Electrochemistry Communications 11 (2009) 1555–1558

Contents lists available at ScienceDirect

Electrochemistry Communications

journal homepage: www.elsevier .com/ locate /e lecom

Epoxide–quinone transformations: Multi-parametric indicatorsfor assessing animal welfare

Lucy Lee a, Maria Marti Villalba a, Robert B. Smith b, James Davis a,*

a School of Science and Technology, Nottingham Trent University, Clifton Lane, Nottingham, NG11 8NS, UKb Centre for Materials Science, University of Central Lancashire, Preston, PR1 2HE, UK

a r t i c l e i n f o

Article history:Received 28 April 2009Received in revised form 17 May 2009Accepted 19 May 2009Available online 23 June 2009

Keywords:EpoxideGlutathioneAnimal welfareStressThiol

1388-2481/$ - see front matter � 2009 Elsevier B.V. Adoi:10.1016/j.elecom.2009.05.056

* Corresponding author. Tel.: +44 (0) 115 848 3218E-mail address: james.davis@ntu.ac.uk (J. Davis).

a b s t r a c t

The use of a novel epoxide label system as the basis of a new approach to monitoring thiol species rel-evant to animal welfare contexts is presented. The need for the development of on-site/field testing isdemonstrated through the analysis of equine saliva before and after exercise and the implication ofdelays in the analysis process exposed. The system is selective for thiols and, in contrast to simpler label-ling systems, could offer potential for speciation studies.

� 2009 Elsevier B.V. All rights reserved.

1. Introduction

Assessing animal welfare has risen to considerable importancein recent years given increasing public, media and regulatory con-cerns. Obtaining unambiguous measures of maltreatment howeveris contentious and, barring the presence of gross physical injury,can be highly subjective [1,2]. Cortisol measurements have beenthe main biochemical measure of ‘‘stress” but there are many lim-itations to their use and interpretation and there is a need for analternative or a complementary system that could be used to mon-itor molecular level insults to the animal wellbeing [3,4]. Glutathi-one has emerged as a potential candidate given its application inthe study of numerous human clinical conditions [5,6] and it couldbe anticipated that the underlying biochemical similarities wouldallow its extrapolation to animal contexts [7,8]. The aim of thepresent communication has been to assess a new class of molecu-lar label as potential contenders and elucidate the underlying reac-tion mechanisms involved in the detection processes.

The proposed system is based on an epoxide label which selec-tively reacts with the target thiol to yield a conjugate with uniquespectroscopic and electrochemical properties. The basic process ishighlighted in Scheme 1. In contrast to previous studies involvingquinone moieties, in the absence of the target thiol – the indicatordoes not possess any native electrochemistry as the quinone

ll rights reserved.

; fax: +44 (0) 1483 686851.

component has been removed by the presence of the epoxide func-tionality. Upon reaction with the target, a number of options canarise [9–11] and it is the objectives of the current investigationto identify those that predominate, assess selectivity and evaluatethe use of the label as a quantitative measure of glutathione rela-tive to conventional lab based procedures. A critical appraisal ofits potential for being harnessed as the basis of a field test is alsoprovided.

2. Experimental details

2.1. Materials and methods

All reagents were of the highest grade available and usedwithout further purification. Electrochemical measurements wereconducted using a lAutolab computer controlled potentiostat(Eco-Chemie, Utrecht, The Netherlands) using a three electrodeconfiguration consisting of a glassy carbon working electrode(3 mm diameter, BAS Technicol, UK). Platinum wire served as thecounter electrode with a 3 M NaCl Ag | AgCl half cell referenceelectrode (BAS Technicol, UK) completing the cell assembly. Allmeasurements were conducted at 22 ± 2 �C. Chromatographicanalysis was carried out using an Agilent 1100 series high perfor-mance liquid chromatography system, throughout, a 150 �4.60 mm Sphereclone 3 lm column was used with a mobile phaseconsisting of 65/35 methanol–water methanol at a flow rate of1 ml/min. The detection wavelength used throughout was

Fig. 1. UV–vis spectra detailing the epoxide response towards the glutathione,cysteine and homocysteine, (200 lM, pH 7). Inset: molar absorptivity data.

Scheme 1.

1556 L. Lee et al. / Electrochemistry Communications 11 (2009) 1555–1558

250 nm. NMR spectra were measured on a Bruker 250 MHz spec-trometer. Chemical shifts are reported in parts per million (ppm)downfield from tetramethylsilane (TMS). Mass spectra were re-corded at the EPSRC Mass Spectrometry Centre at SwanseaUniversity.

2.2. Syntheses

2,3-Epoxy-1,4-naphthoquinone/2,3-Epoxy-2-methyl-1,4-naphthoquinone

Into a boiling tube (A) was placed either 1,4-naphthoquinone(0.91 g, 5.8 mmol) or 2-methyl-1,4-naphthoquinone (1.0 g,5.8 mmol) and ethanol (10 mL). The tube was heated until the en-tire solid dissolved. Into a second boiling tube (B) was added so-dium carbonate (0.2 g), distilled water (5 mL) and cold 30%hydrogen peroxide (2 mL). Boiling tube A was placed under a coldtap until very slight crystallization occurred, then boiling tube Bwas added to A all in one go. With swirling boiling tube A wascooled in ice water which produced a white solid. The white solidwas filtered using suction filtration and washed with water(3 � 10 ml) and air dried to yield the appropriate epoxide.

The yield of 2,3-epoxy-1,4-naphthoquinone was 0.39 g, 39% aswhite crystals. 1H NMR d: 4.01 (s, 2H, C–H) 7.99–7.95 (m, 2H, Ar-H) 7.77–7.73 (m, 2H, Ar-H); 13C NMR d: 55.10, 127.02, 131.65,134.57, 190.59. LCMS – ESI (m-z) 174 [M-H].

The yield of 2,3-epoxy-2-methyl-1,4-naphthoquinone was0.96 g, 88% as white crystals. 1H NMR d: 1.78 (s, 3H, CH3), 3.90

(s, 1H, C–H), 7.75–8.07 (m, 4H, Ar-H). 13C NMR d: 14.75, 61.40,61.48, 126.88, 127.50, 132.01, 132.12, 134.47, 134.64, 191.84,192.00. LCMS – ESI (m-z) 188 [M-H].

3. Results and discussion

The need for the development of on-site analysis kits or devicesfor the analysis of thiol concentrations is well established given theair sensitivity of the primary analyte and the inevitable degrada-tion that can occur between sample extraction and the subsequentanalysis. It is our contention in the present case that the electro-chemical approach offers a more versatile and indeed viable alter-native over conventional colorimetric tests (i.e. Ellman’s) wherethe sample can be irrevocably contaminated by the highly variablebackground colour of the sample – in this particular case saliva –by grazing.

The spectroscopic properties of the system were neverthelessbriefly assessed given the novelty of the proposed system and toset the benchmark through which the electrochemical propertiescould be assessed. The UV–vis spectra relating to the respectivethiol (200 lM glutathione, cysteine or homocysteine) are com-pared in Fig. 1. It can be seen that the profiles are markedly differ-ent with the cysteine generating a purple colour in contrast to theorange hues of glutathione and homocysteine. The cysteine profileappears to consist of two distinct species (kmax1 = 475 nm andkmax2 = 610 nm) and it is likely that these reflect the singlesubstituted open-ring derivative (similar to that attainable by glu-tathione at kmax = 455 nm) and a second which is attributed tothe ring closed intermediate. The spectra relating to glutathioneexhibit no absorbance at the higher wavelength (kmax2 =610 nm). It is possible therefore to consider that in a mixture ofthe two, the respective concentration of both could be obtainedthrough conventional two component multi-parametric analysis.In contrast to the unsubstituted derivative, there was no purplecolouration upon the addition of cysteine to the methyl substi-tuted epoxide with little difference observed in the spectroscopicprofile between the different thiols. As such it was not investi-gated further.

L. Lee et al. / Electrochemistry Communications 11 (2009) 1555–1558 1557

Two possible reaction routes (I–II–III–IV and I–II–V–VI) are pre-sented in Scheme 1. Previous studies on the reaction of naphtho-quinone [12] and naphthoquinone bromide [13] with glutathioneand cysteine give rise to the production of derivatives identicalto those postulated in III and IV. However, the spectroscopic profileobserved in those cases does not match the products resultingfrom the epoxide–glutathione nor epoxide–cysteine. There wasno purple colouration upon the reaction of naphthoquinone withcysteine and stands in marked contrast to the profile observed inFig. 1. In addition, the naphthoquinone-glutathione conjugate yielda species with a kmax of 420 nm [12,13], not the 455 nm shown inFig. 1 for the epoxide system. It would appear that route V–VI ispredominant though it must be noted that it is unlikely to be thesole process under the assay conditions and a competitive mixtureis inevitable. HPLC analysis (not shown) revealed a spread of prod-ucts and while it was not possible to conclusively identify the var-ious moieties – there were at least three defined products in thechromatograms irrespective of which thiol was used as a probewhich indicates the viability of both routes.

The selectivity of the epoxide system was also assessed with noreaction to lysine or histidine (both 0.2 mM) under the time courseof the reaction. The reaction of such species with epoxides is wellrecognised [14] but, providing the analysis is done within 30 minunder ambient conditions, there is no appreciable interference.N-acetylcysteine (kmax = 475 nm) mimicked the glutathione re-sponse in deference to the purple colouration apparent with cys-teine and corroborates the proposition that the unprotectedprimary amino is required for ring closure and purple colouration.The molar absorptivities of the various components at the differentwavelengths are compared in Table 1 (inset Fig. 1). In all cases itcan seen that the intensity is markedly less than Ellman’s reagent(e = 13,600 L mol�1 cm�1). The main strength of the proposed sys-tem, from a spectroscopic viewpoint, clearly lies in the potentialfor instrumental speciation based on the two component analysis.

The electrochemical properties of the epoxide system wereinvestigated as a means of countering the intensity and matrixcontamination deficiencies of the spectroscopic systems. Voltam-mograms detailing the response of a glassy carbon electrode to

Fig. 2. Cyclic voltammograms detailing the response of a glassy carbon electrodetowards the epoxide (455 lM, pH 7) in the presence of cysteine and glutathione(178 lM, pH 7). Inset: chronoamperometric responses to increasing glutathione(�0.6 V in air).

cysteine and glutathione (both 178 lM) in the presence of theepoxide (455 lM, pH 7) are compared in Fig. 2. In the absence ofany thiol, a single reduction process corresponding to the reduc-tion of the epoxide itself is observed at �0.75 V. The correspondingoxidation peak is found on the reverse scan indicating that the ringhas been opened to yield the corresponding quinone. Upon theaddition of thiol, however, a second reduction process whose posi-tion is dependent upon the nature of the conjugate being formed.In common with the spectroscopic results, different profiles are ob-tained with the cysteine assay yielding two reduction peak pro-cesses indicating the presence of two conjugates representing theopen and closed ring derivatives [15]. The glutathione conjugateprovides a single broad reduction process at potentials more neg-ative than the initial cysteine reaction product (�0.55 V vs�0.4 V, respectively).

While the use of simple naphthoquinones for the electrochem-ical detection of thiols is well established [12,15,16], the mainproblem with developing an onsite based amperometric system re-lates to the fact the measurement would need to be done in air –re-oxidation of the quinone by dissolved oxygen in the samplecould introduce further ambiguities through changing the relativeconcentrations of the oxidised and reduced forms. Such irrepro-ducibility would negate the diagnostic value and is a commonproblem with potentiometric naphthoquinone systems [12].

The great strength of the epoxide label, in contrast to the naph-thoquinone system, relies upon the generation of the oxidised formwith the subsequent measurement based on the reduction of ananalyte that can only arise as a consequence of reaction with athiol. As such, a positive signal should be acquired upon the intro-duction of the thiol. The resulting derivative should be insensitiveto dissolved or ambient oxygen with the reduction of the oxidisedform of the conjugate, but not the epoxide, utilised as the analyti-cal signal. Chronoamperometric responses (Fig. 2: Inset) were mea-sured at a fixed potential of �0.6 V without degassing. The currentwas extracted at 20 s and a linear relationship to glutathione wasobtained (Response/A = 2.53 � 10�3 [GSH/mol L�1] + 3.94 � 10�8;N = 5; R2 = 0.992). An assumption is made whereby glutathione isliable to be the predominant thiol species. It could be anticipatedthat lowering the detection potential to �0.4 V would enable theselective measurement of cysteine as there is no appreciable signalto glutathione and, in principle, provide a degree of speciation.

4. Conclusions

There is at present, a conflict between subjective measures thatcan be taken at the scene and objective measures that require atime delay between collection and the results being available. Anobjective test that can provide immediate in situ results is there-fore a much-needed development in the monitoring of animal wel-fare standards. The epoxide system investigated here lays thefoundation of a new approach that could address many of theshortcomings inherent to the simple naphthoquinone systemsand promote a new avenue for the development of the technology.

References

[1] J.D. Clark, D.R. Rager, J.P. Calpin, Laboratory Animal Science 47 (1997) 586.[2] M.S. Cockram, Animal Welfare 13 (2004) 283.[3] J. Lane, Animal Welfare 15 (2006) 331.[4] B.L. Pearson, P.G. Judge, D.M. Reeder, American Journal of Primatology 70

(2008) 1145.[5] C. Livingstone, J. Davis, British Journal of Diabetes and Vascular Diseases 7

(2007) 258.[6] C.A. Lang, B.J. Mills, W. Mastropaolo, M.C. Liu, Journal of Laboratory and Clinical

Medicine 135 (2000) 402.[7] C.A. Williams, M.E. Gordon, C.L. Betros, K.H. McKeever, Journal of Animal

Science 86 (2008) 576.[8] A. Frankiewicz-Jozko, E. Szarska, Biology of Sport 17 (2000) 217.

1558 L. Lee et al. / Electrochemistry Communications 11 (2009) 1555–1558

[9] Q. Chen, G.A. Doss, E.C. Tung, W. Liu, Y.S. Tang, M.P. Braun, V. Didolkar, J.R.Strauss, R.W. Wang, R.A. Stearns, D.C. Evans, T.A. Baillie, W. Tang, DrugMetabolism and Disposition 34 (2006) 145.

[10] A.S. Kalgutkar, K.R. Henne, M.E. Lame, A.D.N. Vaz, C. Collin, J.R. Soglia, S.X. Zhao,C.E.C.A. Hop, Chemico-Biological Interactions 155 (2005) 10.

[11] M. Kamenjicki, S.A. Asher, Sensors and Actuators B 106 (2005) 373.

[12] S. Gracheva, C. Livingstone, J. Davis, Analytical Chemistry 76 (2004) 3833.[13] G.A.M. Lobo, S.A. Chitre, S.M. Rathod, R.B. Smith, R. Leslie, C. Livingstone, J.

Davis, Journal of Chromatography B – Biomedical Applications 864 (2008) 173.[14] J.W. Newman, B.D. Hammock, Journal of Chromatography A 925 (2001) 223.[15] C.G. Stone, M.F. Cardosi, J. Davis, Analytica Chimica Acta 491 (2003) 203.[16] P.C. White, N.S. Lawrence, J. Davis, R.G. Compton, Electroanalysis 14 (2002) 89.