Original Paper Evaluation of 1,2-Diaminoanthraquinone (DAA) as a ...

Microchim Acta 152, 35–45 (2005) DOI 10.1007/s00604-005-0420-x

Original Paper Evaluation of 1,2-Diaminoanthraquinone (DAA) as a Potential Reagent System for Detection of NO Helen Dacres and Ramaier Narayanaswamy School of Chemical Engineering and Analytical Science, The University of Manchester, Faraday Building, Sackville Street, PO Box 88, Manchester, M60 1QD, UK Received May 22, 2005; accepted August 8, 2005; published online October 17, 2005 # Springer-Verlag 2005

Abstract. The reaction of 1,2-diaminoanthraquinone (DAA) with NO has been investigated for potential development into an optical fibre chemical sensor. In aqueous solutions a linear calibration was obtained between 0 and 40 mM NO with a calculated detection limit of about 5 mM (s=n ¼ 3). DAA was shown to be pH independent between 6 and 10 with a pKa of 5.5 making it a suitable reagent for detection of physiological NO. NO was shown to quench the fluorescence of DAA and this was due to a static quenching mechanism. The effects of nitrite, peroxynitrite, superoxide and hydrogen peroxide were investigated. Peroxynitrite did not react with DAA and nitrite did not interfere with the reaction of NO with DAA. Though attempts to immobilize DAA in silicone rubber and nafion were successful, exposure to NO gas did not result in a significant response. Key words: Nitric oxide detection; fluorescence; nafion; silicone rubber; quenching.

Nitric oxide (NO) has received considerable attention since its identification as a biological signalling molecule [1]. Recent research implicates the involvement of NO in physiological processes including vasodi Author for correspondence. E-mail: Ramaier.Narayanaswamy@ manchester.ac.uk

lation, communication in the brain, and immunological defence against invading organisms [2]. A sensor capable of direct and reversible detection of NO would be invaluable to advance understanding of its physiological role. NO detection methodologies include electrochemical [3, 4], electron paramagnetic resonance (EPR) spectroscopy [5], chromatography [6], chemiluminescence [7], absorbance [8, 9] and fluorescence [10]. All of these methods are reviewed comprehensively in a recent review article [11]. Of these options those based on fluorescence have great potential for investigating biological NO. Early investigations of fluorimetric reagents for NO detection in biological samples involve the synthesis of diaminonapthalene (DAN) [12]. DAN is poorly soluble in solution and a UV excitation wavelength (375 nm) was employed which causes some autofluorescence of endogenous tissue. Diaminofluoresceins (DAFs) [13] and diaminorhodamines (DARs) [14] were synthesised to overcome the problems associated with DAN. Both reagents were prone to instability around neutral pH. Following this 1,3,5,7-tetramethyl-8-(30 ,40 -diaminophenyl)-difluoroboradiaza-sindacene (TMDABODIPY) was synthesised and was shown to be photostable and pH independent over a wide range [15]. However, another problem was encountered on approaching physiological temperatures

36

H. Dacres and R. Narayanaswamy

with TMDABODIPY being quickly protonated which interfered with its response to NO. Optical fibre based NO sensors containing fluorescent dye labelled cytochrome c0 [16] and soluble guanylate cyclase (sGC) [17] have been constructed. Both sensors were fully reversible, exhibited fast response times, stability at neutral pH and excellent specificity for NO. The sensors were applied to detecting intra and extra cellular NO which demonstrates the one of the main advantages of using optical fibre sensors for such an application, namely ease of miniaturisation. We report here the use of 1,2-diaminanthraquinone (DAA), an aromatic vicinal diamine which reacts selectively with NO [18], in the development of an optical sensor for NO with a view to adaptation into an optical fibre sensor. Previously DAA has been used to image NO production in vivo using fluorescence microscopy [18].

DEANO

Experimental

The autoxidation of hydroxylamine in alkaline solutions leads to the formation of peroxynitrite anion as its major product [21, 22]. This method was chosen for peroxynitrite anion synthesis as it is simple, inexpensive and does not require any specialist equipment. The method is based upon the following reaction:

Apparatus Absorbance spectra were recorded using the Perkin Elmer Lambda 5 UV-VIS spectrophotometer (www.perkinelmer.com). All fluorescence studies were carried out using a commercial spectrofluorimeter (Perkin Elmer LS5 for gas exposure studies or Perkin LS55 for solution studies). Quartz cuvettes were used for fluorescence measurements of solutions. For immobilization studies a purpose built gas blender was used to mix appropriate quantities of the gases to produce known concentrations at a controlled flow rate. The spectrofluorimeter was used in conjunction with a purpose built flow cell [19]. Lifetime measurements were recorded with the OB 920 fluorescence life-time spectrometer (Edinburgh Instruments, www. edinst.com).

Solution Studies Reagents All reagents were of analytical grade and used as purchased without further purification. All solutions were prepared with distilled deionised water unless otherwise stated. A stock solution of 1 mM of DAA (Aldrich, www.sigma-aldrich.com) was prepared in water and diluted in phosphate buffer (pH 7.4, 10 mM) to achieve the desired final concentration. The stock solution was sonicated until dissolved prior to dilution and new stock solutions were prepared daily.

Buffer Preparation A 10 mM (pH 7.4) phosphate buffer solution was prepared by mixing 1.6 mmoles of potassium dihydrogen orthophosphate (KH2PO4) (BDH, www.bdhind.com) with 3.4 mmoles of sodium phosphate dibasic (Na2HPO4) (Aldrich) in water. To make buffer solutions of varying pHs (pH 3–10) the contents of hydrion buffer chemenvelopes (Aldrich) were added to 500 mL of distilled water.

A 0.25 mM diethylamine NONOate (DEANO, Sigma, www.sigmaaldrich.com) solution was prepared in phosphate buffer and used immediately after preparation. DEANO is a solid which when dissolved in phosphate buffer (pH 7.4, 22–25 C) dissociates to produce stoichiometric amounts of nitric oxide [20].

Nitrite A 10 mM stock solution of sodium nitrite (Aldrich) was prepared in water by dissolving 69 mg in 100 mL of deionised distilled water. The appropriate amount of nitrite was added to the DAA solutions to give a final concentration of 100 mM.

Hydrogen Peroxide (H2O2) A 30% (w=w) hydrogen peroxide solution was purchased from Sigma and used as received without further purification. From this solution a 0.1 M stock solution was prepared in water. This solution was prepared immediately prior to use and added to the DAA solutions to give the desired final concentration (100 mM). Peroxynitrite Anion (ONOO )

NH2 O þ O2 ! H2 O2 þ NO

ð1Þ

NO þ O2 ! ONOO

ð2Þ

The reaction proceeds by the attack of oxygen on a deprotonated species yielding nitroxyl ion which is further oxidised to peroxynitrite. In the presence of metal sequestering agent ethylenediaminetetraacetic acid (EDTA, Fluka, www.sigma-aldrich.com) the peroxynitrite ion is stabilised. Experiments were carried out with solutions containing 10 mM of hydroxylamine hydrochloride (Aldrich), 0.5 M of sodium hydroxide (NaOH, BDH) and 1 mM of EDTA. The solutions were bubbled with oxygen and stirred vigorously for about 4–5 hours. Following this MnO2 (Manganese (IV) oxide, Fluka) was added to the solution and then filtered and the solution was stored in a freezer. Peroxynitrite ion concentration was determined by UV spectrometry at 302 nm (" ¼ 1670 M1 cm1 ) [22]. The appropriate dilution of the peroxynitrite solution was made to give a solution of 100 mM concentration. Superoxide Anion (O2 ) Potassium superoxide (KO2, Aldrich) solutions can be used to prepare solutions containing about the same concentration of superoxide anion [23, 24]. KO2 is sparingly soluble in dry dimethyl sulfoxide (DMSO, Aldrich) and the use of 18-crown-6 (Aldrich) increases the solubility of KO2 allowing pale yellow solutions to be prepared as long as the crown ether is in at least a 2:1 excess of KO2. A 50 mM solution of KO2 was prepared in a 0.15 M crown-ether:DMSO solution. The solution was sonicated until both KO2 and crown ether had fully dissolved. Superoxide anion concentration was determined by UV spectrometry at 250 nm (" ¼ 2686 29 M1 cm1 ) [25–27]. The appropriate dilution of the superoxide solution was made to give a solution of 100 mM concentration.

37

1,2-Diaminoanthraquinone as a Reagent for Detection of NO Superoxide dismutase (SOD) (Sigma) from bovine blood contained 2800 units mg1 was used. SOD was dissolved in 10 mL of phosphate buffer (0.1 mg mL1 ) and diluted to give a final concentration of 2.5 mg mL1 . This concentration had previously been shown to inhibit the effect of superoxide produced from a 50 mg mL1 KO2 solution by 85–90% [24].

Procedure Varying amounts of DEANO were added to a 100 mM DAA solution to give the desired final concentrations of NO. The temperature was maintained at 22 C and the absorbance was read after 20 minutes reaction time. A final concentration of 100 mM of the appropriate interferent was added to solutions containing 100 mM DAA or 100 mM DAA alongside 50 mM NO. The procedure was repeated three times for each interferent in both solutions.

Immobilization Studies Silicone Rubber Silicone rubber is prepared by cross-linking linear polymeric siloxane chains (polysiloxane polymers). The curing process can be carried out at elevated or room temperature. In the latter case the process is known as room temperature vulcanisation (RTV). In this study a two-part room temperature vulcanisation system (Silgard 184, Dow Corning, www.dowcorning.com) was used to prepare silicone rubber incorporating DAA. Silicone rubber membranes were prepared by mixing a 1 mM DAA solution (toluene=methanol (4:1)) with the polydimethylsiloxane (PDMS) component of a two-part temperature vulcanisation silicone rubber. A typical solution consisted of 350 mL of 1 mM DAA mixed with 1 g of PDMS following the addition of 100 mL cross-linking reagent. The PDMS unit consists of platinum based catalyst to aid the curing process. The process uses a cross-linking reagent between the vinyl group and the silicone-hydrogen (Si–H) functionalities of the cross-linking reagent. 50 mL aliquots were pipetted onto the centre of clean glass cover slips and left to spread under its own weight on a flat surface. After one hour, the films were placed in an oven (120 C) for 30 minutes to complete the curing process. The films were stored in a desiccator until use. The films were placed in phosphate buffer overnight before reaction with NO gas.

Nafion 1 mM DAA solutions were prepared in 5 mL of the Nafion (Aldrich) solution (5% w=v) and 50 mL of this solution was pipetted onto a clean glass cover slip (22 mm2, BDH) and left to dry and spread under its own weight on a flat surface. Dried films were kept immersed in phosphate buffer solution overnight (10 mM, pH 7.4) before reaction with NO gas.

Results and Discussion Solution Studies Absorbance and Fluorescence Spectra Figure 1 shows the absorbance spectrum of DAA. Anthraquinone itself is virtually colourless and its

absorption spectrum consists of at least three – bands in the 240–350 nm wavelength range, with peaks being located at 252 nm, 272 nm and 326 nm. The bands at 252 nm and 326 nm are assigned to the benzenoid chromophore and the 272 nm band to the quinoid chromophore. These peaks are illustrated more clearly in Fig. 2a. A longer wavelength absorption band is also seen at 400 nm [28–32]. The absorption spectrum of monoaminoanthraquinone and diaminoanthraquinones exhibit practically the same bands as anthraquinone [28]. In the visible region of the spectrum the aminoanthraquinone derivatives show absorption bands which are due to – transition with charge transfer characteristics. The visible colour of substituted anthraquinones is dependent on the substituents ability to supply electrons by what is called a ‘‘mesomeric shift’’ to the carbonyl oxygen or the quinoid group as indicated in Fig. 3 and this is a ground state effect [33, 34]. Figure 1 shows that at low concentrations there is a weak double peak in the visible region at 640 nm and 592 nm. This ‘‘double-headed’’ absorbance peak in the visible region suggests that both NH2 groups are supplying some electrons by a mesomeric shift (Fig. 3). It was previously shown that the absorbance spectrum of 1,4-diaminoanthraquinone (1,4-DAA) also had a double-headed peak and a single head represents only one group supplying electrons by the mesomeric shift [33]. The fluorescence spectra of DAA are shown in Fig. 2. The emission and excitation spectrum are shown in Fig. 2a and b, respectively. Using 264 nm (the wavelength of maximum absorbance) the emission peak was elicited at a max of 394 nm, and using this emission wavelength an excitation spectrum was obtained with max at 232 nm. These were the excitation and emission wavelengths used throughout this study. pH Dependency and Photostability The effect of pH on the absorbance spectrum was investigated as the pH dependency often poses a problem for reagents used for NO detection in physiological samples [13, 14]. If DAA is shown to be stable at physiological pH then this reagent would be at an advantage to its predecessors. Figure 4 shows the affect of changing pH on absorbance (264 nm) using 100 mM DAA. An ANOVA test showed that the absorbance of DAA in solutions of pH 6–10 did not differ

38


Fig. 1. Absorbance spectrum of 100 mM DAA (in phosphate buffer, pH 7.4, 0.01 M) recorded before (1) and after (2) reaction with 1 mM NO

significantly (P ¼ 0.05). This suggests that this fluorescent reagent is unaffected by changes in pH around physiological pH. This is a major advantage compared to the originally synthesised DAFs [13]. The pKa value of DAA is approximately 5.5. Previous studies have established that protonation of 1aminoanthraquinone occurs at the nitrogen atom [35]. It has been proposed that the interaction of a proton from HCl with the amino group of 1-aminoanthraqui-

none results in the transition of the nitrogen atom to the pure sp2 state which prohibits the longer wavelength charge transfer characteristics due to the interaction of the lone electron pair of the nitrogen atom with the anthraquinone ring. Protonation of nitrogen should result in decolourisation of the solution and the absorption band in the ultra-violet region of the electromagnetic spectrum should be very similar to the spectrum of the anthraquinone itself in the same

39

1,2-Diaminoanthraquinone as a Reagent for Detection of NO

from sunlight prior to any experiments and the DAA stock solution was prepared daily to avoid any decomposition. NO Exposure

Fig. 2. Example of excitation (a) and emission (b) spectra of 100 mM DAA before and after reaction with NO of concentration (1) 0 mM, (2) 10 mM, (3) 20 mM, (4) 30 mM and (5) 40 mM (from top to bottom) in phosphate buffer (pH 7.4, 0.01 M, 22 C, 20 minutes reaction time) recorded with excitation 232 nm, emission 394 nm

medium [36]. The data presented in Fig. 4 suggests that this is the case as the UV absorbance intensity increases at lower pH. It is also suggested here that due to the close proximity of the nitrogen atoms in 1,2-DAA it behaves as a monoacid base, similar to 1,10-phenanthroline derivatives [37, 38] which have pKa values of 5.8. The stability of DAA in solution was assessed by monitoring the absorbance (264 nm) continuously over a one hour period. The decrease observed in absorbance was much more prominent when DAA is exposed to UV light continuously compared to ambient light in the laboratory. Throughout the study care was taken to ensure the DAA solution was protected

The effect of NO on the absorbance spectrum of DAA is illustrated in Fig. 1. There is an increase in absorbance at 340 nm but only after the addition of excess nitric oxide (1 mM). Figure 2 demonstrates that the fluorescence of DAA was quenched and that the extent of the quenching was dependent on NO concentration. Figure 5 shows the Stern-Volmer plot derived from the data presented in Fig. 2. The detection limit was evaluated to be in the order of 5 mM NO (detection limit calculated from data ¼ 5.13 mM, s=n ¼ 3). Table 1 compares the sensing characteristic of a number of well established optical reagents for detection of nitric oxide and compares them to DAA. The detection limits achieved with DAN, DAFs and DARs are generally in the nM range mainly using fluorescence imaging technique [12–14]. With respect to commercial availability and lower cost, DAA would be the preferred reagent for developing a low cost sensor. The detection limits achieved here are comparable with those achieved using wellestablished colorimetric techniques such as the Griess, the ABTS (2,20 -azinobis (3-ethylbenzthiazoline-6-sulfonic acid)) and the ferrocyanide assays [8], in the mM range. However, the lower detection limits achieved using the optical fibre based NO sensors containing fluorescent dye labelled cytochrome c0 [16] and sGC [17] were 8 mM and 1 mM, respectively. These sensors were used to directly monitor intracellular NO levels correlated with specific activities, such as phagocytosis, suggesting that DAA may also be a useful reagent for measuring NO in this environment. The average NO extracellular concentration was measured to be 210 90 mM for activated macrophages, with an intracellular NO concentration of 160 10 mM [16].

Fig. 3. Electron transfer by ‘‘mesomeric shift’’ from 1,2-NH2 substituents to the carbonyl oxygen of the quinoid group

40


Fig. 4. Absorbance (264 nm) of 200 mM DAA in solutions of pH 3–10, mean S.D., (n ¼ 3)

Fig. 5. Stern-Volmer plot for the quenching of 100 mM DAA fluorescence (ex ¼ 232 nm, em ¼ 394 nm) by nitric oxide in phosphate buffer (0.01 M, pH 7.4, 22 C), (mean S.D., (n ¼ 3))

Mechanism of Reaction It was previously reported that DAA changes from a brown dark violet to a brownish colour upon reaction with NO with a fluorescence emission 580 nm. It was also claimed that DAA is not fluorescent prior

to reaction with NO [18]. Figure 2 demonstrates that this is not the case with strong fluorescence emission being observed with UV excitation. The change in fluorescence was attributed to the formation of a triazole ring by subsequent reaction with NO and the vicinal amino groups. The main difference between this study and the previous one [18] is that fluorescence microscopy was used to study the fluorescence previously and a spectrofluorimeter was used in this case. The Stern-Volmer plot (Fig. 5) is linear suggesting that the observed quenching is either dynamic or static and not a combination of the two. To determine which of the quenching processes is being observed fluorescence lifetime measurements were measured and Table 2 shows the evaluated lifetimes of two exponential decay profiles observed with solutions purged with nitrogen and NO. The standard deviations and chi-squared ‘goodness of fit’ values are also presented. The mean lifetimes ( 1 and 2) do not differ significantly (P ¼ 0.05). This leads to two conclusions, firstly, the nitric oxide quenching mechanism of DAA is static type at the chosen wavelengths (excitation 232 nm, emission 394 nm) and, secondly, oxygen does not interfere with the static quenching of DAA by NO. Further evidence supporting that static quenching is the mechanism by which NO quenches the fluorescence of DAA is provided by the absorbance spectrum presented in Fig. 1 which indicates a ground-state reaction between NO and DAA. One possible explanation for the fluorescence quenching is that in DAA, prior to reaction with NO, the two substituent groups are electron rich resulting in electron transfer as illustrated in Fig. 3.

Table 1. Figures of merits of comparable methods for optical determination of NO Scheme

Detection

Analytical range; LODs

Comments

Refs.

Griess ABTS Ferrocyanide Oxy-hemoglobin DAN DAF

colorimetry colorimetry colorimetry colorimetry fluorescence fluorescence

14–70 mM; 14 mM 8–80 mM; 8 mM 20–200 mM; 20 mM 10–1600 nM; 80 nM 10 nM–250 nM; 9 nM 5–1000 nM; 2–5 nM

[8] [8] [8] [9] [12] [13]

DAR Fluoresceincytochrome c0 Fluorescein-sGC

fluorescence fluorimetric optical fibre sensor fluorimetric optical fibre sensor fluorescence

10 nM–10 mM; 7 nM 8–1000 mM; 8 mM

NO32 and NO22 interfere reducing species interfere NO32 and NO22 interfere oxidising agents interfere poorly soluble at neutral pH detected endothelial NO, unstable at neutral pH pH independent over large range NO2 , NO3 , O2, N2 and ascorbic acid don’t interfere sGC fragile and only useful for few hours stable at neutral pH, cost effective

DAA

1–700 mM; 1 mM 5–40 mM; 5 mM

[14] [16] [17] this method

41

1,2-Diaminoanthraquinone as a Reagent for Detection of NO Table 2. Fluorescence lifetime (ns) of 100 mM DAA solutions in phosphate buffer solutions (pH 7.4, 0.01 M) in the presence and absence of nitric oxide (94.3 ppm) and in the presence of oxygen (100%) (mean S.D., n ¼ 3)

DAA DAA and NO

1 (ns)

2 (ns)

2

2.6230 0.0418 2.6575 0.0116

11.6623 2.420 11.3617 0.3073

1.098 0.026 1.192 0.057

Following reaction with NO a triazole ring is formed which contains two electron withdrawing pyridine type nitrogens and is electron deficient. This would result in the inhibition of electron transfer from the aminosubstituents and decrease the visible absorbance. It has previously been shown that electron attracting substitutes at the 1 position of anthraquinone result in the decrease of the intensity of the UV band [34] as observed in Figs. 1 and 2. This phenomenon was also observed when using TMDABODIPY to detect NO in phosphate buffer solutions [15]. Interferents Effect of Interferents on DAA Fluorescence The effects of the interferent on fluorescence intensity of DAA were investigated. The results are presented in Table 3. The average optical intensity of the reagent solution when in the presence of the interferent was compared to that of the reagent solutions alone to determine if the interferent has a significant effect. Nitrite reacted with DAA. DAA can be used to detect nitrite in acidic solutions [39]. It was anticipated that DAA may react with the products of NO oxygenation reactions as many of the commercially available fluorescent indicators involving triazole ring formation actually interact with these NOx species instead of NO; these include compounds such as DANs, DAFs and DARs. Hydrogen peroxide causes Table 3. Effect of 100 mM of the interferent on 100 mM DAA and significance of the effect (t) (critical value ¼ 2.78, 4 degrees of freedom)

slight bathochromic shifts in the excitation and emission spectra of DAA. Bathochromic shifts in DAA spectra are normally due to dye-solvent interactions [40] suggesting that this effect may be due to intermolecular hydrogen bonding between either H2O2 and the NH2 groups or the carbonyl oxygen, or both. The observed increase in fluorescence suggests that the electron transfer from the NH2 substituents to the carbonyl oxygen has been enhanced by H2O2. This is contrary to the results previously reported which stated that H2O2 did not react with DAA [18]. Peroxynitrite did not react with DAA. Superoxide appears to substantially quench the fluorescence of DAA. Upon inspection of the fluorescence excitation spectra it appeared that this effect may be due to the interaction of DMSO with DAA. This causes a large bathochromic shift in the excitation spectrum which results in the fluorescence intensity at the specified wavelengths (excitation 232 nm, emission 394 nm) being much lower even without the addition of superoxide. This confirms that some intermolecular hydrogen bonding may occur in aqueous solutions of DAA and this can be inhibited by the addition of an aprotic solvent such as DMSO. Only a small amount of DMSO (20 mL) was added to the solution showing how receptive DAA is to small changes in solvent composition. This data renders the data presented in Tables 3 and 4 regarding the effects of superoxide, as inconclusive. Effect of Interferents on the Reaction of NO with Sensing Reagents The effect of the interferents on the reaction of 50 mM NO with the reagents was investigated. The results are presented in Table 4. The average fluorescence intensity of the reagent solution when in the presence of the interferent and NO was compared to that of the reagent solutions in the presence of NO alone. Table 4. Effect of 100 mM of the interferent on the reaction of 50 mM NO with 100 mM DAA and significance of the effect (t) (critical value ¼ 2.78, 4 degrees of freedom) 100 mM DAA þ 50 mM NO

100 mM DAA None Nitrite H2O2 ONOO O2 O2 þ SOD

349.87 22.18 264.29 4.53 492.89 14.05 317.37 6.62 10.74 0.81 21.92 0.70

– 6.55 9.43 2.43 26.47 25.60

Represents a significant interference (P ¼ 0.05).

None Nitrite H2O2 ONOO O2 O2 þ SOD

239.60 5.16 203.81 22.78 361.13 7.34 206.54 12.24 13.59 0.71 18.28 1.17

– 2.65 23.46 4.31 75.16 72.45

Represents a significant interference (P ¼ 0.05).

42

Statistically (P ¼ 0.05) nitrite did not significantly affect the reaction of NO and DAA, this suggests DAA is more selective to NO than nitrite as nitrite was previously shown to react with DAA. Although H2O2 interacts with DAA following this NO can still quench the fluorescence of the product of this reaction by about the same degree as it does on its own (31.5% quenching for NO alone and 26.6% for H2O2 and NO). This may be due to either DAA reacting with H2O2 first followed by NO or the product of the interaction of the latter two chemicals. Peroxynitrite anion interferes with the reaction of NO with DAA. As peroxynitrite did not appear to react with DAA itself it must interfere by reacting with NO to produce nitrite [41]. Superoxide again interferes with the reaction of DAA and NO. Superoxide can react with NO to produce peroxynitrite anion [42, 43]. Superoxide is in excess of NO concentration so it is suggested that as peroxynitrite does not react with DAA it is this excess which causes the interference. Further work should be carried out using DAA and alternative superoxide sources but the results presented in Tables 3 and 4 infer that superoxide will interfere with the reaction of NO with DAA. The measurement of superoxide, peroxynitrite, nitrite and hydrogen peroxide are particularly relevant to the application of the assay for measurement of NO produced by macrophages. In addition to nitric oxide, macrophages can be stimulated to produce superoxide, which either undergoes dismutation to produce hydrogen peroxide, or reacts with NO to produce peroxynitrite [44–46]. In this study much higher concentrations (100 mM) of the interferents have been used compared to that found in the actual macrophage environment. Guinea pig macrophages activated with phorbil myrisate acetate (PMA) produced a maximum of 251.5 5.6 nM of superoxide=0.25106 cells and 280.4 6.4 nM of hydrogen peroxide=0.5106 cells over a 3 hour period [44]. The rate of formation of peroxynitrite by alveolar macrophages activated with PMA was calculated to be as high as 0.11 nmol= 106 cells min1 [45]. The calculated values for accumulated concentrations of superoxide and hydrogen peroxide were found to be 30 mM and 6 mM, respectively [16]. It must be noted that the lifetime of superoxide in aqueous solutions is of the order of subseconds until it becomes concentration limited, so the amount of superoxide present at any time would be much less. However, if lower concentrations of the interferent comparable with those present in the macrophage


environment produced interference effects, then the immobilization matrix can be chosen in order to overcome this problem. In this study both silicone rubber and Nafion have been investigated. It was thought that the hydrophobic nature of silicone rubber and the cation exchange characteristics of Nafion may be useful for reducing eliminating interferents from solution and from anionic species. However, a number of other strategies have been implemented to improve selectivity of NO sensors. The use of polymeric membranes incorporating either polylysine or polypyridinium has been used to minimise interference from cationic species [47]. Alternatively, polymerised eugenol(4-allyl-2-methoxyphenol) (PE) can be used instead of Nafion to reject anionic species and it was shown that PE also had some insensitivity to some cationic species including dopamine and norepinephrine [48]. In conclusion, although the reagent DAA did suffer some interference from common oxidising species found in the macrophage environment, the concentrations used in this study where much higher than would be found in this environment. Furthermore, the use of variety of different immobilization media can minimise interference effects. Gas Exposure Studies Silicone Rubber Initial studies were concerned with attempts to immobilize 10 mM DAA solutions in silicone rubber. This resulted in large crystals appearing on the surface of the film after the curing was completed. This suggests that saturation concentration had probably been reached and the crystals are formed on the surface during the solvent evaporation due to the lower solubility of DAA in PDMS compared to the toluene= PDMS mixed solvent system. Therefore, DAA films were prepared using 1 mM DAA solutions as described earlier and under this condition, no crystals appeared on the surface of these films. Figure 6a shows the excitation and emission spectra of DAA immobilized in silicone rubber. Following immersion in phosphate buffer overnight an increase in fluorescence intensity was observed. Following this the DAA films were exposed to either NO or nitrogen gas and the response is shown in Fig. 6b. It appears from Fig. 6b that NO at a concentration of 94.3 ppm exerts a small effect on the DAA silicone rubber with a slight increase in the fluorescence

1,2-Diaminoanthraquinone as a Reagent for Detection of NO

43

Fig. 6. (a) Fluorescence spectra (excitation 445 nm, emission 500 nm) of DAA immobilized in silicone rubber before and after immersion in phosphate buffer solution (pH 7.4, 0.01 M) overnight and (b) response of DAA immobilized in silicone rubber (excitation 445 nm, emission 500 nm) exposed to NO gas (94.3 ppm, 100 mL min1 ) or N2 (100 mL min1 ) after immersion in phosphate buffer solution (pH 7.4, 0.01 M), (mean S.D., n ¼ 3)

Fig. 7. (a) Fluorescence spectra (excitation 342 nm, emission 420 nm) of DAA immobilized in Nafion before and after immersion in phosphate buffer solution (pH 7.4, 0.01 M) overnight and (b) response of DAA immobilized in Nafion (excitation 342 nm, emission 420 nm) exposed to 94.3 ppm NO or N2 (both 100 mL min1 flow rates) after immersion in phosphate buffer (mean S.D., n ¼ 3)

intensity (excitation 445 nm, emission 500 nm), however, this response was not significantly different to that observed with nitrogen alone (P ¼ 0.05, NO: 1.0117 0.0423 Au, Nitrogen: 0.9899 0.0398 Au (mean S.D., n ¼ 3)). Further work was then carried out using nafion as a potential immobilization medium.

films is in contrast to the colour of DAA in solution and in silicone rubber which appeared to be light purple=pink. It has already been proposed that this purple=pink colour is due to the charge transfer characteristics of the diamine substituents at the 1 and 2 position of the anthraquinone [33, 34]. Figure 8

Nafion The fluorescence spectrum of DAA immobilized in nafion is shown in Fig. 7a before and after immersion in phosphate buffer. The effect of NO exposure on the DAA films is presented in Fig. 7b. There was a small difference in the response of the films to nitrogen and NO gas. However, this difference in response was not significant (P ¼ 0.05, NO: 1.0114 0.0045 Au, Nitrogen: 1.0230 0.0382 Au (mean S.D., n ¼ 3)). The DAA nafion films appeared to be a brown=yellow colour. This suggests that the structure of immobilized DAA may differ from that in solution. The brown colouring of the

Fig. 8. Absorbance spectra of DAA immobilized in Nafion after casting the solution on the glass cover slip recorded immediately (1) and then at 30 minute (2) and 60 minute (3) periods.

44

shows the change in the absorbance spectrum of the nafion solution as the solvent evaporates. The absorbance of the DAA nafion films changes considerably as the films dry. The initial cast solution was a bright pink colour as indicated by an absorbance maximum at 530 nm. As the films dried the colour changed to a brown colour as indicated by the decrease in absorbance at 530 nm and the shifting of the absorption peak to a maximum at about 350 nm. This color change was predicted to occur when NO reacted with DAA [18]. This suggests that the immobilization of DAA in Nafion has resulted in a structural change which involves the diamine moiety thus inhibiting the interaction of NO gas. Although the initial immobilization studies were unsuccessful in this case, due to promising results obtained in solution, it is suggested that further work using alternative immobilization methods, as suggested in the ‘Interferents’ section, would be worthwhile. The initial solution studies suggest that such a sensor would be useful for measuring macrophage NO concentrations. Conclusions Investigations in aqueous solutions have shown that NO quenches the fluorescence of DAA via a static quenching mechanism in phosphate buffer solutions (pH 7.4). The detection limit of NO determination using DAA was 5.13 mM with high linearity in response (r ¼ 0.9956) in the range 0–40 mM. Peroxynitrite anion did not interfere with DAA fluorescence but superoxide, H2O2 and nitrite did. Both H2O2, peroxynitrite and superoxide interfere with the reaction of the complex with NO by scavenging analyte species. In the investigations involving the immobilization of DAA the fluorescence of DAA was not significantly perturbed by NO gas exposure compared to nitrogen exposure (P ¼ 0.05). Spin coating can be used for thinner film preparations [49]. The preparation of thinner silicone rubber films should improve the response times to NO. Further investigations could be carried out using alternative immobilization media. Sol–gel is a suitable candidate for such investigations. These inorganic supports offer several advantages over organic polymer supports since they provide physical rigidity and little swelling in solution, easy entrapment of the dye and are chemically inert, however, DAN immobilized in sol–gel showed no reaction


with NO due to direct interaction between the amine groups of DAN and the hydroxyl groups located in the porous surface of the sol–gel matrix [50]. This is envisaged to be the same case for DAA. A number of other immobilization media could be considered for the production of thin films as it has previously been shown that thin films produce fast response and reversal times. For instance PVC (poly(vinyl chloride)) and PVP (poly(vinyl pyrrolidone)) have good mechanical properties and are simple to prepare and optically transparent. The results presented here suggest that upon successful immobilization of DAA an economically attractive sensor could be produced to detect macrophage NO. Acknowledgements. The authors acknowledge the Engineering and Physical Research Council for the Ph.D. studentship support to H.D.

References [1] [2] [3] [4] [5] [6] [7] [8]

[9] [10] [11] [12] [13]

[14] [15] [16] [17] [18] [19] [20] [21] [22]

Furchgott R F, Zawadzki J V (1980) Nature 288: 373 Culotta E, Koshland D E Jr (1992) Science 258: 1862 Shibuki K (1990) Neurosci Res 9: 69 Malinski T, Taha Z (1992) Nature 358: 676 Kalyanaraman J J B, Hyde J S (1993) Biochem Biophys Res Commun 192: 926 Funazo K, Tanaka M, Shono T (1980) Anal Chim Acta 119: 291 Kikuchi K, Nagano T, Hayakawa H, Hirata Y, Hirobe M (1993) Anal Chem 65: 1794 Nims R W, Cook J C, Krishna M C, Christodoulou D, Poore C M B, Miles A M, Grisham M B, Wink D A (1996) Meth Enzymol 268: 93 Privat C, Lantoine F, Bedioui F, Millanvoye van Brussell E, Devynck J, Devynck M-A (1997) Life Sci 61: 1193 Nagano T (1999) Luminescence 14: 283 Yao D, Vlessidis A G, Evmiridis N P (2004) Microchim Acta 147: 1 Miles A M, Wink D A, Cook J C, Grisham M B (1996) Meth Enzymol 268: 105 Nakatsubo N, Kojima H, Kikuchi K, Nagoshi H, Hirata Y, Maeda D, Imai Y, Irimura T, Nagano T (1998) FEBS Lett 427: 263 Kojima H, Hirotani M, Urano Y, Kikuchi K, Higuchi T, Nagano T (2000) Tetrahedron Lett 41: 69 Zhang X, Wang H, Li J-S, Zang H-S (2003) Anal Chim Acta 481: 101 Barker S L R, Clark H A, Swallen S F, Kopelman R (1999) Anal Chem 71: 1767 Barker S L R, Zhao Y, Marletta M A, Kopelman R (1999) Anal Chem 71: 2071 Heiduschka P, Thanos S (1998) Neuroreport 9: 4051 Brook T E, Narayanaswamy R (1997) Sens Actuators B 38–39: 195 Keefer L K, Nims R W, Davies K M, Wink D A (1996) Meth Enzymol 268: 281 Hughes M N, Nicklin H G (1971) J Chem Soc (A) 64 Yang X-F, Guo X-Q, Zhao Y-B (2002) Talanta 57: 883

1,2-Diaminoanthraquinone as a Reagent for Detection of NO [23] [24] [25] [26] [27] [28] [29] [30] [31]

[32] [33]

[34]

[35] [36]

Valentine J S, Curtis A B (1975) J Am Chem Soc 97: 224 Lokesh B R, Cunningham M L (1986) Toxicol Lett 34: 75 Fee J A, Hildenbrand P G (1974) FEBS Lett 39: 79 Kim S, DiCosimo R, Fillpo J S Jr (1979) Anal Chem 51: 679 Gampp H, Lippard S J (1983) Inorg Chem 22: 357 Borst H-U, Kelemen J, Fabian J, Nepras M, Kramer H E A (1992) J Photochem Photobiol A 69: 97 Yoshida Z, Takabayashi F (1968) Tetrahedron 24: 913 Inoue H, Hida M, Nakashima N, Yoshihara K (1982) J Phys Chem 86: 3184 Ritter J, Borst H-U, Lindner T, Hauser M, Brosig S, Bredereck K, Steiner U E, Kuhn D, Kelemen J, Kramer H E A (1988) J Photochem Photobiol A: Chem 41: 227 Srivatsavoy V J P, Venkataraman B, Periasamy N (1992) J Photochem Photobiol A: Chem 68: 168 Kesswani H I (1963) Some properties of hydroxyl and amino derivatives of anthraquinones. PhD Thesis, UMIST, Manchester, UK Uffindell N D (1964) The purification and photochemistry of some anthraquinone dyes. PhD Thesis, UMIST, Manchester, UK Etaiw S H, Abou-Sekkina M M, El-Hefnawey G B, Assar S S (1982) Can J Chem 60: 304 Kratochvil Y, Nepras M (1972) Coll Czechoslovak Chem Commun 37: 1533

45 [37] Siegel G G, Zeegers-Huyskens T H (1989) Spectrochim Acta (A) 45: 1297 [38] Schilt A A, Dunbar W E (1974) Tetrahedron 30: 401 [39] Centelles J J, Fernandez-Cancio M, Imperial S (2003) Anal Lett 36: 2139 [40] Umavedi M, Ramakrishan V (2003) J Raman Spectro 34: 13 [41] Sigel H, Sigel A (1999) Metal ions in biological systems. Dekker, New York, p 548 [42] Blough N V, Zafiriou O C (1985) Inorg Chem 24: 3502 [43] Nauser T, Koppenol W H (2002) J Phys Chem A 106: 4084 [44] Pick E, Mizel D (1981) J Immunol Meth 46: 211 [45] Ischiropoulos H, Zhu L, Beckman J S (1992) Arch Biochem Biophys 298: 446 [46] Ding A H, Nathan C F, Stuehr D J (1988) J Immunol 141: 2407 [47] Mitchell K M, Michaelis E K (1998) Electroanalysis 10: 81 [48] Ciszewski A, Milczarek G (1998) Electroanalysis 10: 791 [49] Brook T E (1998) Developments leading to optical fibre based gas and vapour sensors. PhD Thesis, UMIST, Manchester, UK [50] Esquembre R, Pastor I, Tormo C, Mallavia R, Mateo C R (2004) Fluorescent biosensor for nitric oxide by insertion of 2,3-diaminonaphthalene in liposomes immobilized in sol–gel glasses. Book of abstracts. Seventh European Conference on Optical Chemical Sensors & Biosensors, EUROPT(R)ODE VII: 93