A Bifunctional Luminogenic Substrate for Two Luminescent Enzymes: Firefly Luciferase and Horseradish Peroxidase

Analytical Biochemistry 271, 159 –167 (1999) Article ID abio.1999.4114, available online at http://www.idealibrary.com on

A Bifunctional Luminogenic Substrate for Two Luminescent Enzymes: Firefly Luciferase and Horseradish Peroxidase T. Sudhaharan and A. Ram Reddy 1 Department of Chemistry, Nizam College, Hyderabad-500001, Andhra Pradesh, India

Received December 11, 1998

Horseradish peroxidase (HRP) catalyzes the oxidative chemiluminescent reaction of luminol, and firefly luciferase catalyzes the oxidation of firefly D-luciferin. Here we report a novel substrate, 5-(5*-azoluciferinyl)2,3-dihydro-1,4-phthalazinedione (ALPDO), that can trigger the activity of HRP and firefly luciferase in solution because it contains both luminol and luciferin functionalities. It is synthesized by diazotization of luminol and its subsequent azo coupling with firefly luciferin. NMR spectral data show that the C5* of benzothiazole in luciferin connects the diazophthalahydrazide. The electronic absorption and fluorescence properties of ALPDO are different from those of its precursor molecules. The chemiluminescence emission spectra of the conjugate substrate display biphotonic emission characteristic of azophthalatedianion and oxyluciferin. It has an optimum pH of 8.0 for maximum activity with respect to HRP as well as luciferase. At pH 8.0 the bifunctional substrate has 12 times the activity of luminol but has 7 times less activity than the firefly luciferin–luciferase system. The specific enhancement of light emission from the cyclic hydrazide part of ALPDO helped in the sensitive assay of HRP down to 2.0 3 10 213 M and of ATP to 1.0 3 10 214 mol. Addition of enhancers such as firefly luciferin and p-iodophenol (PIP) to the HRP–ALPDO–H 2O 2 system enhanced the light output. © 1999 Academic Press Key Words: bifunctional substrate; firefly luciferase; horseradish peroxidase; luciferin; luminol.

Recent advances in biotechnology, especially in the areas of immunology and DNA diagnosis, are basically results of the development of various analytical methods for detecting target molecules. Although various 1

To whom correspondence should be addressed at Department of Chemistry, Nizam College, Osmania University, Hyderabad-500001, A.P., India. Fax: 91-40-717 3882. E-mail: [email protected]. 0003-2697/99 $30.00 Copyright © 1999 by Academic Press All rights of reproduction in any form reserved.

techniques have been introduced for molecular recognition, isotopic labeling remains the most widely exploited technique. Because of the requirement for specialized handling skills in the usage of isotopic probes during experiments and their subsequent disposal, the search for simpler alternatives is constant. Among these, the chemiluminescent, bioluminescent, and fluorescent methods are considered to have the most potential for detection and assay of various enzymes, cofactors, and their metabolites (1). A number of chromogenic and fluorogenic substrates are in use at present for the determination of unmodified enzymes (2). The sensitivity, speed, and specificity of the two enzymes firefly luciferase (3) and horseradish peroxidase (HRP) 2 (4), which catalyze conversion of chemical energy to light energy via the formation of an activated chemical structure whose decay produces light, are comparable to those of radiometric assays. Chemiluminescent systems based on HRP are widely used for immunoassays (5– 8). HRP catalyzes the H 2O 2 oxidation of luminol (5-amino-2,3-dihydro-1,4-phthalazinedione) (I; see Structure 1) to produce a luminol radical (9). This radical then forms an endoperoxide that, upon decomposition, yields an electronically excited 3-aminophthalate dianion, emitting light on its return to the ground state. Firefly luciferase catalyzes the oxidative decarboxylation of 4,5-dihydro-2-(69-hydroxybenzothiozolyl)-4thiazole carboxylic acid (D-LN) (II) in the presence of ATP. Firefly bioluminescence involves the incorporation of oxygen into LN to give a dioxetanone intermediate (10, 11) which is spontaneously cleaved to form excited oxyluciferin (OLN). The excited OLN undergoes radiative decay to the ground state, emitting at ;565 nm at neutral pH (12). The firefly LN–luciferase 2

Abbreviations used: ALPDO, 5-[59-azoluciferinyl-2,3-dihydro1,4-phthalazinedione; LN, luciferin; HRP, horseradish peroxidase; OLN, oxyluciferin; PIP, para-iodophenol; DMSO, dimethyl sulfoxide. 159

160

SUDHAHARAN AND REDDY

STRUCTURE 1

reaction is the best method for rapid, sensitive measurement of ATP and other nucleotides by coupled bioluminescence (13), providing a linear assay over 5– 6 orders of magnitude down to 10 216 mol. Several publications aimed at improving the efficiency and elucidating the chemiluminescent reaction mechanism of luminol and firefly LN (14 –17), respectively, have appeared in recent years. The chemiluminescent active functional moiety in luminol is the hydrazide group, while in LN the reactivity originates from the unsubstituted 5 position of the thiazole ring. White and Rosewell (18) synthesized the dehydroluciferin monoacylhydrazide (III), which combined the above two active functional groups, and found that the chemiluminescence resulted from the excited carboxylate anion via an azadioxetane intermediate. Because of their poor solubility in aqueous medium and their low acidity, these luciferyl hydrazides have not found many applications in water-based chemiluminescent systems. On the other hand, Whitehead and co-workers discovered that peroxidase catalyzed enhanced luminescence of luminol severalfold in the presence of

firefly LN and other 6-hydroxy benzothiazoles (19). Thus, the use of chemiluminescence enhancers such as firefly LN and PIP improved the detection limit of peroxidase making possible the development of simple nonisotopic immunoassay systems. To explain the potential of benzothiazoles as chemiluminophores, Susamoto et al. (20) designed a cyclic hydrazide, 4-(59hydroxybenzothiazolyl)phthalyl hydrazide (IV), a hybrid of LN and luminol. This is the first substrate that has proved to serve as both a fluorogenic and a chemiluminogenic substrate. Similarly, Klel and Obreen (21) observed that diazoluminol, derived from 3-amino-Ltyrosine and luminol, increases the luminescence of the latter 5- to 20-fold. We have coupled luminol and firefly D-LN by the diazo group and obtained a novel bifunctional substrate, 5-(59-azoluciferinyl)-2,3-dihydro-1,4phthalazinedione (ALPDO) (V), which can bind to both HRP and firefly luciferase. It is stable and water-soluble, and the D-LN part is not racemized and has maximum activity at pH 8. Hence, we considered it worthwhile to use ALPDO as a chemiluminescent substrate. To our knowledge, this report is the first of its kind for

161

BIFUNCTIONAL ENZYME SUBSTRATE TABLE 1

NMR and Elemental Analysis Data of ALPDO Elemental analysis NMR a (in DMSO D6) 8.58 7.95 7.42 4.32 3.63 10.8 a

(s, 1H), 8.03 (d, 1H) (d, 1H), 7.46 (s, 1H) (s, 1H), 7.12 (d,d, 1H) (s, 2H, broad), 6.72 (t, 1H) to 3.75 (m, 2H) (offset region, 1H)

Element

Theoretical (%)

Experimental (%)

C H N O S

50.44 2.65 18.59 14.16 14.16

50.38 2.60 18.76 — —

d, doublet; s, singlet; t, triplet; d,d, double doublet; m, multiplet.

single-reaction-vial detection of two biologically active enzymes, HRP and firefly luciferase. MATERIALS AND METHODS

Materials. Firefly D-luciferin and luminol were purchased from Sigma–Aldrich (U.S.A.) and HRP was purchased from Bangalore Genei Pvt. Ltd. (India). North American firefly luciferase was obtained from Amersham, as the ATP bioluminescent assay RPN1630 kit. ATP was obtained from Boehringer-Mannheim. H 2O 2, NaNO 2, p-iodophenol, and HCl were obtained from E. Merck (India) as AR grade and used without any further purification. The original stock solutions of LN/ luciferase were made in 50 mM Tricine buffer containing 10 mM MgCl 2 and 1 mM EDTA at pH 8.0, whereas the stock solutions of ALPDO, luminol, and HRP were made in 0.1 M Tris–HCl (pH 8.0) and stored at 280°C. The working stock solutions were prepared on the day of use by diluting in the respective buffers. Synthesis of ALPDO. Ten milligrams (0.033 mmol) of LN and 6.3 mg (0.033 mmol) of luminol were chilled in ice. Luminol was dissolved in a minimum amount of water. Once the temperature was less than 3°C, 100 ml of 1.2 N HCl was added, drop by drop, with the temperature maintained below 3°C. Meanwhile, 3.1 mg (0.039 mmol) of NaNO 2 was dissolved in a minimum amount of H 2O, chilled, and added to the precooled luminol solution without a rise in the temperature. The excess HNO 2 was tested with KI-starch paper and destroyed by urea solution. The diazonium salt obtained was kept in ice for about 15 min. About 4 mg (0.1 mol) of NaOH was dissolved in a minimum of H 2O and added to the ice-cooled LN. Finally, the chilled diazonium salt solution was added to the alkaline LN slowly with stirring. After complete addition, it was allowed to stand at low temperature for about 1 h. The precipitated ALPDO was separated by centrifugation, washed with chilled dilute HCl and water, and vacuum-dried. It has an R f value of 0.5 in CHCl 3:CH 3OH (90:10, v/v), while luminol and LN have R f values of 0.1 and 0.15,

respectively. The compound decomposed around 340°C. The elemental analysis and NMR data are given in Table 1. Spectral studies. The UV–vis spectra of ALPDO were obtained using a Hitachi 150 UV–vis spectrophotometer in water. The corrected fluorescence spectra of ALPDO were obtained using a Hitachi 4010 spectrofluorimeter with 5-nm entrance and exit slit widths. The concentration of ALPDO used was 1 3 10 26 M and the excitation wavelength was 360 nm. The NMR spectrum was obtained with a Bruker WH 300 spectrometer by dissolving 5 mg of ALPDO in DMSO D6. Elemental analysis data were obtained using a Perkin–Elmer 240C analyzer. Bioluminescence data were obtained using a LKB Luminometer 1250. The chemiluminescence emission spectra of the substrate were obtained using an Hitachi 4010 Spectrofluorimeter by closing the entrance slit and keeping the exit slit wide open (15 nm) in energy mode. The chemiluminescence reaction was triggered for the spectrum by injecting ATP (20 pmol), H 2O 2 (3 3 10 27 M), or both into cuvettes containing premixed solutions of enzymes (5.6 3 10 26 M HRP and 4.3 3 10 26 M luciferase) and substrate ALPDO (2.4 3 10 23 M) in 50 mM Tricine buffer containing 10 mM MgCl 2 and 1 mM EDTA at pH 8.0. The signals emitted from the chemiluminescence system were scanned immediately. Optimization of pH for ALPDO. The activity of ALPDO at various pH ranges was evaluated by fixing H 2O 2 (6.0 3 10 25 M), HRP (5.6 3 10 28 M), ALPDO (8.5 3 10 25 M), luciferase (4.3 3 10 26 M), and ATP z Mg 21 (1.25 3 10 213 mol), and varying buffers of different pH levels. The total reaction volume was made up to 400 ml using the respective buffers. The buffers used to obtain a pH between 7.0 and 12 were 0.1 M Tris–HCl buffer and 0.1 M glycine/NaOH buffer. The mixing was done with an external mixer and readings were taken after a 5-s delay. The standard reactions for LN and luminol at various pH ranges were also carried out

162

SUDHAHARAN AND REDDY

FIG. 1. (A) Absorption spectrum of ALPDO (2 3 10 24 M) in H 2O (OD scaled down by two times). (B) Fluorescence spectrum of ALPDO (1 3 10 26 M) in H 2O l ex at 360 nm.

under identical conditions employing the same buffers. The activity of the analyte was plotted against the pH. HRP estimation using ALPDO. The HRP assay was carried out by keeping ALPDO (8.5 3 10 26 M) and H 2O 2 (6.0 3 10 25 M) at the optimum pH of 8.0 and varying the concentration of HRP. The reaction was initiated by the addition of HRP to the assay mix, and the reaction volume was kept at 400 ml using 0.1 M Tris–HCl buffer. A control assay was also carried out using an identical concentration of luminol as the standard for comparison with ALPDO. ATP assay using ALPDO. The ATP estimation was carried out by injecting various amounts of ATP concentration into the reaction mixture containing firefly luciferase (4.3 3 10 26 M) and ALPDO (1.77 3 10 24 M) in 50 mM Tricine buffer containing 10 mM MgCl 2 and 1 mM EDTA at pH 8.0. The total reaction volume was 200 ml. A standard assay was carried out for LN under identical conditions for comparison. Enhanced chemiluminescence. The experiment was performed by adding various concentrations of LN/PIP to the reaction mixture containing ALPDO (8.5 3 10 25 M), HRP (5.6 3 10 210 M), and H 2O 2 (6.0 3 10 25 M) in 0.1 M Tris–HCl buffer at pH 8.0. The total reaction volume was maintained at 400 ml. A control experiment was also carried out using luminol as the standard. RESULTS AND DISCUSSION

Azobenzene has rich possibilities for various technological and fundamental studies. It is a photochemically reactive chromophore. Coulombic interaction between the two polar groups in ALPDO may stabilize the structure and prevent its decomposition (22). The new substrate is not only stable but also bioactive, and the prepared stock solutions display profound in vitro bioluminescence even after 2 months, when stored at 280°C.

Electronic absorption, fluorescence, and chemiluminescence properties of the ALPDO. The absorption spectrum of ALPDO in water (pH ;7.2) is shown in Fig. 1A. It differs markedly from the absorption spectra of its corresponding precursor components. It exhibited three electronic transitions at 255, 288, and 360 nm in water. The longer and shorter wavelength absorption maxima have more intensity than the 288-nm absorption maximum and have A 360/A 288 5 2.66 and A 255/A 288 5 3. Following Griffths (23), the 288-nm absorption maxima can be attributed to the syn azo form and are probably stabilized by the intramolecular hydrogen bonding between 69-hydroxyl hydrogen of LN and diazo nitrogen. The fluorescence spectrum of the bifunctional substrate in water is shown in Fig. 1B. Note that when excited at 360 nm it gave a broad emission doublet maxima centering at 542 nm. The emission maxima obtained resemble the emission spectrum of LN, which exhibits an emission maximum at 536 nm in water. However, compared to LN, ALPDO is relatively less fluorescent. The 6-nm red shift in the fluorescence maximum and the decreased fluorescence quantum yield of ALPDO over LN are probably due to the intramolecular charge transfer from donor luciferyl part to the acceptor luminol moiety. The large Stokes shift (;175 nm) and the nonintersection of the absorption and emission bands clearly support a role for ALPDO as a fluoroprobe. As expected in phenols, the fluorescence intensities in ALPDO display pH dependence. Increase in pH increases the fluorescence intensity, reaching a maximum at pH 8 –9. Although the compound contains both the luminol and LN moieties, it does not exhibit any wavelength-dependent emission spectrum, revealing that the two moieties are in one plane, with extensive delocalization of p electrons from one part of ALPDO to another taking place via the azo bridge. The chemiluminescence emission spectrum of the

163

BIFUNCTIONAL ENZYME SUBSTRATE

FIG. 2. Chemiluminescence emission spectra of ALPDO (the noisy chemiluminescence emission spectra are converted into smooth curves) in 50 mM Tricine buffer, pH 8.0, containing 10 mM MgCl 2 and 1 mM EDTA. —, ALPDO (2.4 3 10 23 M) in the presence of ATP z Mg 21 (20 pmol) 1 luciferase (4.3 3 10 26 M) 1 HRP (5.6 3 10 26 M) 1 H 2O 2 (3 3 10 27 M). z z z , ALPDO (2.4 3 10 23 M) in the presence of ATP z Mg 21 (20 pmol) 1 luciferase (4.3 3 10 26 M). - - -, ALPDO (2.4 3 10 23 M) in the presence of HRP (5.6 3 10 26 M) 1 H 2O 2 (3 3 10 27 M).

substrate is shown in Fig. 2. It can be seen from the figure that when the reaction was initiated by injection of ATP into the equilibrated mixture of firefly luciferase–ALPDO, a luminescence maximum at 554 nm was observed, whereas when it was triggered by injection of H 2O 2 (3.0 3 10 27 M) into the ALPDO–HRP mixture a luminescence maximum was observed at 448 nm. The chemiluminescence maxima of ALPDO are slightly different from the chemiluminescence maxima of firefly LN and luminol (Table 2), when obtained with their respective enzymes. The difference in chemiluminescence maxima of the ALPDO can be attributed to the differing light-emitting species shown in Scheme 1. In the case of firefly LN–luciferase bioluminescence, the light-emitting species is the excited OLN, while in luminol–HRP it is the excited 3-aminophthalate dianion. When ALPDO reacted with H 2O 2 in the presence of HRP, the light-emitting species is 3-(59-luciferinyl) azophthalate dianion (VI), whereas when it reacted with ATP, the light-emitting species is 59-(39-luminol)-azo oxyluciferin (VII). Interestingly, the azo conjugate substrate exhibits two chemiluminescence emission maxima, one at 448 nm and other at 554 nm, in the presence of both luciferase and HRP enzymes when the chemiluminescent reaction is initiated with a mixture of ATP and H 2O 2. The biphotonic emission at two wavelengths that are marginally shifted from the emission maxima of their respective natural substrates reveals that ALPDO can bind either to one or to both of the enzymes. This is in contrast to the fluorescence spectra, where only emission maxima characteristic of LN appeared. The in vitro bioluminescence spectral intensity of the azophthalate dianion moiety far exceeds the bioluminescence intensity of the azo OLN moiety, depending

on the order of mixing of the reagents. A similar observation was made in case of phthalahydrazide (24). When phthalahydrazide is oxidized to excited phthalate anion, the latter is practically nonfluorescent. However, when coupled by a methylene bridge to either N-methylacridone or diphenylanthracene, the chemiluminescence of phthalahydrazide increased substantially as a result of intramolecular energy transfer. The bioluminescent intensity of ALPDO at 554 nm is always less than the bioluminescent intensity of LN at 550 nm. It was noted in the bioluminometer that once one of the moieties of the substrate is oxidized to the light-emitting part, the light emitted by its counterpart in the second stage, with the addition of H 2O 2/ATP, is always less than the light emitted in the first stage. For example, when the ALPDO is first oxidized by H 2O 2 in the presence of HRP the light output given by the luminol moiety is 7710 mV. But when the same reaction is carried out after the oxidation of the LN moiety, the light output is only 2270 mV, under similar experimental conditions, after the blank correction (Scheme 1). Similarly, when the LN part of the bifunctional substrate was first oxidized, with the injection of ATP, the light output was 19.5 mV. The same reaction yielded only 5 mV when it occurred in the second stage. This is probably because ALPDO binds to both of the enzymes sequentially, preferring HRP, and the partially oxidized substrate can bind to the first enzyme so that it can neither perfectly orient nor be completely available to the active site of the second enzyme. Optimization of pH. Although HRP is stable in the pH range 4.2–12.0, its activity in the soluble state is at a maximum at pH 7.0 and greatly reduced under alkaline conditions. But luminol chemiluminescence is most efficient between pH 10 and 13. This limits its use in biology. However, the optimum pH for maximum luciferase activity is 7.8 at 25°C. Hence, one of the important properties of the new substrate is its lightemitting ability with respect to the two native subTABLE 2

Absorption and Emission Maxima of LN, Luminol, and ALPDO in H 2O Compound D-Luciferin

Luminol ALPDO

a b

l max (nm) (e, mol 21dm 2) 260 (7079) 335 (18197) 305 (7530) 340 (8190) 255 (6470) 288 (2157) 360 (5798)

l emi. flu z (nm) a

l emi. CL (nm) b

536

550

424

425

542

554, 448

Fluorescence emission maxima. Chemiluminescence emission maxima.

164

SUDHAHARAN AND REDDY

SCHEME 1

strates vis-a-vis its pH dependency. In Fig. 3 the activity of luciferase and HRP with their natural substrates and ALPDO against pH is presented. Examination of the figure reveals two things. First, the optimum pH for maximum activity of HRP shifts from pH 10 with luminol to pH 8.0 with ALPDO, while there is no change in the optimum pH for the luciferase–ALPDO system, which retained maximum activity at pH 8.0, as in the case of the luciferase–LN system. Second, with the bifunctional substrate the activity of HRP increased 12 times and that of luciferase decreased 7 times compared to luminol and LN as substrates. This can be attributed to the intramolecular synergistic effect of LN on the light emission from the peroxidasecatalyzed oxidation of luminol moiety. Hence, the diazo phthalazinedione moiety is more chemiluminescent than the luminol moiety. The results obtained reveal that when two different substrates are engineered into one entity, with a diazo bridge, the new synthon can still retain the chemiluminescence properties of the two components, although they are affected differently. Thus, two different enzymes can be assayed with a single substrate. After ascertaining the HRP and luciferase discriminating ability of ALPDO, we have evaluated the specificity factor, f, of the substrate toward the two enzymes employing the relation (25)

f5

n HRP 3 @luciferase# , n luciferase 3 @HRP#

where n HRP and n luciferase are the chemiluminescent light intensities obtained at equal concentrations of HRP and luciferase, under optimal conditions, and assuming that the two enzymes behave well under steadystate conditions, obeying Briggs–Haldane kinetics (26), that is, substrate binding being faster than subsequent stages. The specificity factor obtained is 60,720, and this value shows that ALPDO has a profound preference to HRP over luciferase. The high specificity of ALPDO chemiluminescence toward HRP would allow sensitive detection of the latter and its metabolites. The enhanced chemiluminescence of ALPDO with HRP at pH 8.0 has greater significance, because it is a chemiluminogenic probe. Thus, the biological pH compatible with maximum activity offers the possibility of estimating a variety of luminol-related metabolites, cofactors, enzymes, etc., under physiological conditions, as it has emission at longer wavelengths, has a large Stokes shift, and takes a long time to reach peak height in chemiluminescence. The light yield of the coupled derivative was also linear over the range of concentrations. Chemiluminescence assay of HRP. HRP is one of the most important and smallest enzymes (M r ;44 K) used as a label. The chemiluminescent luminol and isoluminol are its known substrates and are used to measure its activity. As ALPDO proved to be a better substrate for HRP than luminol, it is of profound significance if it can be employed to assay the activity of HRP following the same principle of HRP–luminol, i.e.,

BIFUNCTIONAL ENZYME SUBSTRATE

165

ATP and ALPDO and emits light. At very low ATP concentrations the chemiluminescence decay is much slower as the potential quencher, OLN, forms at a slower rate. In Fig. 5 a typical ATP standard curve employing ALPDO as a firefly luciferase substrate is shown. Addition of ATP up to 20 3 10 212 mol exhibited linearity in increase of the light output. The Michaelis– Menten constant, K m, estimated for ALPDO is 8.25 6 (0.25) 3 10 25 M for luciferase at pH 8.0 in Tricine buffer. From Fig. 5 it can be inferred that although it is less sensitive than LN, ALPDO can be used as a substrate for firefly luciferase and can be employed to estimate ATP. Using ALPDO it is possible to detect ATP in the femtomolar range. Effect of PIP. Light emission from luminol–H 2O 2– peroxidase systems can be greatly enhanced by the addition of substituted phenols, p-hydroxycinnamic acid, benzothiazoles, and LN (8). The underlying mechanism of this enhanced chemiluminescence is not completely understood. The most probable explanation is that these chemiluminescence enhancers can acceler-

FIG. 3. Optimization of pH for maximum activity of ALPDO in comparison with luminol and LN in varied (7–12) pH buffers. (a) ALPDO (8.5 3 10 25 M) 1 HRP (5.6 3 10 26 M) 1 H 2O 2 (6.0 3 10 26 M). (b) Luminol (8.5 3 10 25 M) 1 HRP (5.6 3 10 26 M) 1 H 2O 2 (6.0 3 10 26 M). (c) ALPDO (1.77 3 10 24 M) 1 ATP (1.25 3 10 212 mol) 1 luciferase (4.3 3 10 26 M). (d) LN (1.77 3 10 24 M) 1 ATP (1.25 3 10 212 mol) 1 luciferase (4.3 3 10 26 M).

under the constant but excess concentration of ALPDO and H 2O 2 (#3 mM). The amount of light produced by the photon-emitting final product showed linearity with the HRP concentration. From Fig. 4 it can be seen that ALPDO exhibits linearity from 0.84 to 8.4 3 10 28 M of HRP. Owing to improved signal-to-noise ratio and higher light output, the detection of HRP reached 2 3 10 213 M with 1.6 3 10 24 M ALPDO and 1.5 3 10 24 M H 2O 2, while it was detected up to 1 3 10 213 M with enhanced chemiluminescence of luminol substrate (27). The Michaelis–Menten constant K m calculated for ALPDO is 1.2 6 (0.2) 3 10 25 M for HRP at pH 8.0 in 0.1 M Tris–HCl buffer. ATP assay. Firefly LN–luciferase bioluminescence is a unique technique for accurate and sensitive assay of cellular ATP as well as ATP in solutions (28). As ALPDO is a 59-azo phthalazine derivative of firefly LN and exhibits the activity of LN, it is considered of value to estimate ATP based on its oxidative chemiluminescence reaction, catalyzed by firefly luciferase. The procedure adopted is similar to that for firefly LN–luciferase bioluminescence, wherein the single enzyme uses

FIG. 4. Estimation of HRP activity using ALPDO. ALPDO (8.5 3 10 25 M) and H 2O 2 (6.0 3 10 25 M) at pH 8.0 in 0.1 M Tris–HCl buffer.

166

SUDHAHARAN AND REDDY

FIG. 5. Estimation of ATP employing a ALPDO–luciferase system. ALPDO (1.77 3 10 24 M) 1 luciferase (4.3 3 10 26 M) in 50 mM Tricine buffer, pH 8.0, containing 10 mM MgCl 2 and 1 mM EDTA.

ate one or more of the oxidation steps to generate luminol radicals during the complex reaction pathway of the enzymatic oxidation of luminol (29). We have also made a similar attempt to improve the HRP assay with ALPDO using PIP and firefly LN as chemiluminescence enhancers. The procedure involves the simultaneous use of PIP and ALPDO in the presence of oxidant, H 2O 2, and HRP at pH 8.0. The total light

emission in the chemiluminescent peroxidation of ALPDO in the presence of PIP is more than that in its absence, as shown in Table 3. However, if the PIP concentration is more than 2.5 3 10 24 M, then it acts as a quencher. The reason for quenching of the HRP activity by PIP is not immediately known. The probable reason given for the activity not being enhanced at higher concentrations is that at pH 8.0 ionization of PIP to phenoxyiodophenol may not be occurring, thereby not mediating electron transfer between HRP and ALPDO. The effect of PIP on the ALPDO–H 2O 2– HRP system has not yielded a satisfactory result although activity was enhanced 6.5 times at #2.5 3 10 24 M PIP. Effect of LN. Whitehead and collaborators (7) made the unexpected discovery that addition of firefly LN to luminol–HRP–H 2O 2 not only enhanced the peroxidasecatalyzed chemiluminescence severalfold but also reduced the chemical blank from luminol and H 2O 2. In Table 4 the effect of LN on luminol–ALPDO with and without HRP is shown. It can be observed from the table that LN enhances ALPDO–H 2O 2–HRP chemiluminescence but not to the extent found with the luminol–H 2O 2–HRP system. It is now clear that the molecules that provoke light output from luminol when linked to it by a covalent bond no longer act as efficient radical accelerators. For example, firefly LN enhanced the chemiluminescence of luminol more than 10,000 times, but when it was coupled covalently to luminol as in diazo phthalazinedione, the enhancement is less than 10,000 times. This can be explained due to the presence of the amino group in luminol. Amines and phenols are good electron donors. Like transition metal

TABLE 3

Comparison of Luminol, Luciferin, and ALPDO Activity Catalyzed by HRP and Luciferase in Their Respective Buffers at pH 8.0 Light emission (mV) Compound (M)

[HRP] (M)

[Luciferase] (310 26 M)

Time of max light output

Background

Luminol (8.5 3 10 25) H 2O 2 (6 3 10 25) ALPDO (8.5 3 10 25) H 2O 2 (6 3 10 25) Luciferin (1.77 3 10 24) ATP (1.25 3 10 213 mol) ALPDO (1.77 3 10 24) ATP (1.25 3 10 213 mol) Luminol (8.5 3 10 25) PIP (2.5 3 10 24) H 2O 2 (6 3 10 25) ALPDO (8.5 3 10 25) H 2O 2 (6 3 10 25) PIP (2.5 3 10 24)

5.6 3 10 28

—

20 min

33

75

5.6 3 10 28

—

55 s

29

920

—

4.3

5s

—

290

—

4.3

5s

—

30

5.6 3 10 210

—

15 min

5.6 3 10 210

—

13 s

Note. The values in parentheses are before the addition of PIP and after background corrections.

10

6

With enzyme

3400 (1.8)

52 (10.2)

BIFUNCTIONAL ENZYME SUBSTRATE TABLE 4

Effect of Chemiluminescence Enchancer Luciferin (4.46 3 10 23 M) on Luminol/ALPDO (1.25 3 10 24 M) and H 2O 2 (3.3 3 10 24 M) in the Presence and the Absence of HRP (2.8 3 10 210 M) at pH 8.0 Substrate

Chemical blank (mV)

Light output (mV)

Luminol Luminol 1 LN Luminol 1 LN 1 HRP ALPDO ALPDO 1 LN ALPDO 1 LN 1 HRP

10 5 4 6 3 2

44 .10,000 .10,000 28 801 8,195

ions, the free radicals formed from these are more stable. When LN/PIP is added to luminol during the chemiluminescence reaction, the latter can yield a stable radical at an accelerated rate compared to the rate in the absence of enhancers, resulting in enhanced chemiluminescence (30). As ALPDO is devoid of amino substituent and in its place an electron-withdrawing diazo group is present, it is less susceptible to electron transfer mediators, such as LN and PIP. However, the diazo group can impart more light-emitting characteristics than the amino group, and ALPDO can be irreversibly oxidized more efficiently than luminol. Although amines are good electron donors, they are not fluorescent and, hence, ALPDO has more light yield than luminol. A similar observation was noticed by Yang and Yang (31) in the case of anthracene adducts, in which electron-withdrawing groups at the bridge head enhanced the pericyclic chemiluminescence. ACKNOWLEDGMENTS T. Sudhaharan is employed at Jonaki, Board of Radiation and Isotope Technology, DAE. This research work is a part of his Ph.D. thesis and he thanks Dr. S. Gangadharan, Chief Executive, BRIT, for granting permission to continue his research work and Shri B. Muralidharan, Officer-in-Charge, Jonaki, BRIT, for continual encouragement.

REFERENCES 1. Campbell, A. K. (1988) in Chemiluminescence Principles and Applications in Biology and Medicine, Ellis Harwood, Chichester, England. 2. Geiger, R., Hauber, R., and Misaka, W. (1989) Mol. Cell. Probes 3, 309 –328. 3. Conti, E., Franks, N. P., and Brick, P. (1996) Structure 4, 287– 298. 4. Lasagna, M., Vargas, V., Jameson, D. M., and Brunet, J. E. (1996) Biochemistry 35, 973–979.

167

5. Yakunin, A. F., and Hallenbeck, P. C. (1998) Anal. Biochem. 258, 146 –149. 6. Arakawa, H., Maeda, M., and Tsuji, A. (1991) Anal. Biochem. 192, 238 –242. 7. Whitehead, T. P., Thorpe, G. H. G., Carter, T. J. N., Groucutt, C., and Kricka, L. J. (1983) Nature 305, 158 –159. 8. Thorpe, G. H. G., and Kricka, L. J. (1986) Methods Enzymol. 133, 331–351. 9. Thorpe, G. H. G., Kricka, L. J., Moseley, S. R., and Whitehead, T. P. (1985) Clin. Chem. 31, 1335–1341. 10. Koo, J. Y., Schmidt, S. P., and Schuster, G. B. (1977) Proc. Natl. Acad. Sci. U.S.A. 75, 30 –33. 11. Beck, S., and Koster, H. (1990) Anal. Chem. 62, 2258 –2270. 12. DeLuca, M., and McElroy, W. D. (1978) Methods Enzymol. 57, 3–124. 13. Goodrich, G. A., and Burrell, H. R. (1982) Anal. Biochem. 127, 395– 404. 14. Thorpe, G. H. G., and Kricka, L. J. (1989) J. Biolumin. Chemilumin. 3, 97–100. 15. Miska, W., and Geiger, R. (1987) J. Clin. Chem. Clin. Biochem. 25, 23–30. 16. Ugarova, N. N., Vozny, Y. A., Kutuzova, G. D., and Dementieva, E. I. (1991) in Bioluminescence Chemiluminascence, Current Status (Stanley, P. E., and Kricka, L. J., Eds.), p. 511, Wiley, Chichester. 17. Mayer, A., and Neuenhoffer, S. (1994) Angew. Chem. Int. Ed. Engl. 33, 1044 –1072. 18. White, E. H., and Rosewell, D. F. (1970) Acc. Chem. Res. 3, 54 – 62. 19. Thorpe, G. H. G., Kricka, L. J., Gillespie, E., Moseley, S., Amess, R., Baggett, N., and Whitehead, T. P. (1985) Anal. Biochem. 145, 96 –100. 20. Sasamoto, K., Deng, G., Ushijima, T., Ohkura, Y., and Ueno, K. (1995) Analyst 120, 1709 –1714. 21. Klel, J. L., and Obreen, G. J. (1991) U.S. Patent 5003050. 22. Liu, Z., Zhao, C., Tang, M., and Cai, S. (1996) J. Phys. Chem. 100, 17337–17344. 23. Griffiths, J. (1972) Chem. Soc. Rev. 1, 481– 493. 24. Roberts, D. R., and White, E. H. (1978) J. Am. Chem. Soc. 92, 4861– 4867. 25. Thompson, J. E., and Jordan, D. B. (1998) Anal. Biochem. 256, 7–13. 26. Briggs, G. E., and Haldane, J. B. S. (1925) Biochem. J. 19, 338 –339. 27. Kim, B. B., Pisarev, V. V., and Egorov, A. M. (1991) Anal. Biochem. 199, 1– 6. 28. Hastings, J. G. M., Wheat, P. F., and Oxiey, K. M. (1989) Soc. Appl. Bacteriol. Tech. Ser. 26, 229 –233. 29. Thorpe, G. H. G., Stott, R. A. W., Sankolli, G. M., Catty, D., Raykundalia, C., and Kricka, L. J. (1987) in Bioluminescence and Chemiluminescence: New Perspectives (Scholmerich, J., Andreesen, R., Kapp, A., Ernst, M., and Woods, W. G., Eds.), Wiley, Chichester. 30. Gilbert, A., and Baggott, J. (1991) Essentials of Molecular Photochemistry, Blackwell, London. 31. Yang, N. C., and Yang, X. Q. (1989) J. Am. Chem. Soc. 109, 3804 –3805.