A flow injection analysis coupled dual electrochemical detector for selective and simultaneous detection of guanine and adenine

Electrochimica Acta 123 (2014) 485–493

Contents lists available at ScienceDirect

Electrochimica Acta journal homepage: www.elsevier.com/locate/electacta

A flow injection analysis coupled dual electrochemical detector for selective and simultaneous detection of guanine and adenine Rajendiran Thangaraj a , Subramanian Nellaiappan a , Raja Sudhakaran b , Annamalai Senthil Kumar a,∗ a b

Environmental and Analytical Chemistry Division, School of Advanced Sciences, Vellore Institute of Technology University, Vellore-632 014, India Aquaculture Biotechnology Laboratory, School of Bioscience and Technology, Vellore Institute of Technology University, Vellore-632 014, India

a r t i c l e

i n f o

Article history: Received 1 November 2013 Received in revised form 9 January 2014 Accepted 10 January 2014 Available online 23 January 2014 Key words: Flow injection analysis Dual electrochemical detector Bipotentiostat Guanine Adenine.

a b s t r a c t Adenine (A) and guanine (G), important bases of nucleic acids, are often analyzed by separation coupled spectroscopic detection methods. Herein, we are demonstrated a new flow-injection analysis (FIA) coupled dual electrochemical detector (DECD), where a chitosan-carbon nanofiber (Chit-CNF) modified glassy carbon electrode prepared by a simple technique and pH 7 phosphate buffer solution as a carrier system, for separation-less quantification of G and A. This method is highly selective and no interference by the presence of the other DNA bases (Thymine and Cytosine). The FIA-DECD was operated at two different operating potentials, E1 = 0.80 V and E2 = 0.95 V vs Ag/AgCl, where G and {G + A} get oxidized, respectively. Amount of A was calculated from the difference between the FIA current signals, measured at E20.95V and E10.80V . The GCE/Chit-CNF was characterized by cyclic voltammetry with potassium ferricyanide system and Raman spectroscopy. The modified electrode showed unique electron-transfer feature with metal like conductivity. Under an optimal condition, FIA-DECD showed linear calibration plots for G and A in a concentration range, 200 nM—50 ␮M with current sensitivity values 13.83 ± 0.48 and 4.84 ± 0.11 nA ␮M−1 respectively. Calculated detection limit (signal-to-noise ratio = 3) values were 46.8 nM and 73.8 nM for G and A respectively. Applicability of the present technique was further demonstrated by detecting G and A in beef kidney sample and DNA hybridization process. © 2014 Elsevier Ltd. All rights reserved.

1. Introduction Adenine (A) and guanine (G) are the building blocks of both deoxyribonucleic acid (DNA) and ribonucleic acid (RNA) that plays a crucial role in protein biosynthesis and the storage of genetic information [1]. Selective and sensitive detection of the purine bases provides valuable insights in fundamental fields such as understanding of DNA sequence, oxidative damage and hybridization and protein metabolism in cells, protein-DNA interactions, etc [2–4]. Commonly used quantification technique for G and A are separation coupled spectroscopic methods. For instance, ion-pair reversed phase high performance liquid chromatography and capillary electrophoresis coupled UV [1,5,6], micellar electrokinetic chromatography with indirect laser-induced fluorescence detection (ILIFD) [7] and high performance liquid chromatography coupled mass spectrophotometer (HPLC-MS) [8] techniques were reported for the detection of G and A. Each method has its own

∗ Corresponding author. Tel.: +91 416 2202754; fax: +91 416 2243092. E-mail address: [email protected] (A.S. Kumar). 0013-4686/$ – see front matter © 2014 Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.electacta.2014.01.066

strengths and weakness in terms of analytical performance. UV based detectors are versatile; however owing to low extension coefficient, sensitivity of the signals are very low. Similarly, the ILIFD and mass spectroscopy based detection technique allow to detect low concentration of purine bases; beside with respect to the instrumentation cost, off-line sampling preparation, run-time and skilled person requirement, the above techniques are not suitable for routine analytical measurements. Hence, it is highly challenging research to develop a new technique which full fills all majority of the above mentioned criteria. Herein, we introduce a dual electrochemical detector (DECD) based flow injection analysis technique (FIA-DECD) for rapid and simultaneous detection of G and A without any derivatization and separation procedure. Electro-analytical techniques offer simple, less-expensive, highly sensitive and selective analytical approach extendable to disposable type screen printed electrode and miniaturization. In the past, there were several electrochemical methods (cyclic and pulse voltammetric techniques), in which various chemically modified electrodes (CMEs) as working systems were reported for simultaneous detection of G and A with the test sample volume about 10 mL. Following are the representative CMEs:

486

R. Thangaraj et al. / Electrochimica Acta 123 (2014) 485–493

hybridization were successfully demonstrated. Note that CNF has cylindrical nanostructure with different stacking arrangements of graphene sheets such as stacked platelets, ribbon or herringbone [29]. The mechanical strength and electrical properties of CNF are closer to that of the MWCNT [30]. The primary distinguishable character of CNF from MWCNT is the stacking of graphene sheets of varying shapes, producing more edge sites on the outer wall of the CNF than the MWCNT. Such edge plane defects may facilitate the electron-transfer of electro-active analytes [31,32]. In further we found in this work that CNF has relatively less adsorption of aromatic organic compounds than the MWCNTs. Such property is highly desirable for easy reproducible working electrode surface in FIA-ECD. Scheme 1. Illustration for the flow injection analysis coupled with dual electrode system (FIA-DECD) for simultaneous detection of guanine and {guanine + adenine} at two different operating potentials, E10.80V and E20.95V . Adenine’s current can be calculated from the current difference between iE2,0.95V and iE1,0.8V .

TiO2 -graphene oxide nanocomposite [9], PbO2 -carboxylated carbon nanotubes (CNT)-ionic liquid composite film [10], CeO2 nano particles decorated multiwalled carbon nanotube (MWCNT) [11], microwave-assisted prepared carbon nanotube/La(OH)3 nanocomposite [12], polythionine/NPAu/MWNTs modified electrode [13], graphene–nafion composite film [14], ␤-cyclodextrin(␤CD)/MWCNTs modified electrode [15], potassium/phenanthrene doped MWCNTs incorporated poly(new fuchsin) composite film (MWCNT-PNF)) [16], and f-MWCNT-gold-hydroxypropyl ␤-CD composite film (f-MWCNT, f = functionalized) (f-MWCNT-␤-CDAu) [17], NiFe2 O4 magnetic nano particles decorated MWCNTs [18], 2,6-pyridinedicarboxylic acid/graphene composite film [19], graphitized mesoporous carbon modified GCE [20], carboxylic acid functionalized graphene modified glassy carbon electrode [21] and TiO2 nano particles-magnesium doped zeolite Y modified carbon paste electrode [22]. Unfortunately, most of the recently reported working electrodes are either expensive or having tedious preparation routes. For instance, f-MWCNT-␤-CD-Au involved with expensive chemicals such as functional carbon nanotube and Au, and the MWCNT-PNF needs over 48 hours time for the electrode preparation [16]. In further, working with 10 mL volume is not practically viable for DNA based real sample analysis; where the samples are expensive and difficulty in getting required volume (10 mL) for the analysis. Alternatively, flow injection analysis (FIA) coupled electrochemical detection (ECD) technique (FIA-ECD; single working electrode) is a suitable one for low volume detection of real sample (5-50 ␮L) with high sensitivity and selectivity. Few reports were available for the detection of either G or A individually [23,24] or inbound DNA [25,26] by FIA-ECD. To the best of our knowledge, simultaneous detection of G and A without any separation technique by FIA-DECD is not yet attempted. Complication due to the co-electro-oxidation of both the purine bases, G and A restricts the technique for the selective analysis [27]. In further, electrochemical oxidation of one of the DNA bases, is often influenced by presence of other bases [28]. In this work, we resolved the problem by taking two discrete working electrodes as electrochemical detectors coupled with a bipotentiostat for simultaneous FIA of G and A. In this technique, G was detected at 0.80 V (E10.80V ), while A along with G was determined at 0.95 V vs Ag/AgCl (E20.95V ) simultaneously. In further by suitably subtracting the E10.80V ’s current signal with E20.95 ’s value one can easily calculate the A’s current contribution (Scheme 1). A new carbon nanofiber (CNF)-chitosan modified glassy carbon electrode (GCE/Chit-CNF) introduced here is a low cost one (ten times lesser price than impure-MWCNT) and can be prepared within 40 ± 2 minutes without any linkers [16,17]. Furthermore, quantification of G and A in beef kidney and DNA

2. Experimental 2.1. Chemicals and Instruments Chitosan, graphitized carbon nanofibers (diameter 100 nm, length 20-200 ␮m), adenine, guanine, single-stranded probe DNA (5 -AACCAGAGTGGTGGATGGAA), complementary probe DNA (5 -TTGGTCTCACCACCTACCTT) and non-complementary DNA (5 GTCGACGAACTTCACTGGGA) were obtained from Sigma-Aldrich. Aqueous solutions were prepared by using deionized alkaline KMnO4 double distilled (DD) water. Unless otherwise stated pH 7 phosphate buffer solution (PBS) of 0.1 mol L−1 ionic strength was used as a supporting electrolyte solution. Electrochemical measurements were carried out using a CHI model 660 C electro-chemical work station, USA with 10 mL working volume. The three electrode system consisted of GCE with 0.0707 cm2 geometrical surface area and its CME as a working electrode, Ag/AgCl with 3 M KCl as a reference electrode, and platinum wire as a counter electrode. The Bio-analytical system (BAS, USA) polishing kit was used to polish the GCE surface. Hydrodynamic amperometric measurements were done using a bipotentiostat instrument (CHI 760D electrochemical workstation, Austin, TX, USA). The FIA system consisted of Hitachi L-2130 pump delivery, a Rehodyne model 7125 sample injection valve (20 ␮L loop) with interconnecting Teflon tube and a conventional electrochemical cell (BAS, USA) [33]. A DECD with two glassy carbon electrodes of similar geometric area (0.0707 cm2 ) placed 2 mm apart purchased from BAS was used as a dual working electrode. FEI Quanta FEG 200 instrument was used for FE-SEM analysis. Raman spectroscopic analysis was performed using AZILTRON, PRO 532, (USA) with 532 nm laser excitation. Following optimal DPV parameters were used for electrochemical measurements: increment = 4 mV; amplitude = 50 mV; pulse width = 0.2 s; pulse period = 0.5 s and sampling width = 0.0167 s. 2.2. Preparation of the chemically modified electrode (GCE/Chit-CNF) Initially, the GCE was cleaned both mechanically (polished with 0.05 micron alumina powder, cleaned with acetone and washed with DD water) and electrochemically (by performing cyclic voltammetry (CV) for 10 cycles in the potential window 0.2 V to 1.0 V vs Ag/AgCl at the potential scan rate (v) of 50 mV s−1 in pH 7 PBS), before each experiment and served as an underlying substrate. For preparation of the GCE/Chit-CNF modified electrode, 1 mg of CNF dispersed in 500 ␮L of 0.1% chitosan solution, sonicated for 5 ± 2 minutes in a water bath at room temperature, was drop coated on the cleaned GCE surface and allowed to air dry for 30 ± 2 minutes at room temperature (25 ± 2 ◦ C) to get near uniform thin layer on GCE. In total, 40 ± 2 min time required for the electrochemical detector preparation.

R. Thangaraj et al. / Electrochimica Acta 123 (2014) 485–493

487

Fig. 1. A Comparative CV responses of GCE/Chit-CNF (a) and unmodified GCE (b) with 5 mM of potassium ferricyaninde in 0.5 M KCl solution at a scan rate 10 mV s−1 , (B) FE-SEM image of Chit-CNF composite and (C) Raman spectroscopic responses of CNF (a) and Chit-CNF (b).

2.3. DNA sample preparation Single-strand probe DNA (p-ssDNA), complementary ssDNA (c-ssDNA), non-complementary DNA (nc-ssDNA) samples were diluted with pH 7 PBS and kept in the refrigerator. For hybridization process, equal concentration of p-ssDNA and c-ssDNA was mixed in a glass vial and kept at 55 ± 2 ◦ C for 10 ± 1 minute. Similar procedure was adopted for the mixing of p-ssDNA and nc-ssDNA. Beef real sample was prepared according to the previous reported literature [20]; where 5 g of the sample was homogenized with 10 mL of PBS using a mortar and pestle, filtered and made upto to 30 mL with PBS for further analysis.

3. Results and Discussion 3.1. GCE/Chit-CNF characterization To study the electronic characteristics, the modified electrode was subjected to cyclic voltammetric investigation with a standard benchmark redox system, ferricyanide/ferrocyanide. Fig. 1(A)a shows typical CV response of GCE/Chit-CNF electrode with 5 mM of potassium ferricyanide at a scan rate of 10 mV s−1 . A well defined redox couple with peak-to-peak potential (Ep = Epa -Epc ) and apparent standard electrode potential (Eo ’ = Epa + Epc /2) values of 70 ± 2 mV and 250 ± 5 mV vs Ag/AgCl respectively, were noticed. For comparison, unmodified GCE is also subjected to the ferri cyanide experiment as in Fig. 1(A)b. Corresponding Ep and Eo values obtained on the GCE are 80 ± 2 and 250 ± 5 mV vs Ag/AgCl.  The near similarity in the Ep and Eo values between GCE and GCE/Chit-CNF indicate appreciable electron-transfer behavior of the GCE/Chit-CNF. Morphology of the Chit-CNF surface was examined by FE-SEM. Fig. 1B shows the SEM image of Chit-CNF composite. Smooth fiberlike structure with variable diameters in a range, 200-480 nm, which is about 2—5 times higher than the size of unmodified fiber system (dia.100 nm), was noticed. Possibly, the chitosan polymer may be wrapped on the CNF and inturn increased the diameter size of the carbon nano fiber. The graphitic micro-structure of CNF and its chemical interaction, if any with chitosan, was probed using Raman spectroscopy. Fig. 1 C showed Raman spectra responses of unmodified CNF (a) and Chit-CNF (b) systems. As can be seen, characteristic peaks corresponding to D-band radial breathing (disorder graphite with sp3 carbon) mode and the G-band tangential stretching mode (graphitic structure with sp2 carbon) were observed at ca. 1250 ± 2 cm−1 and ca. 1570 ± 2 cm−1 , respectively. The ratio of the intensities of the D and G-bands, ID /IG relates degree of graphitic structure disorder (oxygen functionalized graphitic sites) [20]. If

ID /IG > 1, which infers high disorder-ness of the graphitic unit whilst ID /IG < 1 indicates a high crystalline associated with rich graphitic structure. Calculated ID /IG ratio values for the CNF and Chit-CNF are 0.24 and 0.18 respectively. The observation reveals apparent decrement in the oxygen functionalized graphitic sites of CNF after the chitosan modification. Presumably, amino functional groups of chitosan are responsible for the reduction of the oxygen functional sites. Overall the Chit-CNF composite chemically modified electrode showed unique structure and heterogeneous electrontransfer activity. 3.2. Electrochemical oxidations of G and A on GCE/Chit-CNF Fig. 2A shows comparative cyclic voltammograms for the simultaneous detection of G (100 ␮M) and A (100 ␮M) on GCE/Chit-CNF and unmodified GCE electrode with 10 mL pH 7 PBS working system. As can be seen, two well separated irreversible peaks at 0.73 ± 0.002 V and 0.99 ± 0.002 V vs Ag/AgCl, which are due to the simultaneous electrochemical oxidations of G and A, was noticed. As like CV, DPV also showed well-defined and well-separated current signals for G and A oxidation reactions at 0.66 ± 0.002 V and 0.92 ± 0.002 V vs Ag/AgCl, respectively (Fig. 2B). The slight negative shifts observed in the oxidation potentials of the G and A in DPV, when those values compared with the CV response, is due to the applied amplitude and pulse parameters. In contrast, at bare GCE (curve b) no such significant peak response was noticed for G and A oxidations. This observation indicates inability of the unmodified GCE for the electro-analytical oxidation reactions. The voltammetric peak positions for the G and A were confirmed by testing with individual analytes discretely by CV and DPV techniques (Fig. S1). Fig. 3A and B are typical DPV responses of a GCE/Chit-CNF for simultaneous sensing of G and A with respect to one fixed analyte concentration in pH 7 PBS. As can be seen, systematic increase in the peak current responses with an increase in G and A concentration were noticed. Plots of base-line corrected DPV peak current signals (ipa ) against respective G and A concentrations were linear in the range 10—140 ␮M and 10—120 ␮M, respectively. Corresponding current sensitivity/detection limit (signal-to-noise ratio = 3) values are 43.8 ± 0.90 nA ␮M−1 /5.07 ␮M and 40.0 ± 3.30 nA ␮M−1 /4.11 ␮M. The detection limit values for G and A are comparable and lower than some of the previous literature reports [9–22]. To check the stability of the modified electrode, ten successive DPV measurements of 15 ␮M of G was done (data not given). A relative standard deviation (RSD) value 1.42% was observed. Similarly, three numbers of GCE/Chit-CNF prepared discretely at different timings tested for DPV detection of 15 ␮M G in pH 7 PBS was showed a RSD value of 3.82% (data not given). These observations clearly evidence good stability and reproducibility of the present system. The above

488

R. Thangaraj et al. / Electrochimica Acta 123 (2014) 485–493

A.

B. DPV

CV 40

20

15

I/ A

I/ A

30

A

20 a.

10

b.

A 10

G

G

b.

a.

5 0 0 0.2

0.4

0.6

0.8

1.0

0.2

1.2

0.4

0.6

0.8

1.0

1.2

E vs (Ag/AgCl)/V

E vs (Ag/AgCl)/V

Fig. 2. Simultaneous CV (A) and DPV (B) responses of guanine (G, 100 ␮M) and adenine (A, 100 ␮M) on GCE/Chit-CNF (a) and unmodified GCE (b) in pH 7 PBS. CV scan rate = 50 mV s−1 ; DPV parameters: amplitude = 50 mV; increment potential = 4 mV; pulse width = 0.05 s; pulse period = 0.5 s.

preliminary results attribute suitability of the GCE/Chit-CNF for the electro-analytical assays of the G and A. In aim to work with low working volume (␮L), FIA-ECD technique is further adopted. ipa / A

6

3.3. Flow injection analysis 4 2

slope = 0.0438 40

2 A

0

80

[G]variable 10 - 140 M

120

[G] / M

[A]fixed

Initially, interrelated hydrodynamic parameters such as flow rate (Hf ) and applied potential (Eapp ) on discrete FIA of G and A were systematically optimized (Fig. S2) as Hf = 0.8 mL min−1 and Eapp = 0.80 V and 0.95 V vs Ag/AgCl for G and A, respectively. Fig. 4 is a typical FIA-ECD (single electrode system; GCE/Chit-CNF detector) response for increasing concentration of G in a range 200 nM—50 ␮M. FIA calibration plot was linear with a current sensitivity value of 13.83 ± 0.48 nA ␮M−1 . Twelve continuous injections of 200 nM of G yielded a RSD value 3.72% (Fig. 4, inset c), which indicating good stability of the present working electrode. Calculated detection limit (signal-to-noise ratio = 3) value is 46.8 nM.

I

[A]variable

[G] = 200 nM

c.

n = 12

E app = 0.80 V

10 nA

10 - 120 M

50 M

4 800

slope = 0.040

2

0

40

80

b.

600 i / nA

ipa / A

2 A

6

400

slope = 13.83

200

120

[A] / M

0

[G]fixed

0

10 20 30 40 50 [G] / M 200 nM

10 M 100 nA

I / nA

a. 30 20

2 M

0.2

0.4

0.6

0.8

1.0

E vs (Ag/AgCl)/V Fig. 3. Typical DPV responses for the simultaneous detection of guanine and adenine with respect to one fixed analyst on GCE/Chit-CNF in pH 7 PBS. Plots of ipa vs analyte concentration were given as respective inset figures. DPV parameters are same as in the Fig. 2.

250 s

Fig. 4. FIA responses of GCE/Chit-CNF with increasing the concentration of guanine from 200 nM to 50 ␮M with pH 7 PBS as carrier solution. Inset figures are: (a) enlarged view of lower part of the concentration, (b) plot of FIA current vs [G] and (c) twelve successive injections of 200 nM guanine. FIA parameters: Eapp = 0.80 V; Hf = 0.8 mL min−1 .

R. Thangaraj et al. / Electrochimica Acta 123 (2014) 485–493

489

Table 1 Comparison of proposed method with other conventional electrochemical methods for the simultaneous determination of G and A. Sl. No.

Modified electrode

Technique

Analyte

Linear range

Detection limit (␮mol L−1 )

Ref.

1

TiO2 -graphene oxide nanocomposite

DPV

PbO2 -carboxylated carbon nanotubes (CNT)-ionic liquid composite film CeO2 nanoparticles decorated MWCNT

DPV (Individual) DPV DPV

5

Microwave-assisted prepared carbon nanotube/La(OH)3 nanocomposite Polythionine/NPAu/MWCNTs modified electrode

DPV

6

Graphene–Nafion composite film

DPV

7

␤-cyclodextrin)/MWCNTs modified electrode

DPV

8

DPV

10

Potassium/phenanthrene doped MWCNTs incorporated poly(new fuchsin) composite film Functionalized-MWCNT-gold-hydroxypropyl ␤-CD composite film NiFe2 O4 magnetic nano particles decorated MWCNTs

LSV

11

2,6-pyridinedicarboxylic acid/graphene composite film

DPV

12

Graphitized mesoporous carbon modified GCE

DPV

13

Carboxylic acid functionalized Graphene/GCE

DPV

14

TiO2 nanoparticles-magnesium doped zeolite Y modified carbon paste electrode

DPV

0.5–200 ␮M 0.5–200 ␮M 0.07–20 ␮M 0.37–37 ␮M 5–50 ␮M 5–35 ␮M 1–45 ␮M 1–45 ␮M 0.2–1 ␮M 0.2–1 ␮M 2–120 ␮M 5–170 ␮M 100–280 ␮M 4–20 ␮M 0.10–8.5 mM 0.01–3.9 mM 0.59–3.25 mM 0.14–0.57 mM 0.05–3.0 ␮M 0.10–4.0 ␮M 0.05–4.5 ␮M 0.10–6.0 ␮M 25–200 ␮M 25–150 ␮M 0.5–200 ␮M 0.5–200 ␮M 0.1–100 ␮M 0.1–100 ␮M

0.15 0.10 0.006 0.03 0.01 0.02 0.26 0.22 0.01 0.008 0.58 0.75 0.034 0.00075 95.76 7.40 380 240 0.006 0.01 0.01 0.02 0.76 0.63 0.05 0.025 0.013 0.020

[9]

2

G A G A G A G A G A G A G A G A G A G A G A G A G A G A

15

GCE/Chit-CNF

FIA-ECD

G A

0.2–50 ␮M 0.2–50 ␮M

0.046 0.074

This Work

3 4

9

Semi-derivative DPV

[10] [11] [12] [13] [14] [15] [16] [17] [18] [19] [20] [21] [22]

MWCNT: multi-walled carbon nanotube, NPAu: Gold nanoparticle, Chit: Chitosan, CNF: carbon nanofiber, GCE: glassy carbon electrode

Similar to the above, FIA of A was studied on GCE/Chit-CNF discretely at an optimized applied potential, Eapp = 0.95 V vs Ag/AgCl (Fig. 5). The modified electrode showed a linear calibration plot in a concentration range 200 nM—50 ␮M with current sensitivity value 4.84 ± 0.11 nA ␮M−1 . Twelve successive injections of 200 nM of A showed a RSD value 3.14% (Fig. 5, inset b) and calculated detection limit value is 73.8 nM. The linear range and detection limit for G and A by FIA-ECD was compared with other electrochemical techniques in Table 1 [9–22]. As can be seen, although the analytical parameters are comparable with conventional techniques, with respect

[A] = 200 nM

n = 12

Eapp = 0.95 V

10 nA

b.

50 M a.

i / nA

200 100

slope = 4.84

0 0

10 M

50 nA

200 nM

10 20 30 40 50 [A] / M

2 M

250 s

Fig. 5. FIA response of GCE/Chit-CNF with increasing the concentration of adenine from 200 nM to 50 ␮M with pH 7 PBS as carrier solution. Inset is: (a) FIA current vs [A] and (b) twelve successive injections of 200 nM adenine. FIA parameters are: Eapp = 0.95 V; Hf = 0.8 mL min−1 .

to low working volume (20 ␮L) and convenience for real sample analysis, the FIA-DECD is the best choice. Effect of interference of other DNA bases, thymine (T) and cytosine (C) on the two applied potentials were also tested. Fig. 6A is a FIA-ECD of individual injections of G, A, T and C, at an Eapp = 0.80 V vs Ag/AgCl (single electrode mode). Interestingly, except G, none of them showed any current response. Similarly, at 0.95 V vs Ag/AgCl only G and A showed FIA current signal (Fig. 6B); while the T and C resulted to negligible current response. Oxidation of T and C occurred at higher potentials, >1.2 V vs Ag/AgCl [20], is the possible reason for the selective observation. In continuation, FIA of following mixtures; {G + A} (Peak-V, Fig. 6B) and {G + A + T + C} (Peak-VI, Fig. 6B) were also tested. Both of them showed appreciably similar FIA responses. Following can be summarized from the above individual FIA-ECD studies: (i) G can be selectively detected at 0.80 V vs Ag/AgCl with a current sensitivity value of 13.83 ± 0.48 nA ␮M−1 , (ii) A can be detected at 0.95 V vs Ag/AgCl (sensitivity= 4.84 ± 0.11 nA ␮M−1 ), (iii) there is no interference of T and C both at 0.80 and 0.95 V vs Ag/AgCl and (iv) at 0.95 V, G gets oxidized with 26% higher response than that of the value observed at 0.80 V vs Ag/AgCl (Fig. 6A, Peak-I). Fig. 7A and B are FIA-DECD responses of {G + A} mixture (dual electrode/bipotentiostat mode), where the system was operated at different applied potentials, 0.80 V (E10.80V ) and 0.95 V vs Ag/AgCl (E20.95V ). As can be seen, at E10.80V , regular increase in the FIA current signals upon increase in the mixture concentration of {G + A} was noticed. Calculated current sensitivity at E10.80V (SE1,0.80V , S = Sensitivity) is 12.99 ± 0.45 nA ␮M−1 . But at E20.95V , the mixture get electro-oxidized and detected with a current sensitivity value, SE2,0.95V = 19.58 ± 0.42 nA ␮M−1 . Linear range, sensitivity and detection limit values are consolidated in Table 2. Followings are the important observations noticed from the above

490

R. Thangaraj et al. / Electrochimica Acta 123 (2014) 485–493

Table 2 Analytical results for individual and simultaneous determination of guanine and adenine by FIA-ECD and FIA-DECD using GCE/Chit-CNF working electrode. Condition

Eapp /V

1. Individual (FIA-ECD) 2. Bipotentiostat (FIA-DECD)

Analyte

Calibration range (␮mol L−1 )

Linear equation

Detection limit (nmol L−1 )

Injected

Detected

0.80

G

G

0.2-50

iG = 13.83 ± 0.48[G] + 19.12

46.8

0.95 0.80 (E1)

A {G + A}

A G

0.2-50 0.2-50

iA = 4.84 ± 0.11[A] + 9.84 iE1 = 12.99 ± 0.45[G] + 19.75

73.8 49.9

0.95 (E2) E2-E1

{G + A} —

{G + A} Ca.A

0.2-50 0.2-50

iE2 = 19.58 ± 0.42[G + A] + 18.90 iA = 6.58 ± 0.05[A] + 0.805

54.2

result: (i) the current sensitivity obtained with the individual FIA of G (13.83 ± 0.43 nA ␮M−1 , Fig. 4; Single electrode mode) is closely matches (2.9% error) with the value, 12.99 ± 0.45 nA ␮M−1 obtained with the bipotentiostat for the {G + A} mixture at E10.80V (dual electrode; Fig. 7A), (ii) Cumulative current sensitivity values of A (4.84 ± 0.11 nA ␮M−1 ; Single electrode mode) and G (13.83 ± 0.48; Single electrode mode), which is nothing but SA,0.80V + SG,0.95V (18.67 ± 0.59 nA ␮M−1 ) obtained with the individual FIA systems, is nearly equal (4.6% error) to the value of 19.58 ± 0.42 nA ␮M−1 for the mixture system, i.e., {G + A} (S{G+A},0.95V ) detected with the dual electrode at E20.95V and (iii)

A. E1 = 0.80 V

G

over all current sensitivities; SG,0.80V + SA,0.95V (single electrode mode; individuals) = S{G+A},0.95V mixture (bipotentiostat). Following conclusions can be made from the above observations: (i) Analyt responsible for current signal at E10.80V (iE1 ,0.80V ) is G, (ii) Analyts responsible for the current signal (iE2,0.95V ) at E20.95V is G + A. (ii) Subtraction of iE1,0.80V current (iE1 ) from iE2,0.95V (iE2 ) (i.e., iE2,0.95V –iE1,0.80V = iA ; Fig. 7 inset e) gives quantification of A current value (iA ). However, back calculated A’s current value (iA ) by subtraction of iE1,0.80V from iE2 ,0.95V , those obtained under bipotentiostat mode, showed 26% deviation in the current sensitivity over the expected A value measured by single electrode system (entry no 5, Table 2). The deviation may be due to the relatively higher oxidation current value of G at E20.95V over E10.80V (Fig. 6B). Thus, the value, 26% can be taken as a correction factor for the A.

1.0

I/ A

i / nA

600

a.

Bipotentiostat

E1 = 0.80 V

400 slope = 12.994

200 0 0

0.5

50 M

10 20 30 40 50 [G+A] / M

A

T

20 nA

b. 200 nM 200 nA

C

10 M

0.0 2 M

400

600

800

1000 B.

250 s

B. i / nA

{G+A} {G+A+T+C}

c.

d.

E2 = 0.95 V

800 400

slope = 19.58

300

i / nA

1200

E2 = 0.95 V

10 20 30 40 50

0

10 20 30 40 50 ca. A / M

[G+A] / M

I / nA

slope = 6.58

0

0

G

50 M

200 100

0

200

i E2- iE1

e. 10 nA

A

200 nM

10 M 200 nA

100

2 M

T

C 250 s

0 1000

1500

2000

2500

3000

3500

t/s Fig. 6. Interference study on GCE/Chit-CNF at Eapp = 0.80 V (A) and 0.95 V vs Ag/AgCl (B) with pH 7 PBS as a carrier solution. FIA Hf = 0.8 mL min−1 . For the injections, 100 ␮M each of G, A,T and C were injected.

Fig. 7. Typical simultaneous FIA responses of a mixture containing guanine and adenine with concentrations from 200 nM to 50 ␮M at two different applied potentials, E1 = 0.80 V (A) and E2 = 0.95 V vs Ag/AgCl (B) on a GCE/Chit-CNF containing dual electrode detector. Carrier solution is pH 7 PBS. Hf = 0.8 mL min−1 . Inset plots are: (a) i vs. [G + A] at E10.80V , (b) enlarged view of lower part of the concentration at E10.80V , (c) i vs. [G + A] at E20.95 , (d) i vs. ca. [A] by E20.95V -E10.80V and (e) enlarged view of lower part of the concentration at E20.95V .

R. Thangaraj et al. / Electrochimica Acta 123 (2014) 485–493

A.

491

B.

R+S4 E2 = 0.95 V

E1 = 0.80 V

R+S3 R+S2

R+S4 R+S3

R+S1 100 nA

100 nA

R+S2 R+S1

R

R

200

400

600

800

1000

1200

200

400

600

800

1000

1200

t/s

t/s

Fig. 8. Beef kidney real sample (R) analysis using FIA-DECD technique at applied potentials, (A) E1 = 0.80 V and (B) E2 = 0.95 V vs Ag/AgCl by standard addition method. pH 7 PBS as a carrier solution and Hf = 0.8 mL min−1 .

A calculation under bipotentiostat mode. In consideration to the above, following simplified equations are used for G and A current value calculations under the bipotentiostat mode: iG = iE1,0.80V

(1)

iA = (iE2,0.95V − iE1,0.80V ) × (100 − 26)/100

(2)

(iii) Since the mixture current signals are cumulative of the individuals, no other chemical complications such as adsorption of analyts and chemical interaction of G and A, on the electrode surface of FIA-ECD will be occurred in this system. (iv) The dual electrode approach can give wonderful platform for the simultaneous detection of G and A, unlike to the conventional separation coupled techniques [5–8]. In order to validate the assays, a beef kidney sample which contain rich concentration of purine bases is subjected to FIA-DECD. Fig. 8A and B are FIA of a beef kidney real sample solution at a dual electrode (E10.80V and E20.95V ) for G and A quantifications by standard addition approach; where first injection is a real sample solution (R) and other injections are 4 ␮M each of standard mixture of of {G+A} with R (R+S1-4 ). Systematic increase in the current signals both at E10.80V and E20.95V upon increasing concentrations of standard mixture, were noticed. Table 3 provides detailed analytical results for the beef kidney sample analysis. Detected G and A contents are 53.9 and 38.6 ␮g g−1 (corrected with the 26% error). The FIA-ECD technique is further applied to DNA hybridization studies. Note that probing of DNA hybridization is highly demanding subject in clinical diagnosis of several carcinogenic diseases like breast cancer [34], lung cancer [35] and prostate cancer [36], etc. In general, real-time quantitative polymerase chain reaction

(PCR) >method was often used for residual DNA determination. The above method is highly complicated and time consuming. On the other hand, electro-analytical techniques (CV, DPV and electrochemical impedance spectroscopy) in which respective probe ssDNA and target ssDNA suitably modified on a working electrode for selective hybridization reaction, were recently considered as a simple and sensitive method for DNA sensing [37,38]. >Meanwhile, the difference in oxidation current signals of G in ssDNA and dsDNA is properly tuned for specific DNA hybridization detection study [27]. Note that the ssDNA G’s electro-oxidation current value is much higher than that of respective dsDNA system [27]. Possible reasons for the behavior may be due to different structure, either free or hydrogen bonded nature, of G in the DNA samples. First time in this work, we employed FIA-ECD technique to probe the DNA hybridization. FIA applied potential, 0.95 V vs Ag/AgCl (E2) is chosen, where both and G and A get oxidized, as an optimal potential to check the nature of the DNAs. It is expected that the probe ssDNA, which has rich free G and A contents (non-hydrogen bonded), will yield higher FIA current signal than c-ssDNA (which has fewer G and A) and dsDNA (hybridized; hydrogen bonded) systems. As can be seen in Fig. 9, continuous injections of probe ssDNA (a), c-ssDNA (b), nc-ssDNA (c), probe-ssDNA + cssDNA (d) and probe-ssDNA + nc-ssDNA (e) showed different FIA current signals. As expected, probe-ssDNA yielded 5-fold higher current response over the respective hybridized system (i.e., probe-ssDNA + c-ssDNA). Meanwhile, a control sample of probessDNA + nc-ssDNA failed to show any such marked decrement in the FIA current value. These preliminary observations attribute the usefulness of the FIA-ECD technique for selective DNA biosensor studies.

Table 3 FIA-DECD of G and A in a beef kidney real sample using GCE/Chit-CNF working electrodes. Eapp /V

Detected (R)/␮mol L−1

Standard Spiked (R + Sn )/␮mol L−1

Found after standard spiked (␮mol L−1 )

Recovery (%)

Total detection* (␮mol L−1 )

0.80 (E1)

11.85 (G)

24.71 (G + A)

(E2-E1)

12.86 ca. (A)

3.63 8.57 10.60 15.83 6.85 18.09 23.29 31.80 -

90.75 107.13 88.37 98.94 85.63 113.06 97.04 99.38 -

59.25

0.95 (E2)

4 8 12 16 8 16 24 32 -

*Corrected with dilution factor = 5

123.55

64.30

492

R. Thangaraj et al. / Electrochimica Acta 123 (2014) 485–493

a.p-ssDNA

c.nc-ssDNA

1000

1500

100 nA

d.p-ssDNA + c-ssDNA

b.c-ssDNA

500

e.p-ssDNA + nc-ssDNA

2000

t/s Fig. 9. FIA-ECD (single electrode system) responses of p-ssDNA (a), c-ssDNA (b), ncssDNA (c), p-ssDNA + c-ssDNA (d) and p-ssDNA + nc-ssDNA (e) on FIA-GCE/Chit-CNF modified electrode at an applied potential = 0.95 V vs Ag/AgCl, Hf = 0.8 mL min−1 and carrier solution = pH 7 PBS. Inset figures are typical cartoons for the various DNAs.

4. Conclusions A carbon nano fiber-chitosan modified glassy carbon electrode was prepared and used as voltammetric sensor and flow injection analysis dual detector for simultaneous detection of G and A, without any separation technique. Following two potentials; E1 = 0.80 V and E2 = 0.95 V vs Ag/AgCl were chosen as an optimal for the simultaneous FIA-DECD of G and {G + A}. Quantification of A was done by subtracting the E10.80V ’s current signal from E20.95V ’s value. Analytical utility of this technique was further demonstrated by analyzing G and A concentrations in a beef kidney sample and selective DNA hybridization and non-hybridization processes. Acknowledgement The authors are grateful to Department of Science and Technology - Technology System Development (DST-TSD) for the financial support, India. RT thanks the Council of Scientific and Industrial Research (CSIR) for the award of his Senior Research Fellow (SRF). Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/j.electacta. 2014.01.066. References [1] Pilar Garcia del Moral, Maria Jesus Arin, Jose Antonio Resines, Maria Teresa Diez, Simultaneous determination of adenine and guanine in ruminant bacterial pellets by ion-pair HPLC, J. Chromatography B 826 (2005) 257–260. [2] M. Elizabeth, Boon, E. Julia, Salas, K. Jacqueline, Barton, An electrical probe of protein–DNA interactions on DNA-modified surfaces, Nat. Biotechnol. 20 (2002) 282–286. [3] M. Patricia de-los-Santos-Alvarez, Jesus Lobo-Castanf on, J. Arturo, MirandaOrdieres, Paulino Tunf on-Blanco, Electrochemistry of Nucleic Acids at Solid Electrodes and Its Applications, Electroanalysis 16 (2004) 1193–1204. [4] Dai Kato, Mayuri Komoriya, Kohei Nakamoto, Ryoji Kurita, Shigeru Hirono, Osamu Niwa, Electrochemical Determination of Oxidative Damaged DNA with High Sensitivity and Stability Using a Nanocarbon Film, Anal. Sci. 27 (2011) 703–707. [5] D. Brendon, Gill, E. Harvey, Indyk, Development and application of a liquid chromatographic method for analysis of nucleotides and nucleosides in milk and infant formulas, Int. Dairy J. 17 (2007) 596–605.

[6] Ching-Fen Yeh, Shiuh-Jen Jiang, Determination of monophosphate nucleotides by capillary electrophoresis inductively coupled plasma mass spectrometry, Analyst 127 (2002) 1324–1327. [7] Weiping Wang, Lei Zhou, Shumin Wang, Zhi Luo, Zhide Hu, Rapid and simple determination of adenine and guanine in DNA extract by micellar electrokinetic chromatography with indirect laser-induced fluorescence detection, Talanta 74 (2008) 1050–1055. [8] L. Claire, Camp, L. Barry, Sharp, J. Helen, Reid, John Entwisle, Heidi Goenaga- Infante, Analysis of mono-phosphate nucleotides as a potential method for quantification of DNA using high performance liquid chromatography–inductively coupled plasma-mass spectrometry, Anal. Bioanal. Chem. 402 (2012) 367–372. [9] Y. Fan, K.-J. Huang, D.-J. Niu, C.-P. Yang, Q.-S. Jing, TiO2 -graphene nanocomposite for electrochemical sensing of adenine and guanine, Electrochim. Acta 56 (2011) 4685–4690. [10] T. Liu, X. Zhu, L. Cui, P. Ju, X. Qu, S. Ai, Simultaneous determination of adenine and guanine utilizing PbO2 -carbon nanotubes-ionic liquid composite film modified glassy carbon electrode, J. Electroanal. Chem. 651 (2011) 216–221. [11] Y. Wei, Q.-A. Huang, M.-G. Li, X.-J. Huang, B. Fang, L. Wang, CeO2 nanoparticles decorated multi-walled carbon nanotubes for electrochemical determination of guanine and adenine, Electrochim. Acta 56 (2011) 8571–8575. [12] Bin Fang, Wei Zhang, Guangfeng Wang, Hongying Liu, Shouping Wei, Microwave-assisted preparation of a carbon nanotube/La(OH)3 nanocomposite, and its application to electrochemical determination of adenine and guanine, Microchim, Acta 162 (2008) 175–180. [13] H. Liu, G. Wang, D. Chen, W. Zhang, C. Li, B. Fang, Fabrication of polythionine/NPAu/MWNTs modified electrode for simultaneous determination of adenine and guanine in DNA, Sens. Actuator B 128 (2008) 414–421. [14] H. Yin, Y. Zhou, Q. Ma, S. Ai, P. Ju, L. Zhu, L. Lu, Electrochemical oxidation behavior of guanine and adenine on graphene–nafion composite film modified glassy carbon electrode and the simultaneous determination, Process Biochem. 45 (2010) 1707–1712. [15] Q. Shen, X. Wang, Simultaneous determination of adenine, guanine and thymine based on b-cyclodextrin/MWNTs modified electrode, J Electroanal. Chem. 632 (2009) 149–153. [16] C. Tang, U. Yogeswaran, S.-M. Chen, Simultaneous determination of adenine guanine and thymine at multi-walled carbon nanotubes incorporated with poly(new fuchsin) composite film, Anal. Chim. Acta 636 (2009) 19–27. [17] U. Yogeswaran, S. Thiagarajan, S.-M. Chen, Pinecone shape hydroxypropyl-bcyclodextrin on a film of multi-walled carbon nanotubes coated with gold particles for the simultaneous determination of tyrosine, guanine, adenine and thymine, Carbon 45 (2007) 2783–2796. [18] A. Ali, M. Ensafi, B. Jafari-Asl, A.R. Rezaei, Allafchian, Simultaneous determination of guanine and adenine in DNA based on NiFe2 O4 magnetic nanoparticles decorated MWCNTs as a novel electrochemical sensor using adsorptive stripping voltammetry, Sens. Actuators B 177 (2013) 634–642. [19] Chunxiang Li, Xiyang Qiu, Yulin Ling, Electrocatalytic oxidation and the simultaneous determination of guanine and adenine on (2, 6-pyridinedicarboxylic acid)/grapheme composite film modified electrode, J. Electroanal. Chem. 704 (2013) 44–49. [20] R. Thangaraj, A.S. Kumar, Simultaneous detection of guanine and adenine in DNA and meat samples using graphitized mesoporous carbon modified electrode, J Solid State Electrochem. 17 (2013) 583–590. [21] K.-J. Huang, D.-J. Niu, J.-Y. Sun, C.-H. Han, Z.-W. Wu, Y. -Li Li, X.-Q. Xiong, Novel electrochemical sensor based on functionalized graphene for simultaneous determination of adenine and guanine in DNA, Colloids Surf. B 82 (2011) 543–549. [22] M. Arvand, R. Motaghed Mazhabi, A. Niazi, Simultaneous determination of guanine, adenine and thymine using a modified carbon paste electrode by TiO2 nanoparticles-magnesium(II) doped natrolite zeolite, Electrochim Acta 89 (2013) 669–679. [23] Hung-Wei Chu, R. Thangamuthu, Shen-Ming Chen, Zinc Oxide/zinc hexacyanoferrate hybrid film-modified electrodes for guanine detection, Electroanalysis 19 (2007) 1944–1951. [24] Erbao Liu, Bingchun Xue, Flow injection determination of adenine at trace level based on luminol–K2 Cr2 O7 chemiluminescence in a micellar medium, J. Pharmaceut. Biomed. 41 (2006) 649–653. [25] Aysegul Uygun, D.N.A. hybridization electrochemical biosensor using a functionalized polythiophene, Talanta 79 (2009) 194–198. [26] Chaker Tlili, Hafsa Korri-Youssoufi, Laurence Ponsonnet, Claude Martelet, J. Nicole, Jaffrezic-Renault, Electrochemical impedance probing of DNA hybridization on oligonucleotide-functionalised polypyrrole, Talanta 68 (2005) 131–137. [27] Joseph Wang, Liang Chen, Manuel Chicharro, Trace measurements of nucleic acids using flow injection amperometry, Anal. Chim. Acta 319 (1996) 347–352. [28] M.S. Goh, M. Pumera, Oxidation of DNA bases influenced by the presence of other bases, Electroanalysis 24 (2012) 1147–1152. [29] P. Serp, M. Corrias, P. Kalck, Carbon nanotubes and nanofibers in catalysis, Appl. Catal. 253 (2003) 337–358. [30] V. Vamvakaki, K. Tsagaraki, N. Chaniotakis, Carbon nanofiber-based glucose biosensor, Anal. Chem. 78 (2006) 5538–5542. [31] S.-U. Kim, K.-H. Lee, Carbon nanofiber composites for the electrodes of electrochemical capacitors, Chem. Phys. Lett. 400 (2004) 253–257. [32] C.E. Banks, R.G. Compton, Exploring the electrocatalytic sites of carbon nanotubes for NADH detection: an edge plane pyrolytic graphite electrode study, Analyst 130 (2005) 1232–1239.

R. Thangaraj et al. / Electrochimica Acta 123 (2014) 485–493 [33] S. Nellaiappna, A.S. Kumar, Selective flow injection analysis of iodate in iodized table salts by riboflavin immobilized multiwalled carbon nanotubes chemically modified electrode, Electrochim. Acta 109 (2013) 59–66. [34] C.D. Curtis, D.L. Thorngren, A.M. Nardulli, Immunohistochemical analysis of oxidative stress and DNA repair proteins in normal mammary and breast cancer tissues, BMC Cancer 10 (2010) 9. [35] S.V. Vulimiri, X. Wu, W. Baer-Dubowska, M. de Andrade, M. Detry, M.R. Spitz, J. DiGiovanni, Analysis of aromatic DNA adducts and 7, 8-dihydro-8-oxo-2deoxyguanosine in lymphocyte DNA from a case-control study of lung cancer involving minority populations, Mol. Carcinog. 27 (2000) 34–46.

493

[36] J. Zhang, I.B. Dhakal, G. Greene, N.P. Lang, F.F. Kadlubar, Polymorphisms in hOGG1 and XRCC1 and risk of prostate cancer: effects modified by plasma antioxidants, Urology 75 (2010) 779–785. [37] Ying Wang, Congjuan Li, Xiaohong Li, Yongfang Li, Heinz-Bernhard Kraatz, Unlabeled hairpin-DNA probe for the detection of single-nucleotide mismatches by electrochemical impedance spectroscopy, Anal. Chem. 80 (2008) 2255–2260. [38] Chaker Tlili, Hafsa Korri-Youssoufi, Laurence Ponsonnet, Claude Martelet, J. Nicole, Jaffrezic-Renault, Electrochemical impedance probing of DNA hybridisation on oligonucleotide-functionalised polypyrrole, Talanta 68 (2005) 131–137.