Voltammetric and sonovoltammetric studies on the oxidation of thymine and cytosine at a glassy carbon electrode

Voltammetric and sonovoltammetric studies on the oxidation of thymine and cytosine at a glassy carbon electrode

ELSEVIER Journal of Electroanalytical Chemistry 429 (1997) 95-99 Voltammetric and sonovoltammetric studies on the oxidation of thymine and cytosine ...

610KB Sizes 1 Downloads 10 Views

ELSEVIER

Journal of Electroanalytical Chemistry 429 (1997) 95-99

Voltammetric and sonovoltammetric studies on the oxidation of thymine and cytosine at a glassy carbon electrode Ana Maria Oliveira Brett *, Frank-Michael Matysik Departamento de Qufmica, Faculdade de Ci~ncias e Tecnologia, Universidade de Coimbra, 3000 Coimbra, Portugal

Received 13 August 1996; revised 21 October 1996

Abstract

The voltammetric behaviour of the pyrimidine bases thymine and cytosine is studied using a glassy carbon electrode. In contrast to previous reports that assume electroinactivity at carbon electrodes, it is demonstrated that both compounds undergo well-defined oxidation at a glassy carbon electrode. The experimental conditions that influence the electrode reaction, in particular the pH of the electrolyte solution, are varied systematically and optimized for voltammetric determinations. The application of ultrasound in combination with differential pulse voltammetry results in a reliable analytical procedure for thymine and cytosine measurements avoiding electrode fouling and maintaining the electrode characteristics. The effect of ultrasound is mainly to enhance transport of electroactive species and to clean the electrode in situ. Besides studying the sonovoltammetric behaviour of cytosine and thymine separately, simultaneous determinations are also performed and extended to the analysis of adenine, guanine, thymine and cytosine in the same solution. © 1997 Elsevier Science S.A. Keywords: Cytosine; Thymine; Sonoelectrochemistry; Ultrasound; Voltammetry

1. Introduction

Pyrimidine and purine derivatives play an essential role in various biological processes. In particular, the nucleotides of thymine and cytosine together with those of adenine and guanine represent the monomer units of nucleic acids. The genetic information of deoxyribonucleic acid (DNA) is determined by the sequence of the purine and pyrimidine bases, whereas ribose and phosphate groups of the nucleotide units have a structural role. Investigations of the redox behaviour of biologically occurring compounds by means of electrochemical techniques have the potential for providing valuable insights into biological redox reactions of such biomolecules. In the case of the pyrimidine bases thymine and cytosine several studies have been undertaken concerning their redox behaviour at mercury electrodes. Cytosine was found to be reducible at the mercury electrode [1-4], whereas thymine was reported to show no polarographic reduction wave [1,2]. In addition, both bases form sparingly soluble compounds with the mercury electrode which allows their

* Corresponding author. 0022-0728/97/$17.00 © 1997 Elsevier Science S.A. All rights reserved. PII S0022-0728(96)05018-8

determination by cathodic stripping voltammetry [5]. However, according to previous literature reports cytosine and thymine have been assumed to exhibit no electrochemical activity at graphite electrodes [3,6,7] or carbon based electrodes in general [8,9]. During the course of this work we found that both pyrimidine bases undergo oxidation at a glassy carbon electrode. The present paper deals with the optimization of the conditions that influence the voltammetric response, e.g. the pH of the electrolyte solution and the conditioning of the electrode surface. Ultrasonic pretreatment of the glassy carbon electrode was utilized in order to permit reliable voltammetric determinations of thymine and cytosine. A recently developed small-volume sonovoltammetric cell [10] in which the glassy carbon electrode is exposed to high intensity ultrasound was found to be suitable for these studies. The main effects of ultrasound that can be exploited for electrochemical purposes are the enormous enhancement of mass transport [11-13] and electrode surface modification or cleaning which helps to avoid progressive electrode fouling [14,15]. The latter aspect was found to be particularly useful in the present study in order to enhance the reliability of cytosine and thymine determinations.

96

A.M. Oliveira Brett, F.-M. Ma~.'sik/ Journai of Electroanalytical Chemistry 429 (1997) 95-99

As a result of separate studies on the voltammetfic behaviour of cytosine and thymine, optimized conditions were selected for the determination of both compounds in the same solution and it will be demonstrated that even all four DNA bases, i.e. adenine, guanine, thymine and cytosine, can be determined simultaneously by ultrasound-assisted differential pulse voltammetry.

2. Experimental The cell configuration used for the sonovoltammetric experiments has been described in detail in a previous paper [10]. The thermostatted glass cell (25°C) contained 20 ml electrolyte solution. A glassy carbon electrode (GCE) of 6 mm diameter was positioned at the bottom of the cell so as to face the tip of the sonic horn at precisely measured and calibrated distances. A platinum coil was used as a counter electrode and a laboratory-made silver[silver chloridel3 M KCi electrode served as the reference electrode. The horn was connected to a tapered microtip (d = 3 mm) fabricated from high grade titanium alloy. The ultrasonic processor was a Model VCS01 (Sonics and Materials Inc., USA) capable of delivering up to 500W at 20kHz frequency. The ultrasonic processor L, d:'signed to deliver a constant amplitude that can be selected via the amplitude control setting, ranging from 0 to 100; however, in conjunction with the microtip the amplitude control setting must not be higher than 40. The actual power intensity entering the system was calibrated calotimetrically according to the procedure of Mason et al. [16]. For relevant amplitude control settings of 10, 15 and 20 the corresponding power intensities were 16 + 3, 30 + 3 and 72 + 5 W cm-2 respectively. The sonovoltammetric cell and the sonic horn were housed in a sound-proofed cage in order to protect the operator from high-intensity acoustic noise. All voltammetric experiments were done using an Autolab PGSTATI0 potentiostat (Eco Chemie, Utrecht, Netherlands) equipped with an ECD low current module. The current signal was filtered through a third order Sallen-Key filter with a time constant of 0.1 s in order to remove high-frequency a.c. components. The pH values of electrolyte solutions were measured using a Crison Model micropH 2001 pH-meter and an Ingold combination glass electrode. The GCE (a gift from Professor G. Jenkins, A&M University, Normal, Alabama, USA) was prepared for measurement by polishing using plastic foils (Hirschmann, Germany) with adherent alumina of decreasing particle size ranging from 9 to 0.3 Ixm, followed by thorough rinsing with Milli-Q water. Prior to recording voltammograms of electroactive species, several cyclic voltammograms were performed within the same potential limits in the background solution until a stable voltammetric response was obtained. Adenine, guanine, thymine, cytosine, guanosine, adeno-

sine, thymidine, cytidine and the mono-, di-, and triphosphates of thymidine and cyddine were obtained from Sigma Chemical Co. and used as received. All solutions were made using high-purity water from a Millipore MilliQ system (resistivity > 18 M II cm) and analytical-reagent grade chemicals. The following supporting electrolytes were used: 0.1 M perchloric acid, acetate buffer solutions containing 0.1 M sodium acetate/acetic acid covering the pH range between 4.0 and 5.5, 0.1 M phosphate buffer solutions (pH 6.5 to 7.5), 0.1 M borate/sodium hydroxide solutions for the pH range between 9.2 and 12.3, and a 0.1 M sodium hydroxide solution. The purine and pyrimidine derivatives were dissolved directly in the buffer solutions except in the case of guanine. Stock solutions of I mM guanine were made either in 0.1 M NaOH or in 0.1 M HCIO4. Working solutions of guanine were prepared by adding small volumes of stock solution to the corresponding buffer solution. The solutions were then sonicated in order to ensure homogenization.

3. Results and discussion Initial studies of the voltammetric behaviour of cytosine and thymine were performed in acetate buffer (pH 4.50). Cyclic voltammograms were recorded in the absence and presence of cytosine and thymine respectively. In the case of cytosine an increase in current occurred at the positive limit of the accessible potential range (ca. 1.45 V), whereas thymine already showed a well-defined oxidation wave with a half-wave potential of 1.27 V. The pH dependence of the oxidation of both pyrimidine bases was studied systematically in the pH range between 1 and 13. In 0.1 M HCIO4 only thymine gave an oxidation signal; cytosine was not oxidizable at the glassy carbon electrode within the accessible potential region. Fig. 1 illustrates the dependence of differential pulse peak potentials of cytosine and

1.5-" 1.41.3-

>

1.2 1.1 1.0 O.9

0.7

~

~

1'0

~'2

~

pH Fig. 1. Dependence of differential pulse (DP) peak potential of thymine (Q) and cytosine (ll) on pH of the supporting electrolyte solution. Thymine and cytosine concentrations 5× 10-4 M. Scan rate 5mVs -l, amplitude 50mV.

A.M. Oliveira Brett, F.-M. Matysik / Journal of Electroanalytical Chemistry 429 (1997) 95-99 A

I

,

I

0.2

,

I

0.4

0.6

t

i

.

I

0.8

.

I

1.0

,

I

1.2

"':.4

E/V ',,/(1)I , f " / , ~'2)1

B

/

.,llO)l

lO0 (b)

,

0.2

0.4

..a..---

0.6

,

0.8

.

.

1.0

1.2

1.4

EIV Fig. 2. Cyclic voltammograms at a glassy carbon electrode in borate buffer: (A) 2.6 mM thymine (pH 10.0) and (B) 5 mM cytosine (pH 10.79). Scans (1) to (4) are successive recordings, (b) background response. Scan rate 50mVs -t.

thymine on pH. For thymine the slope of the Ep-pH plot is - 6 0 mV per pH unit over the whole pH range studied, which suggests that the number of protons and electrons involved in the oxidation mechanism is equal. In the case of cytosine we find a change in slope of the Ep-pH plot at about pH 10. For pH values lower than 10 the slope is - 6 0 m V per pH unit and in the more alkaline region aEp/0pH is - 8 5 mV per pH unit. This indicates that the ratio of the number of protons and transferred electrons shifts at pH 10 from 1 to 3/2. This could be explained if, for example, a product of oxidation undergoes deprotona-

~

1

oo w~

(1)

97

tion at pH > 10 while the prote,l}~ic state of the educt remains as it was at lower pH. Delia et al. [ 17] reported the formation of cytosine 3-N-oxide by homogeneous reaction between cytosine and m-chloroperbenzoic acid. They determined pK values of 4.82 and 10.3 for cytosine 3-Noxide, as comp,'u'ed with 4.60 and 12.16 for cytosine. Speculating that cytosine 3-N-oxide is the oxidation product formed voltammetrically, this could explain the change in slope of the Ep-pH function for cytosine as discussed above, because cytosine 3-N-oxide becomes deprotonated at pH higher than 10.3. However, there is no proof that the electrochemical oxidation follows the same mechanism. More work has to be done concerning the analytical identification of the products of electro-oxidation of thymine and cytosine respectively. From the results obtained by studying the pH dependence of the oxidation potentials of cytosine and thymine, optimum pH values were selected for further characterization by cyclic voltammetry. Fig. 2 shows the cyclic voltammetric responses of both compounds. The oxidation signals are well resolved from the background response. In contrast, the nucleosides and nucleotides of thymine and adenine were found not to be oxidizable under the same conditions. The cyclic voltammograms shown in Fig. 2 exhibit an obvious signal decrease during successive recordings. This behaviour is similar to that of the corresponding purine compounds [15] and results from the adsorption of oxidation products that block the electrode surface. The situation can be improved considerably by applying ultrasound while recording the cyclic voltammogram. This is illustrated in Fig. 3 where successive cyclic sonovoltammograms of thymine are presented that show no tendency to signal decrease. However, the limiting current region of the sonovoltammograms is not very extended and consequently not well suited for quantitative evaiuation. Therefore, the analytical procedure for quantitative determinations of thymine and cytosine is based on ultrasound-assisted differential pulse voltammetry (DPV) where ultrasound is applied (1 to 2 min) but is switched off in the

(2)

(3)

(1)

, (2)

, (3)

, (4)

2o

, (5)

,

/

y/)'/y .a...--

'o'. 4

0.8 ........

1.0

1.2

1.4

EIV Fig. 3. Cyclic sonovoltammograms of 1 × 10-4M thymine in borate buffer (pH 10.0) at a glassy carbon electrode. Scans (1) to (3) represent successive sonovoltammograms. Ultrasound conditions: 5 mm horn tipelectrode separation, power intensity 30Wcm -2. Scan rate 50mVs-

I

I

I

0.8

1.0

1,2

E/V Fig. 4. Repetitive ultrasound-assisted differential pulse voltammograms of 5x IO-4M cytosine in borate buffer (pH 10.79). Ultrasound (power intensity 30Wcm-2 horn tip-electrode separation 6 mm) is applied until 0.8 V. The potential window is given for voltammogram (3). DP conditions as in Fig. 1.

98

A.M. Oliveira Brett. F.-M. Ma~. sik / Joun~al of Electroanalytical Chemistry 429 (1997) 95-99

~

o~

02

,

0,

,

0'6

,

0'8

,

I

,

i2

I

,

I

/////~

1

I

I

I

I

0.4 0.6 0.8 1.0 1.2 1.4 16 E/V

EIV Fig. 5. Ultrasound-assisteddifferential pulse voltammetric determinations of thymine and cytosine in borate buffer (pH 10.02). Ultrasonic pretreatment (power intensity 30Wcm -2, horn tip-electrode separation 5 mm) until 0.75V. Concentrations: (!) background response, (2) 5x lO-4M thymine, (3) 5 x l O - 4 M thymine and 2.5x lO-4M cytosine, (4) 5x 10-4 M thymine and 5 x 10-4 M cytosine. DP conditions as in Fig. !.

relevant potential region of the differential pulse voltammetric recording in order to avoid negative effects on the precision of the signal due to ultrasonically induced mass transport fluctuations. By means of this analytical strategy excellent reproducibility is obtained, as illustrated for repetitive determinations of cytosiPe shown in Fig. 4. No electrode fouling effects occur over measuring periods of several hours. The differential pulse voltammetric response of thymine is similar and an even more symmetrical peak is obtained because the signal appears at about 200mV ~ess positive potentials. Results of calibration measurements are shown in Table 1, which demonstrate the utility of ultrasound-assisted DPV for quantitative determinations of thymine and cytosine. The limit of detection (defined as the concentration that leads to a signal which is three times the standard deviation of the baseline noise) is 1 × 10-5 M for both compounds. The simultaneous determination of thymine and cytosine was performed at pH 10, which is the optimum with respect to peak separation. Fig. 5 shows ultrasound-asbisted DPV recordings for a constant concentration of thymine in the absence and at two different concentrations of cytosine. The simultaneous quantitative determination of both compounds is possible, but the effect of cytosine on the signal shape of thymine has to be taken into consideration.

Fig. 6. Differential pulse voltammograms of 6x 10-SM adenine and 5 × 10-5 M adenosine. Voltammograms(I), (2), (3) and (6) are recorded in combination with ultrasonic pretreatment, (4) and (5) without ultrasonic pretreatment. Arrows indic;,te the potential where the ultrasound is switched off; ultrasound conditions: 2 mm horn tip-electrode separation, power intensity 30Wcm-2, DP conditions as in Fig. !. Supporting electrolyte acetate buffer (0. ! M, pH 4.50).

This procedure was tested for the determination of all four DNA bases, i.e. adenine, guanine, thymine and cytosine in the same solution by performing just one differential pulse voltammetric run. The sonovoltammetric behaviour of guanine has been reported in detail in a previous paper [15], which demonstrates the sonovoltammetric determination of guanine. Cyclic sonovoltammograms of adenine (not shown) indicate that there are more serious problems of electrode blocking effects than in the case of guanine, at least for adenine concentrations higher than l0 -5 M. However, by means of ultrasound-assisted DPV which allows the selection of higher power intensities a n d / o r closer horn tip-electrode separations, reliable adenine determinations are possible up to adenine concentrations of 10 -4 M. For example, Fig. 6 illustrates repetitive determinations of adenine in the presence of the corresponding nucleoside adenosine by DPV. It is obvious from Fig. 6 that those DPV recordings after ultrasonic pretreatment lead to reproducible signals for adenine (Ep = 1.09 V) and adenosine (Ep = 1.30 V). In contrast, as demonstrated by recordings (4) and (5) of Fig. 6 without ultrasonic pretreatment, the DP signals tend to decrease and even an additional peak (Ep = 0.86V) resulting from some adsorbed product occurs in scan (5).

Table ! Results of linear regression of calibration data for cytosine and thymine determination by means of ultrasound-assisted DPV. Concentration range 5 x 10-5 to 5 x 10-4 M for both compounds. Experimental conditions as in Fig. 5 Analyte thymine cytosine

Buffer borate buffer (0.1 M, pH 10.02) borate buffer (0.1 M, pH 10.79)

'~;lope/gA gM-i 0.031 0.051

Intercept/izA -0.40 0.11

Regression coefficient 0.9986 (n = 10) 0.9998 (n = 10)

m

A.M. Oliveira Bretl, F,-M. Matysik / Journal of Electroanalytical Chemistry 429 (1997) 95-99

T

/ . I

tl

~

'-~..,'--~

,_,

/;7,"

(b>

(a) -----""~

0.0

0.2

. . . .

0.4

0.6

0.8

1.0

1.2

1.4

99

show no electrochemical activity at carbon electrodes [3,6-9]. The problem of progressive electrode fouling encountered when performing repetitive determinations of thymine and cytosine has been solved by applying ultrasound-assisted DPV. On the basis of this approach cytosine and thymine can be reliably determined either separately or in a mixture. Finally, it has been shown that all four DNA bases, adenine, guanine, thymine and cytosine, can be measured simultaneously by ultrasound-assisted DPV. This is obviously a clear advantage over measurements based on the dropping mercury electrode which does not offer such versatility in the context of the analysis of DNA bases.

EIV

Fig. 7. Differential pulse voltammetric determination of purine and pyrimidine bases guanine (2× l0 -5 M), adenine (3 x 10-5 M), thymine (3 x l0 -4 M) and cytosine (3 × l0 -4 M) in borate buffer (pH 10.02), (a) with ultrasonic pretreatment (power intensity 72 Wcm-2 horn tip-electrode separation 5ram), (b) successive scan without ultrasonic pretreatment. DP conditions as in Fig. 1. Finally, a mixture of adenine, guanine, thymine and cytosine was studied by DPV under conditions that give the best peak separation for thymine and cytosine. Fig. 7 shows well separated signals for each compound and demonstrates the excellent multi-component capability of the proposed method. The difference between the DPV recordings 7(a) and 7(b) is that the latter was measured without ultrasonic pretreatment. It can be seen from Fig. 7 that the adenine and guanine peaks in 7(b) are somewhat larger than in 7(a) and shifted slightly towards more negative potentials. This is due to contributions to the signal from adsorbed adenine and guanine [15]. However, the adsorption processes are difficult to control and have negative effects on the reproducibility, which is much better when the voltammetric measurement is combined with ultrasonic pretreatment.

4. Conclusions It has been demonstrated that the pyrimidine bases thymine and cytosine undergo oxidation at glassy carbon electrodes, whereas previously it was assumed that they

Acknowledgements We thank the European Community for financial support (Contract No. CHRX CT94 0475) under the Human Capital and Mobility Scheme.

References [1] E. Palecek and B. Jan~, Arch. Biochem. Biophys., 98 (1962) 527. [2] B. Jan~ and E. Palecek, Arch. Biochem. Biophys., 105 (1964) 225. [3] G. Dryhurst and P.J. EIving, Talanta, 16 (1969) 855. [4] B. Jan~ and P.J. Elving, Chem. Rev., 68 (1968) 295. [5] E. Palecek, Elecuoanalysis, 8 (1996) 7. [6] D. Smith and P.J. Elving, Anal. Chem., 8 (1962) 930. [7] V. Brabec, Bioelectrochem. Bioenerg., 8 (1981)437. [8] J.P. Hart, Electroanalysis of Biologically Important Compounds, Ellis Horwood, Chichester, 1990, p. 46. [9] K. Stul~ and V. Pacfikovfi,ElectroanalyticalMeasurements in Flowing Liquids, Ellis Horwood, Chichester, 1987, p. 262. [10] A.M. Oliveira Brett and F.-M. Matysik, Electrochim. Acta, in press. [1 !] R.G. Compton, J.C. Eklund, S.D. Page, G.H.W. Sanders and J. Booth, J. Phys. Chem., 98 (1994) 12410. [12] C.R.S. Hagan and L.A. Coury, Anal. Chem., 66 (1994) 399. [13] D.J. Walton, S.S. Phull, A. Chyla, J.P. Lorimer, T.J. Mason, L.D. Burke, M. Murphy, R.G. Compton, J.C. Eklund and S.D. Page, J. Appl. Electrochem., 25 (1995) 1083. [14] H. Zhang and L.A. Coury, Anal. Chem., 65 (1993) 1552. [15] A.M. Oiiveira Brett and F.-M. Matysik, Bioelectrochem. Bioenerg., in press. [16] T.J. Mason, J.P. Lorimer and D.M. Bates, Ultrasonics, 30 (1992)40. [17] T.J. Delia, M.J. Olsen and G.B. Brown, J. Org. Chem., 30 (1965) 2766.