Talanta 136 (2015) 68–74

Contents lists available at ScienceDirect

Talanta journal homepage: www.elsevier.com/locate/talanta

Development of rolling circle amplification based surface-enhanced Raman spectroscopy method for 35S promoter gene detection Burcu Guven a, Ismail Hakki Boyaci a,b,n, Ugur Tamer c, Esra Acar-Soykut b, Uzeyir Dogan c a

Department of Food Engineering, Faculty of Engineering, Hacettepe University, Beytepe, Ankara 06800, Turkey Food Research Center, Hacettepe University, Beytepe, Ankara 06800, Turkey c Department of Analytical Chemistry, Faculty of Pharmacy, Gazi University, Ankara 06330, Turkey b

art ic l e i nf o

a b s t r a c t

Article history: Received 17 September 2014 Received in revised form 14 November 2014 Accepted 22 November 2014 Available online 12 January 2015

In this study, we developed the genetically modified organism detection method by using the combination of rolling circle amplification (RCA) and surface-enhanced Raman spectroscopy (SERS). An oligonucleotide probe which is specific for 35S DNA promoter target was immobilised onto the gold slide and a RCA reaction was performed. A self-assembled monolayer was formed on gold nanorods using 5,50 -dithiobis(2-nitrobenzoic acid) (DTNB) and the second probe of the 35S DNA promoter target was immobilised on the activated gold coated slide surfaces. Probes on the nanoparticles were hybridised with the target oligonucleotide. Quantification of the target concentration was performed via SERS spectra of DTNB on the nanorods. SERS spectra of target molecules were enhanced through the RCA reaction and the detection limit was found to be 6.3 fM. The sensitivity of the developed RCA–SERS method was compared with another method which had been performed without using RCA reaction, and the detection limit was found to be 0.1 pM. The correlation between the target concentration and the SERS signal was found to be linear, within the range of 1 pM to 10 nM for the traditional assay and 100 fM to 100 nM for the RCA assay. For the developed RCA–SERS assay, the specificity tests were performed using the 35S promoter of Bt-176 maize gene. It was found out that the developed RCA–SERS sandwich assay method is quite sensitive, selective and specific for target sequences in model and real systems. & 2014 Elsevier B.V. All rights reserved.

Keywords: SERS DTNB Traditional and RCA sandwich assay

1. Introduction The importance attached to the genetically modified organism (GMO) detection is increasing on a daily basis. Sensitive, rapid and reliable methods for GMO detection are gaining worldwide attention, especially in molecular biology and clinical diagnostics [1]. Recently, two marker genes, the 35S promoter from cauliflower mosaic virus (CaMV35S) and the NOS terminator from Agrobacterium turmefaciens, were used for detection of GMO as they are present in nearly all genetically modified plants [2]. Most GMOs also contain the CaMV35S promoter, e.g. the Roundup Ready (RR) soybean, Bt176-, Bt11-, T25-, and Mon810-maize and the FlavrSavr tomato. The widespread use of the CaMV35S makes it the primary target for GMO identification [3]. Nowadays, surface enhanced Raman spectroscopy (SERS) has attracted considerable attention thanks to its high sensitivity [4,5], small sample volume [6], and stable and specific signals [7]. SERS has been reported to detect numerous types of biomolecules such

n

Corresponding author. Tel.: þ 90 312 297 61 46; fax: þ 90 312 299 21 23. E-mail address: [email protected] (I.H. Boyaci).

http://dx.doi.org/10.1016/j.talanta.2014.11.051 0039-9140/& 2014 Elsevier B.V. All rights reserved.

as carbohydrates [8], proteins [9], enzymes [10], bacteria [11,12], viruses, and nucleotides [13–15]. In the literature, there are two types of SERS methods; which are label-free [16–18] and those with label dye detection methods [10,11,15,19–22]. In the label dye detection method, a SERS active label dye such as 5,50 -dithiobis(2nitrobenzoic acid) (DTNB), rhodamine or Texas red is used for indirectly monitoring the presence of a specific molecule the direct identification of which would be considerably difficult due to low amounts of intrinsic Raman cross sections. This method enables the detection of lower concentrations relatively more accurately in comparison with label-free detection methods. In addition, SERS intensity can be enhanced using signal amplification methods such as polymerase chain reaction (PCR) and rolling circle amplification (RCA) without SERS active dyes. A newly-emerging amplification method is RCA, which polymerises nucleotides with a 1000-fold increase at a constant temperature of 30 1C in a much shorter time (1–2 h) [1,23]. In RCA, there is a circular-shaped RCA template that hybridises with the primer sequence, which is then extended with the RCA template's cycling and, this process, leads to the generation of long nucleic acid products [24]. The RCA method has come to the fore as an alternative for PCR, which is the basic method for biological research and

B. Guven et al. / Talanta 136 (2015) 68–74

diagnostics. In PCR, the main limitations are the need for thermal cycling steps and the cost of instruments [23,25]. RCA has gained remarkable attention for the ultrasensitive detection of DNA [26], RNA [27] and proteins [28]. In the literature, there are various RCA applications, such as fluorescence-based RCA [29,30], immuno-RCA [31,32], single nucleotide polymorphism (SNP) detection [33], DNA microarrays [26,34], clinical applications [35], surface plasmon resonance [36], radiolabeling, UV absorbance, electrochemical detection [24,37] and SERS applications [1,38]. In our previous studies, we developed SERS based methods for detection of proteins, carbohydrates, enzymes, bacteria, nucleotides (DNA and RNA). The aim of this study was to enhance the sensitivity of DNA detection developing RCA based SERS method and compare to the developed traditional SERS based sandwich method. In the traditional sandwich assay (without RCA), two target probes were immobilised onto the gold slide and rodshaped gold nanoparticles respectively. Then, they were hybridised with the target oligonucleotide sequence. However, in the RCA sandwich assay, the amplification was performed following the first hybridisation with the probe on the gold slide and the target sequence. After the ϕ29 DNA polymerase amplification was completed, the target sequence was detected sensitively. The analytical performances of traditional and RCA based sandwich methods were examined according to the linear range and detection limits. As the RCA–SERS method was found out to be more sensitive than the traditional SERS sandwich method, selectivity and specificity tests with control and real sample studies were carried out only for the developed RCA–SERS assay.

69

solutions and the electrophoresis running buffer. Agarose gels (1%) were prepared using agarose, ethidium bromide and 1X TAE buffer. PBS (20 mM, pH 7.4) was prepared using Na2HPO4, KH2PO4, NaCl, and adjusting the pH with HCl or NaOH. For the washing and hybridisation buffer, PBS including 750 mM NaCl was used. MES buffer (50 mM, pH 6.5) was prepared by MES adjusting the pH with NaOH. EDC/NHS solution (200 mM EDC and 50 mM NHS) and alanine solution (4 mg/mL) was prepared with 50 mM MES buffer. All solutions were prepared with Milli-Q quality water (18 MΩ cm) to reach the desired concentration levels. 2.3. Instrumentation

2. Experimental

The electrophoresis tank was provided by Cleaver Scientific Ltd. (Rugby, United Kingdom). Micro-Bio-Tec Brand (Giessen, Germany) was used as electrophoresis power supply and the agarose gels were monitored using the Kodak Gel Logic 200-Imaging System (New York, NY, USA). After nanoparticle synthesis, absorption spectra were measured using an Agilent 8453 UV–visible spectrophotometer (Agilent Technologies, Inc., Santa Clara, CA, USA). The quantity analysis of RCA and PCR products was measured using Alpha Innotech Corp (San Leandro, CA, USA). DeltaNu Examiner Raman Microscopy system (Deltanu Inc., Laramie, WY, USA) with a 785 nm laser source, a motorised microscope stage sample holder, and a cooled charge-coupled device (CCD, at 0 1C) detector was used to detect target oligonucleotides. The instrument parameters were as follows: 20  objective, 2.5 mm laser spot size, 100 mW laser power, and 20 s acquisition time. Baseline correction was performed for all the measurements. Transmission electron microscopy (TEM) images were captured with the Tecnai G2 F30 instrument (FEI Company, Hillsboro, OR, USA) at 120 kV.

2.1. Reagents and materials

2.4. RCA optimisation

Hydrogen tetrachloroaurate (HAuCl4), hexadecyltrimethyl ammonium bromide (CTAB), L-ascorbic acid (AA), 11-mercaptoundecanoic acid (11-MUA), 98% ethanolamine, N-(3-Dimethylaminopropyl)-N0 ethylcarbodiimide hydrochloride (EDC), iron (III) chloride (FeCl3), 2(N-morpholino)ethanesulphonic acid (MES) monohydrate, glacial acetic acid and agarose were obtained from Sigma-Aldrich (Steinheim, Germany). Silver nitrate (AgNO3), sodium borohydride (NaBH4), iron (II) sulphate heptahydrate (FeSO4 U7H2O), perchloric acid (HClO4) and absolute ethanol were purchased from Merck (Darmstadt, Germany). DTNB was obtained from Acros (Morris Traditionals, NJ, USA). NHydroxysulphosuccinimide sodium salt (NHS) was purchased from Pierce Biotechnology (Bonn, Germany). Tris(hydroxymethyl)aminomethane (Tris) and ethylenediaminetetraacetic acid (EDTA) disodium salt dihydrate were obtained from Amresco (Solon, OH, USA). Sodium chloride (NaCl), disodium phosphate (Na2HPO4), and potassium dihydrogen phosphate (KH2PO4) were obtained from J.T. Baker (Deventer, The Netherlands). Ethidium bromide was purchased from AppliChem (Darmstadt, Germany). T4 DNA ligase, 10X T4 DNA ligase buffer, ϕ29 DNA polymerase, 10X ϕ29 DNA polymerase buffer, bovine serum albumin (BSA), deoxyribonucleotide triphosphates (dNTPs), λ EcoRIþHindIII and GeneRuler Ultra-Low Range DNA ladders were provided by Thermo Fisher Scientific (Waltham, MA USA).

Primer, RCA template, ϕ29 DNA polymerase, dNTP concentrations and amplification times were optimised; the results obtained were evaluated using an electrophoresis system. In the first step, primer and RCA template concentrations were optimised. For this purpose, different concentrations of primer (1 and 2 mM) and RCA template (0.1 and 0.2 mM) were hybridised at 67 1C for 5 min. After cooled to room temperature, ϕ29 DNA polymerase (2.0 U/mL) and dNTP (200 mM) were added into the hybridised complex. The reaction was carried out at 30 1C for 2 h. Second, the effect of increasing ϕ29 DNA polymerase (0.83, 1.67 and 2.50 U/mL) and dNTP (83, 153, 214 and 262 mM) concentrations for different amplification time periods (1, 2 and 3 h) was optimised following the hybridisation of the primer (1 mM) and RCA template (0.1 mM) at 67 1C for 5 min. In the final step, appropriate combinations of dNTP concentration and ϕ29 DNA polymerase concentration was experimented. To this end, different concentrations of dNTPs (100, 200 and 400 mM) and ϕ29 DNA polymerase (1 and 2 U/mL) were used and incubated at 30 1C for 2 h. All optimisation samples were run in an agarose gel (1%) and monitored with a UV-imaging system. The quantity of RCA amplification products were also verified in different incubation time periods (0, 60, 120, and 180 min). For different time periods, the quantity of RCA amplification product was measured using a small volume UV spectrometer and an electrophoresis system.

2.2. Oligonucleotides and buffers All oligonucleotides were purchased from Alpha DNA (Quebec, Canada). The sequences are shown in Table 1. The RCA template was circularised from Aptagen LLC (Jacobus, PA, USA). EDTA buffer (500 mM, pH 8) was prepared using EDTA and adjusting the pH with HCl. TAE buffer (20X) was prepared using tris base, glacial acetic acid, EDTA buffer (500 mM, pH 8). It was diluted with deionised water to 1X in order to prepare agarose

2.5. Fabrication and self-assembled monolayer (SAM) structure formation of nanoparticles Au-sphere magnetic [39] and Au-rod nanoparticles [40] were fabricated by using seed-mediated growth techniques. Aumagnetic nanospheres were used as part of the sandwich structure for TEM measurements. In order to form SAM, the surfaces of

70

B. Guven et al. / Talanta 136 (2015) 68–74

Table 1 Sequences of the oligonucleotides used in the study. Oligonucleotide

Sequence

Probe 1 Probe 2 Target Nonsense

50 -NH2-AAA AAT CGG CAG AGG CAT-30 50 -CGA TGG CCT TTC CAA AAA-NH2-30 50 -GGA AAG GCC ATC GTT GAA GAT GCC TCT GCC GA-30 50 -CCC CAA ATT TGG TCA AAG GGC CTT TT A ACC CC-30 50 -PO4-CTT CAA CGA TGG CCT TTC CGG TGC TTA GTC-30 50 -PO4-TGT CTT CGC CTT GTT TCC GAT GGC CTT TCC TTC TTC CTT TCT TTC TTT CGA CTA AGC ACC-30 50 -TTG CTT TGA AGA CGT GGT TG-30 50 -GAT TCC ATT GCC CAG CTA TC-30 50 -TCG GAT TCC ATT GCC CAG CTA TCT GTC ACT TTA TTG TGA AGA TAG TGG AAA AGG AAG GTG GCT CCT ACA AAT GCC ATC ATT GCG ATA AAG GAA AGG CCA TCG TTG AAG ATG CCT CTG CCG ACA GTG GTC CCA AAG ATG GAC CCC CAC CCA CGA GGA GCA TCG TGG AAA AAG AAG ACG TTC CAA CCA CGT CTT CAA AGC AA-30

Primer RCA Forward Primer Reverse Primer Bt-176 maize DNA template

the Au-sphere magnetic nanoparticles were modified by means of leaving them in absolute ethanol containing 20 mM 11-MUA overnight. The nanoparticles were collected by using a permanent magnet and washed with 50 mM MES buffer. Au-nanorods were used as a Raman label by assembling DTNB (20 mM) in absolute ethanol to form free carboxyl groups on the surface of the nanoparticle for 18 h at room temperature. Then, the gold nanorods were washed with 50 mM MES buffer and collected by centrifugation (13,500 rpm, 6 min).

2 and gold slide-probe 1 complexes were hybridised with target sequence. The integration time was 20 s and all experiments were performed in triplicate. SERS spectra corresponding to each concentration of target sequence were recorded. The band intensity at 1327 cm  1 versus the target concentration calibration curve was plotted and the linearity range and coefficient of determination (R2), LOD and LOQ values were calculated in accordance with IUPAC definition [42].

2.6. Traditional and RCA sandwich assay designs

2.8. Comparison of TEM measurements between traditional and RCA sandwich assays

Based on our previous studies, the surface of the gold slide modification, activation, immobilisation, and non-specific binding prevention were performed with 11-MUA, EDC/NHS, probe 1 and alanine, respectively [19,41]. At the first hybridisation step of the traditional sandwich assay, different concentrations of target solutions (100 fM to 100 nM) were added directly to the Au-slide and hybridised at 60 1C for 15 min. Then, Au-slides were washed with PBS buffer. At the first hybridisation step of the RCA sandwich assay, 10 mL solutions containing 1 mM primer probe, 1X T4 DNA ligase buffer, and 1 mL target sequences (10 fM to 100 nM) were added to the Au-slide. The ligation mixture was incubated at 65 1C for 10 min and then slowly cooled to room temperature which was followed by the addition of T4 DNA ligase (1 U/mL) and BSA (200 mg/mL). After incubation at 37 1C up to 2 h, the Au-slide was washed with PBS buffer. After DNA ligation, 10 mL solutions containing 0.1 mM RCA template and 1X ϕDNA polymerase buffer were added to the Auslide at 67 1C for 3 min. After cooling to room temperature, ϕDNA polymerase (1 U/mL), dNTPs (200 mM for each dATP, dCTP, dGTP and dTTP), and 200 mg/mL BSA were added and incubated at 30 1C for 2 h. After the RCA step, the Au-slide was washed with PBS buffer. As mentioned above, Au-sphere magnetic nanoparticles and Au-slide were used for the same purpose; therefore, the same protocol was applied for magnetic nanospheres. Similar to the Au-slide modification, surface activation of the Au-rod nanoparticles, probe immobilisation and then blocking of the particle surfaces were performed. After each step, the washing procedure was applied twice using MES buffer. Following the first hybridisation step of the traditional sandwich assay and RCA procedure of the RCA sandwich assay, DTNB-labelled and probe 2-modified Au-rod nanoparticles were added to the modified Auslide. The second hybridisation was performed at 60 1C for 15 min, and the surface was washed with PBS buffer six times. 2.7. Detection of target sequence using SERS For both sandwich assays, SERS spectra of DTNB labels were obtained after DTNB-labelled rod-shaped gold nanoparticle-probe

TEM samples were prepared by pipetting 10 mL of nanoparticle solution onto formvar–carbon coated cupper grids and left to dry for 10 min. TEM images were determined to distinguish between traditional and RCA sandwich assay platforms. Au-sphere magnetic nanoparticles were used as a substitute for Au-slide platform for TEM measurements and thus, sandwich structures were formed between Au-sphere magnetic and Au-rod nanoparticles.

2.9. Control studies of RCA–SERS sandwich assay The nonsense sequence (100 nM) was used to evaluate selectivity, and the results were compared with 0 nM (blank) and 100 nM target sequence concentrations. Different concentrations of 200 base pair (bp) PCR products were used to show the specificity of the developed RCA–SERS assay. The accuracy of the developed RCA sandwich method was assessed with real sample studies. Accordingly, 200 bp of the 35S promoter region of the Bt-176 maize genome was amplified using PCR. PCR amplification was carried out in 50 mL 1X PCR master mixes. The PCR mix was prepared with 1X PCR master mix (MgCl2 (2 mM), dNTPs (0.2 mM), Taq DNA polymerase (0.025 U/mL)), Bt-176 maize DNA template (2 mL), and forward and reverse primers (0.5 mM). Amplification was achieved through an initial denaturation step at 95 1C for 4 min followed by a 50 cycle process that included a denaturation step for 1 min at 95 1C, an annealing step for 1 min at 62 1C and an extension step for 1 min at 72 1C; a final extension at 72 1C for 10 min. The amount of PCR product was measured using a small volume UV spectrometer and the proper dilution was performed. Then, the amount of diluted PCR product was analysed using the developed RCA–SERS assay in order to check the accuracy of the developed assay method. The SERS intensities of target and PCR product sequence concentrations were calculated using calibration graphs, and the results were compared. Experiments were performed in triplicate, average and standard deviations were calculated, and data analysis was performed using Microsoft Excel and OriginPro 7.5.

B. Guven et al. / Talanta 136 (2015) 68–74

71

Fig. 1. Schematic illustration of (A) traditional and (B) RCA based sandwich assays for target oligonucleotide sequence.

3. Results and discussion

preparation, immobilisation, hybridisation and selective detection of the target are shown in Fig. 1A and B.

3.1. Fabrication of the nanoparticles 3.3. RCA optimisation In this study, Au-magnetic sphere nanoparticles were used as the substitute for Au-slide platform for TEM measurements and DTNB-labelled Au-rod nanoparticles were used as the SERS reporter due to the strong SERS signals, based on the research of Temur et al. [40]. While Au-magnetic sphere nanoparticles have a plasmon band at 521 nm, Au-rod nanoparticles have transverse plasmon and longitudinal plasmon bands at 524 nm and 665 nm, respectively. The TEM images of Au-magnetic sphere and Au-rod nanoparticles were given in Fig. S1. The effect of the nanoparticles' geometric shapes on SERS intensity was also reported, and sharpedged nanoparticles were verified to enhance SERS intensity and to act as electromagnetic hot spots for SERS. Rod-shaped Aunanoparticles act as electromagnetic hot spots for SERS owing to their sharp corners and edges [43].

3.2. Traditional and RCA sandwich assay designs The schematic illustration of the stepwise traditional and RCA sandwich assay design processes of the gold slide and nanoparticle

Agarose gel electrophoresis was used to assess the replication product. Lambda EcoRI þHindIII (125–21,216 bp) and GeneRuler Ultra-Low Range DNA (10–300 bp) ladders were used to evaluate the amplification inputs and outputs. The optimisation parameters for RCA amplification were investigated by agarose gel electrophoresis and monitored with a UV-imaging system. The optimum primer and RCA concentrations were found to be 1 mM and 0.1 mM, respectively (Fig. S2). The optimum ϕ29 DNA polymerase (1 U/mL) and dNTP concentrations (100 mM) were determined for different amplification time periods (Fig. S3) and at 30 1C for 2 h (Fig. S4). By using the RCA parameters (0.1 mM RCA template, 1 mM primer, 1 U/mL ϕ29 DNA polymerase, 100 mM dNTP concentrations) already optimised before, the quantity of RCA amplification products were also determined in different incubation time periods (0, 60, 120, and 180 min). The optimum quantity of RCA amplification product was measured using a small volume UV spectrometer (Fig. 2A) and an electrophoresis system (Fig. 2B) and determined as 120 min according to the amplified amount and agarose band intensity.

72

B. Guven et al. / Talanta 136 (2015) 68–74

Amplified RCA product (ng/µL)

A 3000 2800 2600 2400 2200 2000

0

50

100

150

200

Time (min)

B 1

2

3

4

No product

5

6

RCA product bands

Residue reactants

Fig. 2. Verification of RCA amplification with (A) UV spectrometer (B) agarose gel electrophoresis ((1) GeneRuler Ultra-Low Range, (2) λ EcoRI þHindIII, (3) 0 min, (4) 60 min, (5) 120 min, (6) 180 min).

3.4. Comparison of sensitivities between traditional and RCA sandwich assays The SERS spectra for the developed traditional and RCA sandwich assays were obtained by measuring the DTNB signal. The typical DTNB bands (the symmetric nitro stretch (vs. (NO2)) at 1330 cm1) in the presence of the target oligonucleotide sequence were obtained in both sandwich assay systems. The detection was based on the characteristic features of DTNB response and then quantified depending on the changes in the target concentration by intensity. The SERS spectra of different concentrations of target molecules are given in Fig. 3A and B for traditional and RCA sandwich assays, respectively. As shown in Fig. 4A and B, the calibration curves were plotted with the changes in the band intensities of DTNB vs. the concentration of target sequence (R2 ¼0.999 for traditional assay and R2 ¼0.983 for RCA assay). The calibration curve of the developed traditional sandwich assay is between 1 pM and 10 nM, with a detection limit of 0.1 pM, and the calibration curve of the RCA sandwich assay is between 100 fM and 100 nM, with a detection limit of 6.3 fM. 3.5. Comparison of TEM measurements between traditional and RCA sandwich assays Fig. 5 displays TEM images of both sandwich assay platforms. The intense sandwich structures of RCA sandwich assay were distinguished from the base and simple sandwich structure of traditional sandwich assay by means of using TEM image (Fig. 5A). Besides, the amplified and branched structure of RCA strategy was

Fig. 3. (A) Symmetric NO2 stretching bands of DTNB range from 100 fM to 100 nM target concentration in traditional SERS sandwich assay (a) 100 nM, (b) 10 nM, (c) 1 nM, (d) 100 pM, (e) 10 pM (f) 1 pM, (g) 100 fM, target concentration, (B) symmetric NO2 stretching bands of DTNB range from 10 fM to 100 nM target concentration in RCA– SERS sandwich assay (a) 100 nM, (b) 10 nM, (c) 1 nM, (d) 100 pM, (e) 10 pM, (f) 1 pM (g) 100 fM, (h) 10 fM target concentration.

determined in RCA sandwich assay (Fig. 5B). The TEM images' results show that the RCA sandwich method was functional for amplification of desired nucleotide sequences, and the amount of bounded DTNB labelled part was much higher than traditional sandwich assay. 3.6. Control studies of RCA–SERS sandwich assay In the presence of the higher sensitivity of the developed RCA amplification sandwich assay, it was shown that amplification protocol had become functional; therefore, the control parameters (selectivity and specificity) of this system were studied. The nonsense oligonucleotide sequence was used to show the selectivity of the RCA sandwich assay; the relative SERS intensities were given in Fig. 6. The obtained results were compared with target and blank trials at the same concentration, and it was found out that the SERS intensities of nonsense sequence and blank control were lower than the target sequence intensity at the same concentration. The specificity of the developed RCA assay was studied with 200 bp of the 35S promoter of the Bt-176 maize gene using PCR amplification. In order to observe the variation in intensity,

B. Guven et al. / Talanta 136 (2015) 68–74

different concentrations of PCR products were obtained and analysed with the small volume UV spectrometer and the developed RCA assay. By using the UV spectrometer, the concentrations were found to be 0.8 7 0.3 nM and 6.3 7 2.1 fM. The concentrations measured by the developed RCA–SERS assay were

Fig. 4. (A) Calibration curve for target oligonucleotide sequence in range of 1 pM to 10 nM in traditional sandwich assay, (B) calibration curve for target oligonucleotide sequence in range of 100 fM to 100 nM in RCA sandwich assay using gold nanoparticles.

73

1.0 7 0.3 nM and 5.0 7 0.3 fM. As a consequence, similar results between the small volume UV spectrometer and the developed RCA assay showed that the proposed method could be applied to real samples. Our results indicated that under isothermal conditions, a specific oligonucleotide sequence can be detected relatively more sensitively and selectively in the RCA–SERS sandwich method than PCR-based methods [44,45], which need thermal cyclers and specific instruments. In addition, other traditional SERS-based sandwich methods, especially the detection limit of these studies were evaluated, including 0.1 nM [46], 10 nM [47], 1 pM [13], 10 pM [45], 45 pM [48], 77 pM [49], and our developed RCAbased method was found to have higher sensitivity. The detection limit of the developed SERS–RCA sandwich assay was found to be 6.3 fM. It was achieved as 10 fM in chemiluminescence-RCA [50], 2.8 aM in electrochemical-RCA [24], 9 pM in bead-based padlockRCA [51], and 100 fM in fluorescence-RCA [52]. When the detection limit of various methods was compared with our developed method, it was seen that our developed SERS–RCA sandwich assay is more sensitive. RCA-based methods have gained importance due to their speed and reduced cost. The amplified products can be visualised by gel electrophoresis, but this method is also suitable for gel-free systems, such as fluorescence, and it can also be used for the identification and quantification of detected samples.

Fig. 6. The selectivity control of developed RCA–SERS assay with 100 nM concentration of nonsense sequence.

Fig. 5. TEM images of (A) traditional and (B) RCA sandwich assay platforms.

74

B. Guven et al. / Talanta 136 (2015) 68–74

4. Conclusion We developed a rapid, selective and sensitive assay for enumeration of the target oligonucleotide in Bt-176 maize sample by using the combination of RCA and SERS methods. The results obtained from the RCA–SERS assay in this study indicated that RCA amplification and SERS can be used for the determination of target oligonucleotides. We also focused on the different concentrations of the PCR-amplified 200 bp of the 35S promoter of the Bt-176 maize gene sequence and compared the SERS intensity results using calibration curves. It was found that the target oligonucleotide was detected in a real matrix in a rapid, reliable and selective manner. It is, therefore, pointed out that the developed RCA–SERS assay can be used for the selective, sensitive and specific detection of target oligonucleotide sequences. Acknowledgements This study was supported by The Scientific and Technological Research Council of Turkey; Project number: 111T096. We also thank to Assoc. Prof. Dr Demet Cetin and Prof. Dr. Zekiye Suludere for TEM measurements. Appendix A. Supporting information Supplementary data associated with this article can be found in the online version at http://dx.doi.org/10.1016/j.talanta.2014.11. 051. References [1] J.A. Hu, C.Y. Zhang, Anal. Chem. 82 (2010) 8991–8997. [2] M. Lipp, P. Brodmann, K. Pietsch, J. Pauwels, E. Anklam, J. AOAC Int. 82 (1999) 923–928. [3] H.J. Brunnert, F. Spener, T. Borchers, Eur. Food Res. Technol. 213 (2001) 366–371. [4] S. Shanmukh, L. Jones, J. Driskell, Y.P. Zhao, R. Dluhy, R.A. Tripp, Nano Lett. 6 (2006) 2630–2636. [5] S. Boyd, M.F. Bertino, D.X. Ye, L.S. White, S.J. Seashols, J. Forensic Sci. 58 (2013) 753–756. [6] P. Etchegoin, R.C. Maher, L.F. Cohen, H. Hartigan, R.J.C. Brown, M.J.T. Milton, J.C. Gallop, Chem. Phys. Lett. 375 (2003) 84–90. [7] M. Vendrell, K.K. Maiti, K. Dhaliwal, Y.T. Chang, Trends Biotechnol. 31 (2013) 249–257. [8] H. Torul, H. Ciftci, F.C. Dudak, Y. Adiguzel, H. Kulah, I.H. Boyaci, U. Tamer, Anal. Methods 6 (2014) 5097–5104. [9] J. Baniukevic, I.H. Boyaci, A.G. Bozkurt, U. Tamer, A. Ramanavicius, A. Ramanaviciene, Biosens. Bioelectron. 43 (2013) 281–288. [10] N.N. Yazgan, I.H. Boyaci, E. Temur, U. Tamer, A. Topcu, Talanta 82 (2010) 631–639. [11] B. Guven, N. Basaran-Akgul, E. Temur, U. Tamer, I.H. Boyaci, Analyst 136 (2011) 740–748. [12] U. Tamer, A. Onay, H. Ciftci, A.G. Bozkurt, D. Cetin, Z. Suludere, I.H. Boyaci, P. Daniel, F. Lagarde, N. Yaacoub, J.M. Greneche, J. Nanopart. Res. 16 (2014) 2624. [13] C. Fang, A. Agarwal, K.D. Buddharaju, N.M. Khalid, S.M. Salim, E. Widjaja, M.V. Garland, N. Balasubramanian, D.L. Kwong, Biosens. Bioelectron. 24 (2008) 216–221.

[14] K. Gracie, E. Correa, S. Mabbott, J.A. Dougan, D. Graham, R. Goodacre, K. Faulds, Chem. Sci. 5 (2014) 1030–1040. [15] B. Guven, F.C. Dudak, I.H. Boyaci, U. Tamer, M. Ozsoz, Analyst 139 (2014) 1141–1147. [16] S.S.R. Dasary, A.K. Singh, D. Senapati, H.T. Yu, P.C. Ray, J. Am. Chem. Soc. 131 (2009) 13806–13812. [17] H.T. Temiz, I.H. Boyaci, I. Grabchev, U. Tamer, Spectrochim. Acta A 116 (2013) 339–347. [18] W. Xie, B. Walkenfort, S. Schlucker, J. Am. Chem. Soc. 135 (2013) 1657–1660. [19] J. Kneipp, H. Kneipp, A. Rajadurai, R.W. Redmond, K. Kneipp, J. Raman Spectrosc. 40 (2009) 1–5. [20] Z.Y. Wang, S.F. Zong, H. Chen, H. Wu, Y.P. Cui, Talanta 86 (2011) 170–177. [21] C.M. MacLaughlin, N. Mullaithilaga, G.S. Yang, S.Y. Ip, C. Wang, G.C. Walker, Langmuir 29 (2013) 1908–1919. [22] M.E. Pekdemir, D. Erturkan, H. Kulah, I.H. Boyaci, C. Ozgen, U. Tamer, Analyst 137 (2012) 4834–4840. [23] P. Tong, W.W. Zhao, L. Zhang, J.J. Xu, H.Y. Chen, Biosens. Bioelectron. 33 (2012) 146–151. [24] C.F. Ding, N.N. Wang, J. Zhang, Z.F. Wang, Biosens. Bioelectron. 42 (2013) 486–491. [25] C.Q. Lin, Y.X. Zhang, X. Zhou, B. Yao, Q. Fang, Biosens. Bioelectron. 47 (2013) 515–519. [26] G. Nallur, C.H. Luo, L.H. Fang, S. Cooley, V. Dave, J. Lambert, K. Kukanskis, S. Kingsmore, R. Lasken, B. Schweitzer, Nucleic Acids Res. 29 (2001) e118. [27] S.P. Jonstrup, J. Koch, J. Kjems, RNA 12 (2006) 1747–1752. [28] B. Schweitzer, S. Roberts, B. Grimwade, W.P. Shao, M.J. Wang, Q. Fu, Q.P. Shu, I. Laroche, Z.M. Zhou, V.T. Tchernev, J. Christiansen, M. Velleca, S.F. Kingsmore, Nat. Biotechnol. 20 (2002) 359–365. [29] D. Liu, Y.Y. Fan, J. Li, X.G. Gao, M. Hao, H.T. Xue, G.H. Tai, Mol. Med. Rep. 6 (2012) 220–226. [30] M. Nilsson, M. Gullberg, F. Dahl, K. Szuhai, A.K. Raap, Nucleic Acids Res. 30 (2002) e66. [31] B. Schweitzer, S. Wiltshire, J. Lambert, S. O’Malley, K. Kukanskis, Z.R. Zhu, S.F. Kingsmore, P.M. Lizardi, D.C. Ward, Proc. Natl. Acad. Sci. USA 97 (2000) 10113–10119. [32] A.T. Christian, M.S. Pattee, C.M. Attix, B.E. Reed, K.J. Sorensen, J.D. Tucker, Proc. Natl. Acad. Sci. USA 98 (2001) 14238–14243. [33] S.B. Zhang, Z.S. Wu, G.L. Shen, R.Q. Yu, Biosens. Bioelectron. 24 (2009) 3201–3207. [34] R.M. Anthony, T.J. Brown, G.L. French, J. Clin. Microbiol. 38 (2000) 781–788. [35] A. Rector, G.D. Bossart, S.J. Ghim, J.P. Sundberg, A.B. Jenson, M. Van Ranst, J. Virol. 78 (2004) 12698–12702. [36] Y.Y. Huang, H.Y. Hsu, C.J.C. Huang, Biosens. Bioelectron. 22 (2007) 980–985. [37] L. Zhou, L.J. Ou, X. Chu, G.L. Shen, R.Q. Yu, Anal. Chem. 79 (2007) 7492–7500. [38] J. Yan, S. Su, S.J. He, Y. He, B. Zhao, D.F. Wang, H.L. Zhang, Q. Huang, S.P. Song, C. Fan, Anal. Chem. 84 (2012) 9139–9145. [39] U. Tamer, Y. Gundogdu, I.H. Boyaci, K. Pekmez, J. Nanopart. Res. 12 (2010) 1187–1196. [40] E. Temur, I.H. Boyaci, U. Tamer, H. Unsal, N. Aydogan, Anal. Bioanal. Chem. 397 (2010) 1595–1604. [41] B. Guven, I.H. Boyaci, U. Tamer, P. Calik, Analyst 137 (2012) 202–208. [42] G.L. Long, J.D. Winefordner, Anal. Chem. 55 (1983) A712–A724. [43] Y.L. Wang, K. Lee, J. Irudayaraj, Chem. Commun. 46 (2010) 613–615. [44] K.K. Strelau, A. Brinker, C. Schnee, K. Weber, R. Moller, J. Popp, J. Raman Spectrosc. 42 (2011) 243–250. [45] T. Kang, S.M. Yoo, I. Yoon, S.Y. Lee, B. Kim, Nano Lett. 10 (2010) 1189–1193. [46] Y. Li, Y. Zeng, Y.N. Mao, C.C. Lei, S.S. Zhang, Biosens. Bioelectron. 51 (2014) 304–309. [47] S.E.J. Bell, N.M.S. Sirimuthu, J. Am. Chem. Soc. 128 (2006) 15580–15581. [48] F.L. Gao, J.P. Lei, H.X. Ju, Anal. Chem. 85 (2013) 11788–11793. [49] J.A. Dougan, D. MacRae, D. Graham, K. Faulds, Chem. Commun. 47 (2011) 4649–4651. [50] W. Cao, Y. Ma, S. Qiao, P.X. Gong, S.Z. Chen, J.H. Yang, Anal. Lett. 44 (2011) 105–116. [51] K. Sato, R. Ishii, N. Sasaki, K. Sato, M. Nilsson, Anal. Biochem. 437 (2013) 43–45. [52] Y. Long, X.M. Zhou, D. Xing, Biosens. Bioelectron. 46 (2013) 102–107.