Selective aptamer conjugation to silver-coated magnetite nanoparticles for magnetic solid-phase extraction of trace amounts of Pb2+ ions

Herein, a novel aptamer-functionalized magnetic adsorbent was developed and combined with magnetic solid-phase extraction (MSPE) for the specific enrichment of Pb2+ ions prior to flame atomic absorption spectrometric detection. First, silver-coated magnetite core–shell nanoparticles (Fe3O4@Ag MNPs) were synthesized by the chemical reduction of silver ions on the surface of magnetite nanoparticles. After that, the selective DNA aptamer against Pb2+ was conjugated on the surface of the synthesized nanoparticles to form aptamer-modified magnetic nanoparticles (Fe3O4@Ag-APT). The characterization of the prepared adsorbent was performed through SEM imaging, XRD, FT-IR, EDX, and DRS instruments. The influence of the various experimental parameters on the adsorption and desorption steps in MSPE was investigated via Taguchi experimental design to optimize different parameters. Under the optimized conditions, the Pb2+ calibration graph was linear in the range of 33–1000 μg L−1. The relative standard deviation (RSD%) of the method for six replicates containing 100 μg L−1 of Pb2+ ions was 0.34%. Furthermore, the limit of detection (LOD) and the limit of quantification (LOQ) were 10 μg L−1 and 33.3 μg L−1, respectively. Finally, the applicability of the proposed method was successfully confirmed by preconcentration and determination of trace amounts of Pb2+ ions in tap and seawater samples. We showed a proof of concept for Fe3O4@Ag-APT as an efficient bio-adsorbent, offering a promising strategy for the specific binding/removal of toxic heavy metal ions.


Introduction
One of the most serious ecological problems affecting the quality of water is chemical pollution by inorganic or organic compounds, including heavy metals, pharmaceuticals, dyes, drugs, pesticides and biocides. 1 Various toxic heavy metals can accumulate in water and different food chains, such as human bodies via food, drinking water, breathing airborne particles and road dust . 2 Therefore, the continuous monitoring of trace levels of metal ions in the environment is of major importance. Among the inorganic pollutants, lead is highly toxic owing to its accumulative toxicity to humans and animals. At high exposure levels in blood, it can affect human health by causing coma, convulsion, anemia, damage to organs, such as brains, bones, kidneys, and muscles, and also many serious illnesses or even mortality. 3,4 Atomic absorption spectrometry (AAS) is known as the common instrumental method for the determination of heavy metals. However, the conventional determination of metal ions at low concentration by ame atomic absorption spectrometry (FAAS) is frequently not possible because their concentrations in water samples are typically at mg L À1 levels and below the detection limit of FAAS. To solve this problem, preconcentration of an analyte and/or separation procedures coupled with FAAS have been proposed. [5][6][7][8][9] Solid-phase extraction (SPE) is the most popular pretreatment method for sample matrix simplication and the enrichment of trace amounts of analyte in samples. SPE has become a strong method compared to preparation techniques such as liquid-liquid extraction, oatation, membrane ltration, and cloud point extraction because of the advantages such as high reproducibility, simplicity, low cost, lower consumption of organic solvents, high enrichment factor and applicability in different formats for various organic or inorganic analytes. The ability to be coupled with different detection techniques and a wide variety of sorbents are some of the benets of SPE. [10][11][12] Several types of adsorbent, such as mesoporous materials, activated carbon, ion-imprinted and carbon nanotubes, resins and nanoparticles, have been used in SPE. 13,14 Among these sorbents, magnetic nanoparticles have attracted great interest because the magnetic separation shows excellent ability for the collection of magnetic nanomaterials from sample solutions by using an external magnetic eld. [15][16][17][18][19][20][21] Over the last decade, it has become popular for detection systems to use DNA for the detection of metal ions, mainly because DNA has good stability, is highly soluble in aqueous solution, and is a specic binder of some metal ions. 22,23 Aptamers are short single-stranded oligonucleotides, either DNA or RNA (ssDNA or ssRNA), with affinity and specicity for the recognition, separation and detection of target compounds, namely peptides, toxins, proteins, drugs, microbes, organic and inorganic molecules, metal ions or even entire cells. 24,25 Aptamers can be natural or synthetic in origin. They are isolated and chemically synthesized using combinatorial enrichment (SELEX). 26,27 Aptamers bind constantly to a specic molecular target and as a result of their specic three-dimensional (3D) structure they can form particular complexes with the target molecule, which is a powerful tool for the identication of the goal molecules. [28][29][30] Aptamers have been used in SPE for the enrichment, purication and extraction of particular analytes. 31,32 The great merit of extraction using aptamers is the high selectivity, simple analytical process, extraction recovery, and fast processing. Because of the numerous advantages of aptamers, such as easy and economical synthesis and chemical modication, excellent stability against temperature and pressure variations, high affinity and specicity, small size, and reversible denaturation, 33 they can be used for conjugating/ modifying nanomaterials for the detection and diagnosis of heavy metals, proteins, and cancer cells. [34][35][36] Recently, several strategies have been suggested for applying aptamers conjugated with nanomagnetic materials, including the use of aptamers combined with magnetic nanoparticles as modied adsorbents in magnetic solid-phase extraction (MSPE). 16,17,37,38 The aim of this study was to develop a novel selective aptamer-based MSPE for the preconcentration and detection of Pb 2+ ions by FAAS. First, Fe 3 O 4 MNPs were synthesized via chemical precipitation method. The core-shell-structured Fe 3 O 4 @Ag MNPs were prepared by chemical reduction of silver ions on the surface of magnetite nanoparticles. The 15mer DNA aptamer selective for Pb 2+ ions 39,40 was immobilized on the surface of the Fe 3 O 4 @Ag MNPs, giving the aptamerbased MSPE (APT-MSPE) sorbent. The inuence of various experimental factors on the adsorption and preconcentration of Pb 2+ , including pH of solution, contact time, amount of nanoparticles, volume of aptamer, type and volume of elution solvent, concentration of elution solvent and time of desorption, was investigated via Taguchi experimental design method to optimize the different parameters. The prepared Fe 3 O 4 @Ag-APT sorbent was evaluated in terms of selectivity, binding capacity and extraction ability. Sigma-Aldrich (ST. Louise, Missouri, USA). The 5 0 -thiol-modied DNA oligonucleotide (5 0 -GGTTGGTGTGGTTGG-3 0 ) for the aptamer was synthesized and puried by HPLC in Generay Biotech (Shanghai) Co. Ltd (China). This aptamer is a kind of oligonucleotide (single-stranded DNA) consisting of nine deoxyguanosine and six thymidine residues connected by 3 0 / 5 0 phosphodiester linkages in the sequence 5 0 -HS-G-G-T-T-

Instruments and apparatus
The prepared Fe 3 O 4 @Ag nanoparticles (MNPs) were characterized by X-ray diffraction (XRD) using CuKa with 2q values of 5-80 . Scanning electron microscopy and energy-dispersive X-ray spectroscopy (SEM and EDX, model LEO1430VP, England and Germany) were used to observe the surface morphology and for the elemental analysis of the magnetic nanoparticles. Fourier transform infrared (FT-IR) spectra of the synthesized MNPs were obtained using an FT-IR instrument from Shimadzu (model 8900, Japan) in the range of 400-4000 cm À1 . The UV-vis diffuse reectance spectroscopy (DRS) of the synthesized nanoparticles was recorded on a Scinco spectrophotometer (S-4100, South Korea) equipped with a DRS accessory. In this work, a ame atomic absorption spectrometer (FAAS) model AA240FS (Varian, USA) equipped with a deuterium lamp for background correction was used for the determination of Pb 2+ ions in an air-acetylene ame. The parameters for the FAAS detection of Pb 2+ ions were as follows: hollow cathode lamp (HCL); wavelength, 217 nm; slit width, 1.0 nm; and lamp current, 10 mA. The pH of the solutions was measured using a Bante pH meter equipped with a combined glass electrode (China). For magnetic separation, a strong permanent magnet (1 Â 3 Â 5 cm) with a 1.4 T magnetic eld was applied. A magnetic stirrer (Labinco, The Netherlands) and stirrer bar (4 Â 14 mm) were used in the nanoparticle synthesis. Then, the clear yellow solution was degassed with nitrogen gas for 10 min. Simultaneously, 250 mL of ammonia solution (4.5 mol L À1 ) was degassed with nitrogen gas and heated to 80 C in a reactor. Aer that, the mixture of ferric and ferrous ions was added to the ammonia solution slowly via a dropping funnel with vigorous stirring (1000 rpm) under gas protection for 30 min. During the whole process, the temperature of the solution was held at 80 C and nitrogen gas was purged to remove the dissolved oxygen. Aer the formation of the black magnetic nanoparticles (Fe 3 O 4 MNPs), they were separated from the reaction medium using an external magnet and washed with double-distilled water and sodium hydroxide solution (0.1 mol L À1 ).
The Fe 3 O 4 @Ag MNPs were prepared by chemical reduction of silver nitrate on the surface of the MNPs in advance using the following modied protocol. 42 Briey, 1 g of Fe 3 O 4 MNPs was dispersed in 50 mL of double-distilled water, followed by the addition of 25 mL of AgNO 3 (0.001 mol L À1 ) under vigorous stirring. Aer 30 min, 50 mL of 3 Â 10 À4 mol L À1 NaBH 4 , as a reductant, was added dropwise to the suspension under vigorous mechanical stirring for another 30 min. Aer this time, the green core-shell nanoparticles were collected using a permanent magnet, washed several times with doubledistilled water to remove excess Ag particles and then dried at 70 C.
2.3.2. Immobilization of the aptamer on the magnetic nanoparticles. The purchased aptamer sample was centrifuged at 10 000 rpm (2 min) to settle the aptamer at the bottom of the vial. The stock solution of the thiolated lead-binding aptamer (100 mM) was prepared via addition of 123 mL of 0.1 mol L À1 PBS (pH 7.4) to the aptamer vial. Aer vortexing (1 min), the mixture was kept at À20 C. Aptamer solutions with concentrations of 10 and 5 mM were prepared by diluting the aptamer stock solution (100 mM) with 0.1 mol L À1 PBS solution and vortexing (1 min); the solutions were then stored at À20 C.
For synthesis of the Fe 3 O 4 @Ag-APT MNPs, the volume of aptamer solution and the weight of sorbent were optimized by Taguchi experimental design. For the synthesis, 0.04 g of Fe 3 O 4 @Ag MNPs was weighed and transferred into a 10 mL beaker. Aer addition of 1 mL of PBS, the suspension was sonicated (5 min) to prevent aggregation of the MNPs. A certain  2.3.3. Magnetic solid-phase extraction of Pb 2+ ions using Fe 3 O 4 @Ag-APT sorbent. For each MSPE experiment, 25 mL of an aqueous sample containing certain amounts of Pb 2+ ions was poured into a beaker and, aer adjusting the pH of the solution to 7.0, 0.04 g of Fe 3 O 4 @Ag-APT MNPs was added to the mixture. Aerwards, the mixture was mechanically stirred for 35 min at 400 rpm. Aer this time, the MNPs were isolated from the mixture using an external magnetic eld by applying a strong super magnet. The supernatant was transferred to a vial to check for the presence of Pb 2+ . The preconcentrated ions were desorbed from the collected MNPs during stirring with 3 mL of 0.25 mol L À1 HCl (10 min). Aer that, the solution containing the eluted Pb 2+ ions was collected in a vial for further measurement by FAAS (Fig. 2).

Principle of the MSPE strategy using Fe 3 O 4 @Ag-APT MNPs
Metal coordination by DNA has been extensively studied. [43][44][45] DNA is a polyanion allowing electrostatic attraction with metal ions. DNA phosphates can bind hard/borderline metals, while various bases coordinate with metal ions with different affinities. At the simplest level, metal ions are treated as point charges diffusing around DNA polyanions by pure electrostatic interactions. Metal binding to the phosphate backbone stabilizes the DNA duplex (e.g., increasing DNA melting temperature, Tm). Pb 2+ -dependent G-rich ssDNAs can be integrated into Pb 2+ biosensors. Upon addition of Pb 2+ ions to the solution, the conformation of the Pb 2+ -dependent aptamer changes from a random coil structure to a G-quadruplex one.

Characterization of Fe 3 O 4 @Ag MNPs
The Fe 3 O 4 @Ag MNPs were characterized using XRD, SEM, EDX, FT-IR and DRS analysis.
The crystal phases and crystallinity of Fe 3 O 4 and Fe 3 O 4 @Ag MNPs were analyzed by XRD and measured with Cu Ka radiation in the 2q values, of 5-80 , with a step size of 0.0260 and count time of 0.5 /s. According to the results (Fig. 3(a and b)), the characteristic peaks at 2q of 30.3 (220), 35 (Fig. 3(a)), which is in agreement with the standard of Fe 3 O 4 (JCPDS No: 19-0629) and previous literature. 46 The absence of  (200), 64.6 (220) and 77.5 (311) are related to the planes of silver ( Fig. 3(b)), in agreement with the XRD pattern of Ag NPs (JCPDS No: 04-0783). The XRD patterns conrmed that no impurities were observed.
SEM observations were performed to characterize the morphology and size of the synthesized Fe 3 O 4 @Ag MNPs with magnication of 15.0 and 70.0kx (Fig. 4). Based on the SEM images, the synthesized Fe 3 O 4 @Ag MNPs were uniform spherical particles with an average diameter of less than 38 nm. The nanoparticles could provide a large specic surface area and numerous adsorption sites. Moreover, the addition of the aptamer did not alter the morphology or particle size of the nanoparticles distinctly.
Elemental analysis was also used to conrm the existence of silver element on the Fe 3 O 4 nanoparticles. The atomic composition of the Fe 3 O 4 @Ag was evaluated by EDX analysis.

Optimization of the MSPE adsorption parameters
The 5 0 -thiolated aptamer was attached to the Ag layer of the Fe 3 O 4 @Ag MNPs to form an aptamer-based sorbent. The Pb 2+ ions reacted with the functional groups of the oligonucleotide present in the structure of the aptamer and were separated from the solution.
For the MSPE optimization experiments, 25 mL of a solution containing 1 mg L À1 Pb 2+ was exposed at room temperature with the proper amount of sorbent. Aer magnetic separation, the Pb 2+ adsorption and desorption efficiencies were determined using FAAS (Fig. 2). In preliminary tests, the effect of the volume of aptamer solution used in the synthesis of the sorbent was investigated. For this purpose, 2, 5, 10, 15 and 20 mL of 5 mM aptamer solution were used for 0.02 g of Fe 3 O 4 @Ag MNPs to produce the aptamer-based sorbent. According to the preliminary tests, at aptamer volumes lower than 15 mL, the aptamer did not attach to the MNPs properly. Therefore, in the optimization table, aptamer volumes of 15 mL and higher were considered. All of the parameters affecting the adsorption and desorption of Pb 2+ ions on the sorbent were optimized separately via Taguchi orthogonal optimization to obtain the highest sensitivity of sorbent. To fully develop the aptamer-based sorbent for analysis of Pb 2+ ions, some signicant adsorption factors including the pH of the solution (6.0, 7.0, 7.5), contact time (15, 35, 45 min), sorbent weight (0.02, 0.04, 0.06 g) in  solution and volume of aptamer solution (15, 35, 45 mL) were investigated and optimized via Taguchi orthogonal array design using an L 9 array (4 factors at 3 levels, Table 1). Aer carrying out the Taguchi design optimization experiments, statistical calculations were performed and the average main effects of each factor were obtained at different levels using Minitab soware (version 18) (Fig. 8). Based on the results of the experiments and the analysis of variance (ANOVA), the optimal conditions for the adsorption of Pb 2+ ions were obtained according to the mean main effects for each parameter (Fig. 8).
Solution pH is an important parameter affecting the efficiency of Pb 2+ ion adsorption from aqueous samples. pH plays an important role in the complexation of Pb 2+ ions with the aptamer. In this study, the solution pH was varied between 6 and 7.5 (near the physiological pH). According to the results, the maximum Pb 2+ adsorption efficiency was observed at pH ¼ 7.0. According to the results, at physiological pH, the synthesized aptamer is extremely stable and has a negative charge on its surface. Therefore, a pH of 7.0 was selected for further studies.
The maximum adsorption efficiency was obtained with 0.04 g of aptamer, which can be attributed to the increase in the number of sorbent sites in the solution. However, the extraction efficiency decreased when more than 0.04 g of Fe 3 O 4 @Ag MNPs was used. This is due to that a small quantity of Fe 3 O 4 @Ag MNPs could not be dispersed due to hydrophobic surface of it or attachment of Fe 3 O 4 @Ag MNPs onto magnet. This hindered the collection of all of the sorbent by an external magnet, which resulted in a slight decrease in the extraction efficiency. Based on the results, 0.04 g of sorbent provided the highest extraction efficiency for all analytes.
Moreover, the amount of lead ions bound to the Fe 3 O 4 @Ag-APT is a function of the aptamer surface density, which clearly justies the need for optimization. Therefore, the different amounts of the thiolated DNA aptamer were added to 1 mL of a suspension containing 0.04 g of Fe 3 O 4 @Ag in PBS and then the aptamers were anchored onto the surface of the magnetic nanoparticles via immobilization by thiol chemistry. It was found that a low aptamer density on the surface, which affects the amount of target binding, leads to low signal levels. However, the maximum amount of target bound is a function of the aptamer surface density. The aptamer surface density increased and reached a maximum with 45 mL of aptamer.
In the case of the contact time, the majority of lead ions were captured by aptamer-modied Fe 3 O 4 @Ag MNPs within 20 min and the maximum recovery for Pb 2+ ions was obtained within 35 min. However, a long stirring time may cause partial release of adsorbed Pb 2+ ions from the sorbent.

Optimization of Pb 2+ desorption parameters affecting the MSPE procedure
Aer adsorption of Pb 2+ ions using Fe 3 O 4 @Ag-APT MNPs and prior to their measurement by FAAS, it is necessary to desorb the adsorbed ions using the minimum volume of suitable eluent. For this purpose, several desorption parameters, including type of eluent (HCl in distilled water, HCl in phosphate buffer and HCl in ethanol), eluent concentration (0.05,   Table 2) to achieve the best Pb 2+ ion desorption efficiency. The desorption optimization experiments were carried out in the optimal conditions of adsorption process (pH ¼ 7.0, sorbent weight ¼ 0.04 g, contact time ¼ 35 min and volume of aptamer ¼ 45 mL).
To nd the best eluent type to desorb the Pb 2+ ions from the aptamer, HCl in three different solvents was investigated (distilled water, ethanol (EtOH) and PBS (0.1 mol L À1 )). Among them, HCl in DW was found to be superior for the desorption of Pb 2+ ions from the aptamer surface (Fig. 9). Fig. 9 shows the effect of eluent volume (3 to 10 mL) on the amount of Pb 2+ desorbed. The results indicate that by increasing the volume of the eluent, the Pb 2+ absorbance and therefore its concentration in desorbed solution are reduced owing to the dilution of the eluent phase. Thus, to increase the preconcentration factor, 3 mL of eluent was considered to be the optimal eluent volume.
In order to investigate the effect of the concentration of the eluent solvent on the preconcentration of Pb 2+ ions, different HCl concentrations were investigated in the range of 0.05 to 0.25 M. Based on Fig. 9, quantitative recovery of Pb 2+ ions could be obtained with eluent containing 0.25 M HCl.
To achieve the highest preconcentration and optimal recovery of the adsorbed Pb 2+ ions, the effect of desorption time was also tested between 5 and 15 min. The results showed that the maximum desorption efficiency of Pb 2+ ions was obtained with 10 min of shaking of aptamer in 3 mL of HCl (0.25 mol L À1 ) (Fig. 9).

Analytical gures of merit
Under the optimized adsorption and desorption conditions (pH ¼ 7.0, sorbent weight ¼ 0.04 g, volume of aptamer ¼ 45 mL, contact time ¼ 35 min, desorption time ¼ 10 min with 3 mL HCl in distilled water (0.25 M)), several analytical characteristics of the developed method were investigated, including linear dynamic range (LDR), limit of detection (LOD), limit of quantitation (LOQ) and enrichment factor (EF). The calibration curve was obtained without and aer preconcentration of Pb 2+ ions. The linear range for direct measurement without preconcentration was found to be 300-1500 mg L À1 with a correlation coefficient (R 2 ) of 0.9991 while the LDR aer preconcentration was observed to be in the range of 33-1000 mg L À1 with R 2 ¼ 0.9972. The limit of detection (3 S b /m) and limit of quantitation (10 S b /m) were 10 mg L À1 and 33.0 mg L À1 , respectively. In these equations, m is the slope of the calibration graph aer preconcentration and S b is the standard deviation of four independent replicate preconcentration of blank solution. The precision of the method was assessed as relative standard deviation (RSD, %) and calculated from six independent replicates at a Pb 2+ ion concentration of 100 mg L À1 and was found to be 0.34%. The EF of the proposed method, as the ratio of the  Fig. 9 The means of the main effects of each factor in optimization of the desorption parameters obtained by absorbance reading of UV-Vis measurements. A comparison with other SPE adsorbents used for preconcentration of Pb 2+ ions prior to FAAS is given in Table  4. 11,42,[54][55][56][57] Our proposed aptamer showed broad DLR and good LOD with respect to the other adsorbents. Moreover, by utilizing graphite furnace AAS (GF-AAS), ICP-AES or ICP-MS for the determination of Pb 2+ , the detection limit can be improved by more than 1000 times and it is possible to detect trace levels of Pb 2+ ions. In addition, utilizing lower eluent volumes in GF-AAS improves the preconcentration factor.

Conclusion
In this study, we reported the development of an aptasorbent Fe 3 O 4 @Ag-APT MNP for the selective extraction of Pb 2+ ions. The introduced magnetic aptasorbent can be used as an easy, reliable, inexpensive, selective and sensitive method for the extraction and/or preconcentration of trace levels of lead ions in various aqueous matrices prior to determination. Aer carrying out Taguchi design optimization experiments, the best results for the adsorption and desorption of Pb 2+ ions were obtained at pH 7.0 using 0.04 g of the sorbent modied with 45 mL of aptamer and a total contact time of 35 min for adsorption and 10 min for desorption, with 3 mL of HCl (0.25 mol L À1 ) as the eluent. Under the optimum conditions, Pb 2+ was detected in the range of 33-1000 mg L À1 with a low detection limit of 10 mg L À1 . It is noteworthy that, compared to previous works, this new protocol has several advantages. First, the sorbent is simpler than the previous ones; second, the results showed that the developed Fe 3 O 4 @Ag-APT exhibited a wide working range and impressively high selectivity toward Pb 2+ among other heavy metal ions in aqueous media and also had good potential for the extraction of Pb 2+ ions at mg L À1 concentrations.

Conflicts of interest
There are no conicts to declare.