chemistry

probe, the sensing approach allows to detecting CH3Hg+ down to the picomolar level through fluorescence enhancement, 50fold Hg2+ and 106-fold other metal ions do not interfere the detection. The capabilities of this method were tested by monitoring CH3Hg+ accumulation in tissues and organs from three kinds of daily consumed fishes. To the best of our knowledge, no such a CH3Hg+ detection method has been reported in the literature.

■

EXPERIMENTAL SECTION Chemicals and Apparatus. AgNO3 (99.99%), Hg(NO3)2 (99.999%), and NaBH4 (99.99%) were purchased from SigmaAldrich and used as received. CH3HgCl (>95%) was received from Aladdin-Reagent (Shanghai, China). The dye-labeled oligonucleotides were synthesized by Shanghai Sangon Biological Technology & Services Co., Ltd., China. All sequences were dissolved in highly pure water (sterile Millipore water, 18.3 MΩ) as stock solutions. All metal ion stock solutions were prepared from nitrate salts. All other reagents were of analytical reagent grade and were used without further purification or treatment. All working solutions were prepared with 0.1 mM Tris-HNO3 buffer solution (pH 7.2). UV−vis absorption spectra were obtained on a Hitachi U4100 UV−vis spectrophotometer (Kyoto, Japan). Fluorescence measurements were recorded on a Hitachi F-7000 fluorescence spectrofluorometer. Fluorescence anisotropy was measured on a modular spectrofluorometer (Photon Technology International Inc.) Raman spectra were gained by a confocal microprobe Raman instrument (RamLab-010, Horiba Jobin Yvon, France). The CE−ICPMS hybrid technique was conducted through a laboratory-built CE system combined with a Thermo Electric X7 ICPMS. Transmission electron microscopy (TEM) was performed on a JEOL JEM-3010. Energy-dispersive X-ray (EDX) spectra were obtained with the attachment of a TEM microscope. The pH was measured by a model 868 pH meter (Orion). Interactions of Cation with DNA using CE−ICPMS. A previously developed CE−ICPMS hybrid technique45 was used to study the interactions between metal ions and DNA. Detailed information on the CE−ICPMS could be obtained elsewhere.45 Briefly, the CE−ICPMS hybrid technique is composed of three parts: CE, a laboratory-built thermospray interface and ICPMS. Bare fused-silica capillaries were used for the separation of Ag+, Hg2+, and CH3Hg+ from their respective DNA adducts. The electrolytes were running at 20 kV with various Tris-H3BO3 buffers. The effluent from CE was online detected by a Thermo Electric X7 ICPMS. Fluorescent Detection of CH3Hg+. The stock solution of CH3Hg+ was prepared by first dissolving suitable amounts of CH3HgCl in dimethylformamide (DMF), diluting with highly pure water, and then stocked in 30% DMF at 4 °C. For detection of CH3Hg+ in buffer solutions, 5 μL of the asprepared DNA/Ag+ mixture were added to a 500 μL volumetric pipe containing 465 μL of Tris-HNO3 buffer sample. Then 10 μL of CH3HgCl solution of various concentrations were added. After incubating for 10 min with gentle shaking, 20 μL of NaBH4 (1 mM) were added to the mixture. The samples were incubated for 10 min at room temperature before fluorescence measurement. Sample Preparation and Extraction Procedure for CH3Hg+ Determination. For measuring CH3Hg+ content in fresh tissue/organ, the certified reference material ERM-CE464 (tuna fish, BCR) was purchased from LGC Science (Beijing)

Figure 1. Schematic illustration of CH3Hg+ generating and entering human diet via the food chain.

etry.17−19 These methods, although very sensitive, can detect only total mercury contents but cannot discriminate between CH3Hg+ and inorganic mercury ions. Since the health effects between CH3Hg+ and mercury ions are dramatically different,3 the determination of total mercury is no longer completely acceptable. As a consequence, the above detection methods are often performed after sample treating and separation steps using methods such as gas chromatography (GC) and highperformance liquid chromatography (HPLC) to separate and identify the mercury species,20−22 which are rather complicated, time-consuming, and costly. To overcome this limitation and to allow simple and rapid screening applications, mercury-specific sensors have been developed by employing organic chromophores or fluorophores,23−25 polymeric materials,26,27 metal nanoparticles,28−30 and nucleic acids,31−33 or proteins.34 However, to the best of our knowledge, most studies have been devoted to the detection of Hg2+ ions, while the detection of the CH3Hg+ is seldom reported,35 probably because of the weak complex-forming ability of CH3Hg+ in comparison with that of Hg2+. Regarding the development of fluorescent sensor for CH3Hg+, mercury-specific latent fluorophore and chemodosimeters,36−38 protein-functionalized gold nanoclusters,39 organically functionalized mesoporous inorganic materials,40 and rare-earth up-conversion nanoparticles 41 have been developed. Chemically, the assays are realized via CH3Hg+induced desulfurization or alkyl ether cleavage reactions of the acceptor molecules. However, because CH3Hg+ is less thiophilic than Hg2+, the above methods exhibit limited sensitivity and quite weak response toward CH3Hg+ over Hg2+ ions. Consequently, the development of a new approach that can alleviate these problems and, more importantly, allow to the direct assay of the CH3Hg+ in biota and fish species without interference from Hg2+ ions is highly desirable. Moreover recently, Stellacci et al. reported a class of solidstate electrochemical sensors for toxic cations by employing gold nanoparticles (AuNPs) that were coated with striped nhexanethiol and alkanethiols terminated with ethylene glycol (EG).42 Interestingly, depending on the carbon chain length of EG, the sensors can detect CH3Hg+ with ultrahigh sensitivity and selectivity, although the cation recognition and corresponding electrochemical signal transduction are governed by metalligation interaction of thiol-containing ligands with the cations. We show here that silver/mercury amalgamation, a natural phenomenon occurring between mercury atoms and silver atoms,43,44 combining with a CH3Hg+-specific DNA probe, can provide a simple and alternative approach to selectively detect CH3Hg+. By using a fluorophore-labeled CH3Hg+-specific DNA 2453

DOI: 10.1021/ac504538v Anal. Chem. 2015, 87, 2452−2458

Article

Analytical Chemistry Ltd., the fresh fish species were obtained from a local market of Changsha, China. Crucian carp was obtained from the College of Life Science, Hunan Normal University. Fresh tissue/organ samples were acquired by dissection of the fish. All the fresh samples were homogenized by a blender before further treatment. A classical alkaline digestion protocol was employed to release CH3Hg+ in biological samples,46 followed by filtration to remove the residue.

be formed on the DNA strand due to the strong ligation of thymine base with Hg2+ ions,48 quenching the fluorescence signal in a manner similar to the suppression of AgNPs formation, as noted above. To test the hypothesis, a 25-mer T-rich DNA (HT7, Table S1, see the Supporting Information) was designed according to the literature.32 Binding affinities of HT7 with Ag+, Hg2+, CH3Hg+, and Zn2+ (as a typical interfering agent) were studied by capillary electrophoresis with online inductively coupled plasma mass spectrometry (CE−ICPMS)45 and compared with those of a random DNA (HR, Table S1, see the Supporting Information). As shown in Figure 3a, in each electropherogram, free metal species migrate much faster, corresponding to the first peak. Formations of the metal-DNA adducts cause a weaker backward electrophoretic mobility by the reduction of the effective charge and thus belongs to the second peak.49 Interestingly, different electrophoretic behaviors of the DNAmetal adducts were observed for the HT7 or HR strands. On the basis of the proportion of second peaks, the results suggest that the binding affinity of HT7 with the metal ions follows the trend of CH3Hg+ > Hg2+ > Ag+ ≫ Zn2+, indicating that it is only the CH3Hg+ that can tightly bind HT7 to prevent the DNAtemplated AgNPs formation. However, when it comes for HR, the binding affinity of CH3Hg+ > Ag+ ≫ Hg2+ or Zn2+ can be detected (Figure S1, see the Supporting Information). These results are consistent with the previously reported ones.49−51 The metal-DNA binding constant (Kb) could be determined by using a linear regression of a Scatchard plot based on the equation: CbM/CfM = −KbCbM + CDNAKb,49,52 where CbM and CfM are the concentrations of bound and free metal ions, respectively, and CDNA is the concentration of the binding site on the DNA (Figure S2, see the Supporting Information). Figure 3b shows the Scatchard plot for each metal ion−DNA interaction, where the slope of each plot refers to the corresponding Kb. The data of binding constants are collected in Table S2, see the Supporting Information. As expected, HT7 binds to CH3Hg+ with the largest Kb value of (5.57 ± 0.47) × 106 M−1 over Hg2+ ((1.51 ± 0.18) × 106 M−1) or Ag+ ((0.83 ± 0.11) × 106 M−1). Thus, in the absence of CH3Hg+, the relatively high affinity of Hg2+ or Ag+ to T-rich DNA enables the AgNPs and Ag/Hg amalgam to be formed on the DNA scaffold. However, when CH3Hg+ is present, the superstrong binding of CH3Hg+ to DNA would suppress formation of Ag+DNA adducts, thus, in turn, the formation of AgNPs on the DNA scaffold, as shown in Figure 2.

■

RESULTS AND DISCUSSION The basic design of our approach is originated from the higher affinity of nucleotide for CH3Hg+ over Hg2+ and the formation of Ag/Hg amalgams between Ag0 and Hg0 rather than CH3Hg+. In this design, a thymine (T)-rich DNA was employed as the key point for CH3Hg+ discrimination from Hg2+. As shown in Figure 2, first, in the absence of CH3Hg+ or Hg2+, silver

Figure 2. Schematic illustration for fluorescent detection of CH3Hg+ based upon a dye-labeled T-rich DNA (FAM-DNA). Formation of AgNPs or Ag/Hg amalgams on FAM-DNA template quenches the FAM fluorescence emission; the presence of CH3Hg+ suppresses formation of the metal nanostructures, revealing fluorescence enhancement.

nanoparticles (AgNPs) are grown on a dye-labeled T-rich ssDNA(FAM-DNA) by reduction of associated Ag+ to quench the fluorescent tag of the ssDNA.47 However, in the presence of CH3Hg+, a stronger binding of CH3Hg+ over Ag+ ions for the ssDNA suppresses the formation of AgNPs on the FAM-DNA and, hence, lighting up the FAM fluorescence. On the other hand, when Hg2+ ions are present, the Ag/Hg amalgam would

Figure 3. (a) Electropherograms obtained by CE−ICPMS assay for the interactions of Ag+, Hg2+, Zn2+, and CH3Hg+ with HT7 after 12-h incubation at 37 °C. The concentrations of Ag+, Hg2+, Zn2+, and CH3Hg+ were 5.0, 5.0, 5.0, and 2.5 μM, respectively. (b) The corresponding Scatchard plots of Ag+, Hg2+, Zn2+, and CH3Hg+ binding to HT7 from a series of CE−ICPMS assays after a 12-h incubation of 5.0 μM HT7 with various concentrations of metal ions at 37 °C. 2454

DOI: 10.1021/ac504538v Anal. Chem. 2015, 87, 2452−2458

Article

Analytical Chemistry

Figure 4. Fluorescence emission changes of FAM-HR/Ag+ (a) and FAM-HT7/Ag+ (b) toward CH3Hg+, Hg2+, and Zn2+ ions by forming DNAtemplated silver nanostructures. For each trace, we distinguished three steps: (1) the cuvette was filled with 500 μL of Tris-HNO3 buffer containing 50 nM of the FAM-HR or FAM-HT and 10.0 μM AgNO3; (2) 20.0 μM CH3HgCl, Hg(NO3)2, or Zn(NO3)2 was introduced in the cuvette; and (3) 40.0 μM NaBH4 was added to the cuvette. The transition between each regime is marked with an arrow. Excitation was at 480 nm, and emission was monitored at 518 nm.

Figure 5. (a) UV−vis absorption spectra of FAM-HT7/Ag+ induced by NaBH4. The spectra were collected in the Tris-HNO3 buffer solution containing (1) FAM-HT7 + Ag+; (2) (1) + NaBH4; (3) (1) + CH3Hg++ NaBH4, and (4) (1) + Hg2+ + NaBH4. [FAM-HT7] = 1.0 μM, [AgNO3] = [CH3HgCl] = [Hg(NO3)2] = 10.0 μM, [NaBH4] = 20 μM. Inset: the corresponding color of the solutions of 3 and 4. (b) Gel electrophoresis of reaction products. The composition of each lane is as noted. [FAM-HT7] = 1.0 μM, [AgNPs] = 2.0 μM, [AgNO3] = 2.0 μM, [CH3HgCl] = [Hg(NO3)2] = 1.0 μM, [NaBH4] = 5.0 μM.

Having found the CH3Hg+-specific DNA, we exploit HT7 to construct a new probe for discrimination between CH3Hg+ and Hg2+. To signal fluorescently this CH3Hg+ binding event, 6carboxyfluorescein (FAM) was labeled on the 5′-end of HT7 (FAM-HT7, Table S1, see the Supporting Information). Preliminary experiments were performed to prepare the DNA-self-assembled AgNPs by the NaBH4-reduced method.53 Free FAM-HT7 exhibits strong fluorescence; however, when adding Ag+ subsequently with NaBH4, up to 98% fluorescence was quenched (Figure S3, see Supporting Information), indicating the formation of DNA-templated AgNPs.47 The ultrahigh quenching ability of DNA-templated AgNPs was further confirmed by comparing the quenching effect of colloidal AgNPs as well as other metal nanostructures (Figures S4−S6, see the Supporting Information). The ability of the sensor in distinguishing CH3Hg+ from Hg2+ was tested by recording real-time fluorescence of the Ag+ complexes with different FAM-ssDNAs (Table S1, see the Supporting Information) in the presence of CH3Hg+ or Hg2+ and subsequently NaBH4. As shown in Figure 4, the fluorescence emission of the FAM-ssDNA/Ag+ is almost completely quenched after NaBH4-addition. However, if CH3Hg+ or Hg2+ is present in the solution, the FAM fluorescence quenching is suppressed and, as a result, emission

enhancement is observed in comparison to that without the analyte. It was, moreover, intriguing to find that the magnitudes of fluorescence enhancement induced by Hg2+ and CH3Hg+ depends on the DNA sequences. Figure 4a shows the presence of 10 μM Hg2+ or CH3Hg+ in the FAM-HR solution resulted in similar fluorescence enhancement (∼73% and 88%, respectively). However, since the FAM-HT7 has a higher affinity for CH3Hg+ than Hg2+, around 82% of the fluorescence is restored by 10 μM CH3Hg+, but only ∼18% is retained with the same Hg2+ concentration (Figure 4b), typical interfering agent as Zn2+ shows the similar signal as the blank solution for both FAM-HT and FAM-HR. This comparison clearly demonstrates that the change of DNA sequences improves the fluorescence response toward CH 3 Hg + and, consequently, CH 3 Hg + selectivity. Optimal sequence were chosen by investigating DNA strands containing different numbers of T-rich sequences, FAM-HT5 and FAM-HT9, Table S1, see the Supporting Information, none gave better results than FAM-HT7 (Figure S7a,b, see the Supporting Information). The responses of the four DNA sequences to each metal ion are shown in Figure S7c, see the Supporting Information, as for FAM-HR, the signal of Hg2+ is close to CH3Hg+, especially at the lower concentration. However, all the FAM-HT DNAs show good 2455

DOI: 10.1021/ac504538v Anal. Chem. 2015, 87, 2452−2458

Article

Analytical Chemistry

Figure 6. Performance of CH3Hg+ detection. (a) Fluorescence emission spectra of FAM-HT7/Ag+ in the presence of different amounts of CH3Hg+ and subsequently excess of NaBH4. The arrow indicates the signal changes as increases in the CH3Hg+ concentrations. (b) Plots of the value of response sensitivity, S/B, as functions of the concentrations of Hg2+ and CH3Hg+. (c) The response sensitivity, S/B, of the sensing system for selected metal ions (10.0 μM, x-axis markers) and CH3Hg+ (10.0 μM) + the metal ions. The response toward CH3Hg+ without interfering agent ions is used as the standard (S/B = 100). (d) Time-dependence of CH3Hg+ distribution in various tissue samples of crucian carp. The blank bars refer to the background measured each time. The magnitude of the error bars was calculated from the uncertainty given by three independent measurements.

ability in discriminating CH3Hg+ with Hg2+ in a wide range of concentrations, of which FAM-HT7 reveals the optimal results. To further confirm the sensing mechanism, spectral features and structure compositions of the NaBH4-reduction product of FAM-HT7/Ag+ were studied with different conditions. In Figure 5a, the UV−vis absorption spectrum of FAM-HT7/Ag+ shows two absorption bands centered around 260 and 480 nm, corresponding to the DNA and FAM. Reduction of FAM-HT7/ Ag+ results in a new band centered at ∼400 nm, attributing to the surface plasmon resonance of small AgNPs.54 TEM imaging and EDX spectroscopy show uniform and monodispersed nanoparticles (Figure S9a, see the Supporting Information) with composition of Ag, O, N, and P elements (Figure S9d and Table S3, see the Supporting Information), indicating the formation of AgNPs on the DNA strand. However, the reduction product of FAM-HT7/Ag+/Hg2+ reveals a band-shift from 400 to 320 nm, characteristic of dimetallic nanoparticles,55 with irregular size and composition of Ag, Hg, O, N, and P (Figure S9b,e and Table S4, see the Supporting Information), demonstrating that Ag/Hg amalgams are formed on FAM-HT7. In contrast, the absorption spectrum of the reduction product of DNA/Ag+/CH3Hg+ is similar to that observed from Ag+ alone (Figure 5a), but the EDX spectrum shows that only one main component of Ag existed in the nanoparticles (Figure S9f and Table S5, see the Supporting Information), preliminarily indicating that in the presence of CH3Hg+ the AgNPs are formed out of the DNA strand. The results of agarose gel electrophoresis provide more proof again for illustrating our mechanism. As shown in Figure 5b, the sharp, fast-moving band with strong fluorescence are shared

by lane 1 and lane 2, which means that AgNPs separated from the DNA do not influence the migration of the DNA or cause efficient intermolecular fluorescence quenching. However, the products in lane 3 and lane 4 move significantly slower with weak fluorescence, indicating that AgNPs or Ag/Hg amalgams are formed within DNA strand, decreasing the DNA migration and quenching the fluorescence intramolecularly. For lane 5, the same band as lane 1, 2 indicates the successfully suppression of DNA-AgNPs formation in the presence of CH3Hg+. Additional proofs including the detection of SERS signal (Figure S10, see Supporting Information), as well as fluorescence anisotropy study (Figure S11, see Supporting Information) also agree with our assumed mechanism. Taken together, these corroborated results demonstrate the fact that by selecting properly designed DNA scaffold, the presence of CH3Hg+ would suppress formation of metal nanostructures on the DNA that quench the fluorescent tag, allowing selective detection of CH3Hg+ over Hg2+. On the basis of the promising proof-of-mechanism, we proceeded to evaluate the sensitivity and selectivity of our system for detecting CH3Hg+. Taking into account the response sensitivity, pH 8.0 was chosen as the optimal condition for CH3Hg+ detection (Figure S12a−d, see the Supporting Information). Figure 6a shows typical fluorimetric titration curves of FAM-HT7/Ag+ in the presence of different concentrations of CH3Hg+. The FAM fluorescence increases with the increasing CH3Hg+ concentrations up to ∼20 μM. The fluorescence enhancement observed for Hg2+ is much smaller as a consequence of fluorescence quenching by Ag/Hg amalgams on the DNA (Figure S13a, see the Supporting 2456

DOI: 10.1021/ac504538v Anal. Chem. 2015, 87, 2452−2458

Article

Analytical Chemistry Information). The ratio of signal to background, S/B,56 is plotted as a function of the target concentration (Figure 6b). A linear correlation (R2 = 0.990) was found over the range of 2.0 nM−12.0 μM CH3Hg+ with a detection limit of 0.4 nM. The dynamic response for Hg2+ shifts to a higher range, only >50 nM Hg2+ can cause a real change in fluorescence emission, revealing a >125-fold sensitivity to CH3Hg+ over Hg2+. Thanks to difference of the characteristic absorption of Ag/ Hg amalgam and AgNPs, visual recognition of CH3Hg+ and Hg2+ could be easily achieved. Only Hg2+ induces a distinct color change from brilliant yellow to gray, which corresponds to the shift of the maximum absorption band of AgNPs from 400 to 320 nm (Figure 3a, inset, and Figure S15, see the Supporting Information). The specificity for fluorescent detection of CH3Hg+ was studied by examining the metalinduced fluorescence changes of FAM-HT7/Ag+ in the absence and the presence of CH3Hg+ via NaBH4-reduction (Figure S15, see the Supporting Information). Only CH3Hg+ shows a significantly higher fluorescence, Hg22+ ions enhance the fluorescence signal in a manner similar to Hg2+ ions, whereas other metals, such as Mg2+, Cd2+, Co2+, Pb2+, Mn2+, Fe3+, Ni2+, and Zn2+, show far weaker responses (Figure 6c). Cu2+ exhibits slight fluorescence enhancement because of the formation of Ag/Cu nanoclusters which may decrease the quenching effect of AgNPs,57 while Au3+ leads to an obvious fluorescence decrease due to the adsorption of DNA on AuNPs to quench the FAM-HT7 fluorescence.58 It is exciting that the equivalent Hg2+/Hg22+ have no influence on CH3Hg+ determination, in fact, the fluorescence response of CH3Hg+ with 50-fold excess of Hg2+ and other metal ions (Figure S13b, see the Supporting Information) is superimposable with that of CH3Hg+. It is the strongest binding affinity of CH3Hg+ to FAM-HT7 that suppresses the DNA binding with other metal ions, thus, in turn forms the DNA-assembled metal nanostructures to quench the fluorescent tag. Collectively, these results indicate that our approach is not only insensitive to other cations but also selective toward CH3Hg+ when they are present. The performance of CH3Hg+ detection by our method is compared to the reported ones (Table S6, see the Supporting Information), which shows a lower (at least comparable) limit of detection and a wider dynamic range. Finally, we extended the studies to the determination of CH3Hg+ bioaccumulation in fish samples. To validate the proposed method, a certified reference material ERM-CE464 of tuna fish was used.40 A standard sample dealt procedure (see Supporting Information) was employed for releasing CH3Hg+ from the −SH groups complexing mercury species,46 the concentration of CH3Hg+ in the ERM-CE464 was determined to be 5.78 ± 0.47 μg/g, which matches well with the certified value of 5.5 ± 0.2 μg/g (Table S7, see the Supporting Information). Next, three common fishes (snakehead, bighead carp, grass carp) from a local market were chosen to investigate the CH3Hg+ content (Table 1 and Figure S16, see the Supporting Information). Several ng/g levels of CH3Hg+ in the samples were detected, which meets the standard of 0.3 mg/kg (wet) defined by the U.S. EPA.7 Moreover, we find that the content of CH3Hg+ in the three species follows the order of snakehead > bighead carp > grass carp, revealing that CH3Hg+ can be accumulated through the food chain,6,59 since the snakehead is carnivorous fish, while the grass carp is a herbivorous one, and the bighead carp falls in between. A batch of crucian carp were chosen to study the distribution and change of CH3Hg+ uptake in different tissues/organs of fish

Table 1. Concentrations of CH3Hg+ in Tissues of Different Fish Samples From a Market in Changsha, China [Meana ± SDb, n = 5 (Determinations of Five Individuals)] CH3Hg+ (ng/g wet) sample

muscle

liver

gill

snakehead bighead carp grass carp

47.26 ± 22.42 13.17 ± 5.42 7.05 ± 3.14

18.10 ± 7.67 10.15 ± 3.25 −c

9.65 ± 2.91 6.25 ± 1.93 −c

Mean of determinations of five individuals. bSD, standard deviation. No CH3Hg+ detected.

a c

over time. The fish were fed in a CH 3 Hg +-enriched environment, with water changed each day to keep the concentration of CH3Hg+ at 10 μg/L. The CH3Hg+ contents of different tissues/organs were measured at 0, 5, 10, 20, and 30 days (Figure S17, see the Supporting Information). Figure 6d shows that CH3Hg+ was absorbed seriously by fish through time, the accumulation of CH3Hg+ concentrations at each time follows the order of gill > kidney > liver > brain > muscle. Notably, amounts of CH3Hg+ in the gill, liver, and muscle follow the inverse order in contrast to the market samples. As the filter of the fish, the gill is polluted first, with the shortest time (5 days) to reach saturation at the highest amount of CH3Hg+. However, in fresh water, CH3Hg+ expels from fish quickly and falls to a low level. In muscle, the concentrations increase slowly and continually, indicating the accumulation of CH3Hg+ in the muscle is a long-term procedure.60 In liver and kidney, CH3Hg+ increases largely with time, revealing the chronic toxicity to the two organs. We also observed a staged accumulation for brain, with a little change during first 10 days while a sharp increase after 20 days. The high lipid solubility helps CH3Hg+ readily crossing the blood-brain barrier, accumulating in the brain and causing damage to the central nervous system, which is more time-consuming.

■

CONCLUSION

In summary, we report, for the first time, that metal amalgamation, combining with DNA-protected AgNPs, can be used to detect CH3Hg+ with high sensitivity and superior selectivity over Hg2+. On the basis of the higher affinity of nucleotide for CH3Hg+ over Hg2+, and the formation of Ag/Hg amalgams between Ag0 and Hg0 rather than CH3Hg+, a CH3Hg+-specific DNA was selected as the key point for discrimination between CH3Hg+ and Hg2+ through the fluorescence light up. The system can detect down to the picomolar level of CH3Hg+, without interference from 50-fold Hg2+ and 106-fold other metal ions. The utility of the method was demonstrated through determination of CH3Hg+ content in three fish samples, the results show the accumulated order of snakehead > bighead carp > grass carp through food chain. Finally, the bioaccumulation of CH3Hg+ in the model crucian carp is determined to reveal various behaviors of CH3Hg+ accumulating rates and amounts. This simple, rapid, and costeffective process of our approach should afford potential to better understand the role of CH3Hg+ in human health and the environment. Moreover, this strategy promises to open up a new field in designing a series of sensors by utilizing different metal-specific DNA scaffolding as well as different types of signal transduction. 2457

DOI: 10.1021/ac504538v Anal. Chem. 2015, 87, 2452−2458

Article

Analytical Chemistry

■

(24) Nolan, E. M.; Lippard, S. J. J. Am. Chem. Soc. 2007, 129, 5910− 5918. (25) Ando, S.; Koide, K. J. Am. Chem. Soc. 2011, 133, 2556−2566. (26) Kim, L.-B.; Bunz, U. H. F. J. Am. Chem. Soc. 2006, 128, 2818− 2819. (27) Dave, N.; Chan, M. Y.; Huang, P.-J. J.; Smith, B. D.; Liu, J. J. Am. Chem. Soc. 2010, 132, 12668−12673. (28) Lee, J.-S.; Han, M. S.; Mirkin, C. A. Angew. Chem., Int. Ed. 2007, 46, 4093−4096. (29) Xue, X. J.; Wang, F.; Liu, X. G. J. Am. Chem. Soc. 2008, 130, 3244−3245. (30) Li, D.; Wieckowsk, A.; Willner, I. Angew. Chem., Int. Ed. 2008, 47, 3927−3931. (31) Ono, A.; Togashi, H. Angew. Chem., Int. Ed. 2004, 43, 4300− 4302. (32) Liu, J. W.; Lu, Y. Angew. Chem., Int. Ed. 2007, 46, 7587−7590. (33) Hollenstein, M.; Hipolito, C.; Lam, C.; Dietrich, D.; Perrin, D. M. Angew. Chem., Int. Ed. 2008, 47, 4346−4350. (34) Chen, P.; He, C. J. Am. Chem. Soc. 2004, 126, 728−729. (35) Nolan, E. M.; Lippard, S. J. Chem. Rev. 2008, 108, 3443−3480. (36) Yang, Y.-K.; Ko, S.-K.; Shin, I.; Tae, J. Org. Biomol. Chem. 2009, 7, 4590−4593. (37) Santra, M.; Ryu, D.; Chatterjee, A.; Ko, S.-K.; Shin, I.; Ahn, K. A. Chem. Commun. 2009, 2115−2117. (38) Chen, X. Q.; Baek, K.-H.; Kim, Y.; Kim, S.-J.; Shin, I.; Yoon, J. Tetrahedron 2010, 66, 4016−4021. (39) Tseng, W.-L.; Lin, Y.-H. Anal. Chem. 2010, 82, 9194−9200. (40) Climent, E.; Marcos, M. D.; Manez, R. M.; Sancenon, F.; Soto, J.; Rurack, K.; Amoros, P. Angew. Chem., Int. Ed. 2009, 48, 8519−8522. (41) Liu, Y.; Chem, M.; Cao, T. Y.; Sun, Y.; Li, C. Y.; Liu, Q.; Yang, T. S.; Yao, L. M.; Feng, W.; Li, F. Y. J. Am. Chem. Soc. 2013, 135, 9869−9876. (42) Cho, E. S.; Kim, J.; Tejerina, B.; Hermans, T. M.; Jiang, H.; Nakanishi, H.; Yu, M.; Patashinski, A. Z.; Glotzer, S. C.; Stellacci, F.; Grzybowski, B. A. Nat. Mater. 2012, 11, 978−985. (43) Katsikas, L.; Gutierrez, M.; Henglein, A. J. Phys. Chem. 1996, 100, 11203−11206. (44) Yosypchuk, B.; Barek, J. Crit. Rev. Anal. Chem. 2009, 39, 189− 203. (45) Li, Y.; Liu, J. M.; Han, F.; Jiang, Y.; Yan, X. P. J. Anal. At. Spectrom. 2011, 26, 94−99. (46) Bloom, N. S. Can. J. Aquat. Sci. 1989, 46, 1131−1140. (47) Deng, L.; Ouyang, X. Y.; Jin, J. Y.; Ma, C.; Jiang, Y.; Zheng, J.; Li, J. S.; Li, Y. H.; Tan, W. H.; Yang, R. H. Anal. Chem. 2013, 85, 8594−8600. (48) Miyake, Y.; Togashi, H.; Tashiro, M.; Yamaguchi, H.; Oda, S.; Kudo, M.; Tanaka, Y.; Kondo, Y.; Sawa, R.; Fujimoto, T.; Machinami, T.; Ono, A. J. Am. Chem. Soc. 2006, 128, 2172−2173. (49) Li, Y.; Jiang, Y.; Yan, X. P. Anal. Chem. 2006, 78, 6115−6120. (50) Simpson, R. B. J. Am. Chem. Soc. 1964, 86, 2059−2065. (51) Izatt, R. M.; Christensen, J. J.; Rytting, J. H. Chem. Rev. 1971, 71, 439−481. (52) Scatchard, G. Ann. N.Y. Acad. Sci. 1949, 51, 660−672. (53) Petty, J. T.; Zheng, J.; Hud, N. V.; Dickson, R. M. J. Am. Chem. Soc. 2006, 126, 5207−5212. (54) Kelly, K. L.; Coronado, E.; Zhao, L. L.; Schatz, G. C. J. Phys. Chem. B 2003, 107, 668−677. (55) Henglein, A.; Brancewicz, C. Chem. Mater. 1997, 9, 2164−2167. (56) Yang, C. Y. J.; Lin, H.; Tan, W. H. J. Am. Chem. Soc. 2005, 127, 12772−12773. (57) Su, Y. T.; Lan, G. Y.; Chen, W. Y.; Chang, H. T. Anal. Chem. 2010, 82, 8566−8572. (58) Wang, H.; Li, J. S.; Wang, Y. X.; Jin, J. Y.; Yang, R. H.; Wang, K. M.; Tan, W. H. Anal. Chem. 2010, 82, 7684−7690. (59) Zhu, A.; Zhang, X.; Xu, Z.; Huang, L.; Wang, W. X. Mar. Pollut. Bull. 2013, 77, 437−444. (60) Huang, S. S.-Y.; Strathe, A. B.; Fadel, J. G.; Lin, P.; Liu, T. Y.; Hung, S. S. O. Aquat. Toxicol. 2012, 123, 163−171.

ASSOCIATED CONTENT

S Supporting Information *

Additional information as noted in the text. This material is available free of charge via the Internet at http://pubs.acs.org.

■

AUTHOR INFORMATION

Corresponding Author

*Fax: +86-731-8882 2523. E-mail: [email protected]. Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS We are grateful for the financial support from the National Natural Science Foundation of China (Grants 21135001, 21405038, 21375095, 21305036, and J1103312), The Foundation for Innovative Research Groups of NSFC (Grant 21221003), and the “973” National Key Basic Research Program (Grant 2011CB91100-0). We also sincerely appreciate Professor Lu for his kindly help with manuscript preparation and revision.

■

REFERENCES

(1) Miller, J. R.; Rowland, J.; Lechler, P. J.; Desilets, M.; Hsu, L.-C. Water, Air, Soil Pollut. 1996, 86, 373−388. (2) Toxicological Profile for Mercury; Agency for Toxic Substances and Disease Registry (ATSDR): Atlanta, GA, 1999. (3) Clarkson, T. W.; Magos, L. Crit. Rev. Toxicol. 2006, 36, 609−662. (4) Castoldi, A. F.; Johansson, C.; Onishchenko, N.; Coccini, T.; Roda, E.; Vahter, M.; Ceccatelli, S.; Manzo, L. Regul. Toxicol. Pharmacol. 2008, 51, 201−214. (5) Poulain, A. J.; Barkay, T. Science 2013, 339, 1280−1281. (6) Uppenbrink, J. F. Science 2013, 342, 779. (7) U.S. EPA. Draft Guidance for Implementing the January 2001 Methylmercury Water Quality Criterion [S]. EPA-823-R-01-001; Office of Science and Technology: Washington, DC, 2006; pp 1−20. (8) WHO. Guideline Levels for Methylmercury in Fish [S], CAC/GL71991; WHO: Geneva, Switzerland, 1991; p 1. (9) Diez, S. Rev. Environ. Contam. Toxicol. 2009, 198, 111−132. (10) Magos, L. Nature 1977, 269, 183. (11) Bakir, F.; Damluji, S. F.; Amin-Zaki, L.; Murtadha, M.; Khalidi, A.; Al-Rawi, N. Y.; Tikriti, S.; Dhahir, H. I.; Clarkson, T. W.; Smith, J. C.; Doherty, R. A. Science 1973, 181, 230−241. (12) Bergquist, B. A.; Blum, J. D. Science 2007, 318, 417−420. (13) Daniel, C. Nat. Geosci. 2013, 6, 810−811. (14) Egeland, G. M.; Middaugh, J. P. Science 1997, 278, 1904−1905. (15) Engstrom, D. R. Proc. Natl. Acad. Sci. U.S.A. 2007, 104, 16394− 16395. (16) Kelly, C. A.; Rudd, J. W. M.; Louis, V. L. S.; Heyes, A. Water, Air, Soil Pollut. 1995, 80, 715−724. (17) Erxleben, H.; Ruzicka, J. Anal. Chem. 2005, 77, 5124−5128. (18) Zhu, Z. L.; Chan, G. C. -Y.; Ray, S. J.; Zhang, X. R.; Hieftje, G. M. Anal. Chem. 2008, 80, 7043−7050. (19) Rodrigues, J. L.; Torres, D. P.; Souza, V. C. O.; Batista, B. L.; Souza, S. S.; Curtius, A. J.; Barbosa, F. J. Anal. At. Spectrom. 2009, 24, 1414−1420. (20) Li, Y.; Yin, Y.; Liu, G.; Cai, Y. Advances in Speciation Analysis of Mercury in the Environment. In Environmental Chemistry and Toxicology of Mercury; Liu, G., Cai, Y., Driscoll, N. O., Eds.; John Wiley & Sons: Hoboken, NJ, 2012; pp 15−58. (21) Tutschku, S.; Schantz, M. M.; Wise, S. A. Anal. Chem. 2002, 74, 4694−4701. (22) Lemes, M.; Wang, F. Y. J. Anal. At. Spectrom. 2009, 24, 663− 668. (23) Yang, Y.-K.; Yook, K.-J.; Tae, J. S. J. Am. Chem. Soc. 2005, 127, 16760−16761. 2458

DOI: 10.1021/ac504538v Anal. Chem. 2015, 87, 2452−2458