Conjugated Polyelectrolyte Based Real-Time Fluorescence Assay for

Nov 20, 2008 - The fluorescence turn-on and turn-off assays monitor the forward and reverse .... Hyper-Efficient Quenching of a Conjugated Polyelectro...
1 downloads 0 Views 600KB Size
Download PDF
Anal. Chem. 2009, 81, 231–239

Conjugated Polyelectrolyte Based Real-Time Fluorescence Assay for Adenylate Kinase Yan Liu and Kirk S. Schanze* Department of Chemistry, University of Florida, P.O. Box 117200, Gainesville, Florida 32611-7200 Addition of adenosine 5′-triphosphate (ATP) to a solution of the anionic conjugated polyelectrolyte PPECO2 and copper(II) ion (Cu2+) recovers the Cu2+-quenched fluorescence of PPECO2 to a significantly greater extent compared with the addition of adenosine 5′-diphosphate (ADP) or adenosine 5′-monophosphate (AMP) at the same concentration levels. Taking advantage of the differential response of the PPECO2-Cu2+ system to ATP, ADP and AMP, we have developed fluorescence turn-off and turnon assays that monitor the catalytic activity of adenylate kinase (ADK) in the equilibrium transphosphorylation reaction (ATP + AMP S 2ADP). The fluorescence turnon and turn-off assays monitor the forward and reverse transphosphorylation reactions, respectively. The forward assay operates with ATP substrate present at the submillimolar concentration range and offers a straightforward and rapid detection of ADK catalytic activity with the enzyme present in the nanomolar range, in either endpoint or real-time formats. The real-time fluorescence intensity from PPECO2 can be converted to substrate (ATP) concentration in the forward reaction assay by using an ex-situ calibration curve, allowing ADK catalyzed reaction rates and kinetic parameters to be determined. ADK activation by Mg2+ and inhibition by Ag+ and product are analyzed using the optimized assay system. Non-specific interactions are observed between the assay complex and other proteins, but the signal response to the ADK assay is demonstrated to mainly arise from the specific enzyme catalyzed transphosphorylation reaction. Adenylate kinase (ADK, EC 2.7.4.3) is a ubiquitous adenosine 5′-triphosphate:adenosine 5′-monophosphate (ATP:AMP) phosphotransferase which distributes in variety of organisms and catalyzes the interconversion of the adenine nucleotides as shown in Scheme 1.1 ADK composes the adenylate system with ATP, adenosine 5′-diphosphate (ADP) and AMP, and plays a unique buffering role in maintaining equilibrium among the adenine nucleotides, thereby maintaining energy balance and facilitating the storage and use of the high energy of the adenine nucleotides.2 The enzyme also serves as a regulatory factor for many enzymatic reactions in which the adenine nucleotides participate as substrate, * To whom correspondence should be addressed. E-mail: kschanze@ chem.ufl.edu. Phone: 352-392-9133. Fax: 352-392-2395. (1) Noda, L. Enzyme 1973, 8, 279–305. (2) Bomsel, J. L.; Pradet, A. Biochim. Biophys. Acta 1968, 162, 230–242. 10.1021/ac801908f CCC: $40.75  2009 American Chemical Society Published on Web 11/20/2008

activator, or inhibitor.3,4 In addition, recent research reveals that ADK plays an important role in cell metabolism, including the synthesis of DNA and RNA.5 ADK can also phosphorylate nucleoside analogues used in the treatment of cancer and viral infection.6,7 Because of the importance of ADK in biological systems, the enzyme has been extensively studied in enzymology, human genetics, protein chemistry, crystallography, and molecular spectroscopy. Several ADK assays have been developed based on chromatographic separation,8 pH-stat titration,9 bioluminescence,10-12 and coupled-enzyme methods.13,14 The chromatographic separationbased assay offers direct measurement of changes in the amount of the adenine nucleotides; however, it is laborious and timeconsuming.1 The pH-stat assay suffers from low sensitivity and the inherent disadvantage that large quantities of enzyme and substrate are required.15 Although the bioluminescence assay is sensitive and rapid, it is not widely used because it requires coupling with firefly luciferase. The most popular and commercially available assays are colorimetric, and they couple the ADK catalyzed reaction with other enzymes.16 For example, the assay to detect the forward reaction activity of ADK (Scheme 1) is coupled with pyruvate kinase and lactate dehydrogenase and is carried out by monitoring the absorbance change of NADH at 340 nm concomitant with its oxidation. In a similar way, the assay to detect the reverse activity of ADK is coupled with hexokinase and glucose-6-phosphate dehydrogenase and carried out by monitoring the absorbance change of NADPH at the same wavelength accompanying the reduction of NADP.14 The colorimetric enzyme-coupled assays offer simple operation and poten(3) Storey, K. B. J. Biol. Chem. 1976, 251, 7810–7815. (4) Atkinson, D. E. Annu. Rev. Microbiol. 1969, 23, 47–68. (5) Dzeja, P. P.; Zeleznikar, R. J.; Goldberg, N. D. Mol. Cell. Biochem. 1998, 184, 169–182. (6) Dzeja, P. P.; Terzic, A. J. Exp. Biol. 2003, 206, 2039–2047. (7) Dzeja, P. P.; Bortolon, R.; Perez-Terzic, C.; Holmuhamedov, E. L.; Terzic, A. Proc. Natl. Acad. Sci. U.S.A. 2002, 99, 10156–10161. (8) Noda, L. J. Biol. Chem. 1958, 232, 237–250. (9) Mahowald, T. A.; Kuby, S. A.; Noltmann, E. A. J. Biol. Chem. 1962, 237, 1535–1548. (10) Blasco, R.; Murphy, M. J.; Sanders, M. F.; Squirrell, D. J. J. Appl. Microbiol. 1998, 84, 661–666. (11) Wu, Y.; Brovko, L.; Griffiths, M. W. Lett. Appl. Microbiol. 2001, 33, 311– 315. (12) Shutenko, T. V.; Gavrilova, E. M.; Egorov, A. M. Biotekhnologiya 1988, 4, 542–547. (13) Russell, P. J.; Horenste, J.; Goins, L.; Jones, D.; Laver, M. J. Biol. Chem. 1974, 249, 1874–1879. (14) Li, X.; Pan, X. M. FEBS Lett. 2000, 480, 235–238. (15) Thuma, E.; Schirmer, R. H.; Schirmer, I. Biochim. Biophys. Acta 1972, 268, 81–91. (16) Noda, L. Enzyme 1962, 6, 139–149.

Analytical Chemistry, Vol. 81, No. 1, January 1, 2009

231

Scheme 1. Interconversion of the Adenine Nucleotides Catalyzed by ADK

Scheme 2. Structure of PPECO2

tially real-time detection; however, they are less sensitive compared with fluorometric assays, and their accuracy depends on other system factors such as the activity of coupled enzymes. We have recently reported the amplified quenching of PPECO2 (Scheme 2), an anionic conjugated polyelectrolyte (CPE) featuring carboxylate side groups, by cupric ion (Cu2+)17 via a charge and/ or energy transfer mechanism.18,19 The amplified quenching phenomenon (superquenching) of CPEs is attributed to the delocalization and extremely rapid diffusion of the singlet exciton along the CPE backbone to the quencher “trap site”,20-25 and it offers the enhanced sensitivity in the application of CPEs in biosensors.20,24,26-32 The Cu2+-quenched fluorescence of PPECO2 is efficiently recovered by addition of pyrophosphate (PPi),17 and this effect provides the basis for the recently reported alkaline phosphatase (ALP) turn-off assay.33 By analogy to the effect of addition of PPi, titration of the adenosine phosphates (ATP, ADP, and AMP) individually into a PPECO2/Cu2+ solution gives rise to fluorescence recovery to a different extent, and this effect affords a novel platform for the fluorescent ADK assay described in this paper. In particular, the Cu2+-quenched fluorescence is recovered to the greatest extent by ATP, in contrast to a much (17) Zhao, X.; Liu, Y.; Schanze, K. S. Chem. Commun. 2007, 2914–2916. (18) Fan, L. J.; Jones, W. E. J. Am. Chem. Soc. 2006, 128, 6784–6785. (19) Fabbrizzi, L.; Licchelli, M.; Pallavicini, P. Acc. Chem. Res. 1999, 32, 846– 853. (20) DiCesare, N.; Pinto, M. R.; Schanze, K. S.; Lakowicz, J. R. Langmuir 2002, 18, 7785–7787. (21) Tan, C. Y.; Pinto, M. R.; Schanze, K. S. Chem. Commun. 2002, 446–447. (22) Wang, D. L.; Wang, J.; Moses, D.; Bazan, G. C.; Heeger, A. J. Langmuir 2001, 17, 1262–1266. (23) Gaylord, B. S.; Wang, S. J.; Heeger, A. J.; Bazan, G. C. J. Am. Chem. Soc. 2001, 123, 6417–6418. (24) Chen, L. H.; McBranch, D. W.; Wang, H. L.; Helgeson, R.; Wudl, F.; Whitten, D. G. Proc. Natl. Acad. Sci. U.S.A. 1999, 96, 12287–12292. (25) Zhou, Q.; Swager, T. M. J. Am. Chem. Soc. 1995, 117, 7017–7018. (26) Liu, Y.; Ogawa, K.; Schanze, K. S. Anal. Chem. 2008, 80, 150–158. (27) Liu, M.; Kaur, P.; Waldeck, D. H.; Xue, C. H.; Liu, H. Y. Langmuir 2005, 21, 1687–1690. (28) Kumaraswamy, S.; Bergstedt, T.; Shi, X. B.; Rininsland, F.; Kushon, S.; Xia, W. S.; Ley, K.; Achyuthan, K.; McBranch, D.; Whitten, D. Proc. Natl. Acad. Sci. U.S.A. 2004, 101, 7511–7515. (29) Pinto, M. R.; Schanze, K. S. Proc. Natl. Acad. Sci. U.S.A. 2004, 101, 7505– 7510. (30) Rininsland, F.; Xia, W. S.; Wittenburg, S.; Shi, X. B.; Stankewicz, C.; Achyuthan, K.; McBranch, D.; Whitten, D. Proc. Natl. Acad. Sci. U.S.A. 2004, 101, 15295–15300. (31) Gaylord, B. S.; Heeger, A. J.; Bazan, G. C. J. Am. Chem. Soc. 2003, 125, 896–900. (32) Pinto, M. R.; Schanze, K. S. Synthesis 2002, 1293–1309. (33) Liu, Y.; Schanze, K. S., Anal. Chem. [Online early acess]. DOI: 10.1021/ ac801508y. Published Online: 2008.

232

Analytical Chemistry, Vol. 81, No. 1, January 1, 2009

smaller recovery induced by ADP, and virtually no recovery by AMP. The different recovery efficiency for the three adenosine phosphates arises from their differing propensity to complex with Cu2+: ATP with its higher negative charge density and greater number of oxygen donor units features the highest association constant with Cu2+. As a result, the quenched PPECO2-Cu2+ complex is effectively disrupted by ATP, leading to the recovery of the fluorescence intensity. Moreover, the fluorescence recovery can be used to develop a calibration curve to afford a quantitative measure of ATP concentration. In the current article we report the development and optimization of a real-time fluorescence assay for ADK which is based on monitoring the concentration of ATP which serves as either the substrate in the forward reaction or the product of the reverse reaction. The assay for the forward reaction is a fluorescence “turnoff” assay. By introducing ADK into a solution containing PPECO2, Cu2+, Mg2+, ATP, and AMP, and applying an ex-situ calibration, the decrease of PPECO2 fluorescence intensity is quantitatively related to the (decreasing) ATP concentration as a function of incubation time. (In this assay, Mg2+ serves as an activator and AMP is a second substrate.) The assay to monitor the reverse ADK-catalyzed reaction is a fluorescence “turn-on” assay, where an increase of fluorescence intensity reflects an increase in ATP concentration as the reaction proceeds. Both assays are carried out at physiological pH, they are sensitive, and provide either endpoint or real-time measurement capability, with the latter format allowing enzyme kinetic parameters to be determined. MATERIALS AND METHODS Materials. All solutions were prepared with water that was distilled and then purified by using a Millipore purification system. HEPES buffer solution (10 mM, pH 7.5) was prepared with 4-(2hydroxyethyl)-1-piperazineethanesulfonic acid and hydrochloric acid. The aqueous stock solution of PPECO2 was diluted with HEPES buffer solution to a final concentration of 3 µM. All chemicals were used as received, unless otherwise noted. Copper(II) chloride (CuCl2), adenosine 5′-triphosphate disodium salt (ATP), adenosine 5′-diphosphate sodium salt (ADP), and adenosine 5′-monophosphate sodium salt (AMP) were obtained from Sigma-Aldrich. Magnesium chloride was purchased from Fisher Chemical. Silver nitrate was obtained from Mallinckrodt. Each reagent was dissolved in HEPES buffer, and the concentration of the stock solutions was adjusted to 100 mM. Adenylate kinase (ADK, myokinase from rabbit muscle) and the control proteins were purchased from Sigma-Aldrich. The five control proteins include peroxidase from horseradish type I (HRP), peptidase from porcine intestinal mucosa (PTD), phospholipase D from Arachis hypogaea (peanut) (PLD), glucose oxidase from Aspergillus niger (GOx), bovine serum albumin (BSA). All of the enzymes and proteins were dissolved in HEPES buffer solution and adjusted to 20 µM except for PTD with 3.2 mg · mL-1 as stock concentration. The enzyme and protein stock solutions were prepared immediately before their use in the fluorescence assays. Fluorescence Assay Procedure. The enzyme assays were carried out in 10 mM HEPES buffer (pH 7.5) at 37 °C. For the real-time assays, the fluorescence intensity was recorded with excitation and emission wavelengths of 390 and 525 nm, respectively. All measurements were conducted with stirred solutions.

Figure 1. Mechanism of ADK turn-off and turn-on assay.

In the turn-off approach, ATP was used as a substrate. A typical procedure was carried out as follows: First, a 2.0 mL aliquot of the solution containing PPECO2, Cu2+, Mg2+, and AMP was placed in the cuvette, and the mixture was thermally equilibrated to 37 °C. The quenched polymer fluorescence intensity (Iq) at 525 nm was recorded. Subsequently, the ATP solution was introduced into another 2.0 mL aliquot of PPECO2/Cu2+/Mg2+/AMP solution, this mixture was incubated for 15 min and quickly pipetted into a cuvette containing a microliter range aliquot of the ADK solution, and the fluorescence intensity (It) of the solution was measured at 1 s increments. In the turn-on approach, ADP was used as the substrate, and the procedure was the same as the turn-off approach, except that it was initiated with measurement of Iq of PPECO2/Cu2+/Mg2+ solution and followed by the addition of an ADP solution aliquot before introducing ADK and recording It. For monitoring the assays in an end-point format, the fluorescence intensity (It) of the assay solution versus wavelength spectra were recorded with excitation wavelength at 390 nm. Calculation of Initial Rate of Reaction (ν0). In the turn-off approach, the fluorescence intensity of the assay solution, It, was converted to substrate concentration [ATP]t at 1 s intervals during the assay by using eq 1. Equation 1 is derived from the calibration which relates the fluorescence intensity to ATP: (log(Ir/Iq) ) c × [ATP], where Iq is the quenched fluorescence intensity of the PPECO2 solution in the presence of Cu2+, Mg2+, and AMP, Ir is the recovered fluorescence intensity after introduction of ATP, and c is the proportionality constant. This relationship is discussed thoroughly in the text surrounding Figure 3.

() ()

It Iq [ATP]t ) [ATP]0 × I0 log Iq log

(1)

In eq 1, [ATP]0 is the initial substrate concentration, [ATP]t is the substrate concentration at time t, I0 is the initial fluorescence intensity at t ) 0, that is, the fluorescence intensity after addition of substrate ATP and before addition of enzyme, and It is the fluorescence intensity at time t after initiation of transphosphorylation by enzyme. A plot of [ATP]t versus time was then derived

and the v0 was calculated from the slope of the plot by using the data from the region where [ATP]t is a linear function of the time. RESULTS AND DISCUSSION Overview of ADK Turn-off Assay. An anionic CPE featuring carboxylate side groups, PPECO2 (Scheme 2), is used in the ADK assay. The synthesis and characterization of this polymer is described in an earlier communicaiton.34 In 10 mM HEPES buffer solution (pH 7.5), PPECO2 exhibits a broad fluorescence band with a quantum yield of 0.16 and an emission maximum at 525 nm, suggesting the polymer is aggregated in buffer solution. A divalent metal ion is required for ADK activity.1 Magnesium ion, Mg2+, was chosen in our experiments, because it is reported to be the most effective activator compared with other metal ions, for example, calcium (Ca2+) and manganese (Mn2+).16 The cartoon in Figure 1 illustrates the mechanism of the ADK turn-off and turn-on assays, which depends on the different fluorescence recovery efficiency of Cu2+-quenched polymer by ATP, ADP, and AMP. The fluorescence of PPECO2 is quenched efficiently by Cu2+. The quenched fluorescence is recovered to the greatest extent upon addition of ATP; however, it is recovered to a distinctly smaller extent by introducing ADP and to a negligible extent by introducing AMP. In a turn-off approach, the ATP and AMP substrates are added into the PPECO2/Cu2+ solution containing the activator, Mg2+. AMP has no effect on the polymer-metal complex; however, the PPECO2-Cu2+ complex is disrupted by ATP because of association of ATP with Cu2+, thus leading to the recovery of the fluorescence intensity. Introduction of ADK to the PPECO2/Cu2+/Mg2+/ATP/AMP mixture initiates the forward transphosphorylation reaction which produces ADP. With less negative charge density, the product, ADP, binds to Cu2+ more weakly than ATP. As the reaction proceeds, the amount of ATP available to associate with Cu2+ decreases, and the fluorescence of PPECO2 is quenched by the Cu2+ that is released. Thus, ADK catalytic activity is signaled by the fluorescence switching from the “on” state to the “off” state. The changed fluorescence intensity is related to the substrate concentration, [ATP], as a function of time in the enzymatic reaction by using eq 1, thereby allowing quantitative investigation of ADK activity. (34) Jiang, H.; Zhao, X. Y.; Schanze, K. S. Langmuir 2006, 22, 5541–5543.

Analytical Chemistry, Vol. 81, No. 1, January 1, 2009

233

Figure 2. Comparison of fluorescence intensity increase at 525 nm upon titration of ATP, ADP, and AMP, respectively. Solution conditions: 5 µM PPECO2 and 10 µM Cu2+ in 10 mM HEPES buffer (pH 7.5) at 37 °C, λex ) 390 nm.

In a turn-on approach as shown in Figure 1, the procedure is reversed compared with the turn-off approach in terms of steps and on-off switch. The ADP substrate is added into the PPECO2/ Cu2+ solution containing Mg2+. In contrast to ATP, ADP has a lesser ability to effectively restore the quenched fluorescence of PPECO2. Consequently, the mixture maintains its fluorescent “off” state. Addition of ADK to the mixture of PPECO2/Cu2+/Mg2+/ ADP initiates the reverse transphosphorylation reaction producing ATP, which induces the recovery of the quenched fluorescence of PPECO2. Therefore, the reverse activity of ADK is monitored by the increase of the fluorescence intensity of PPECO2. Fluorescence Recovery by ATP, ADP, and AMP. In the previous communication,17 we reported that PPECO2 exhibits an amplified quenching effect by Cu2+ in HEPES buffer with a Stern-Volmer constant (KSV) of 106 M-1. The quenching mechanism is suggested to be singlet exciton quenching via charge and/or energy-transfer facilitated by coordination of Cu2+ with the carboxylate groups of the polymer.18,19 The Cu2+-quenched fluorescence of PPECO2 is efficiently recovered by PPi because of the strong binding between phosphate groups on PPi and the metal ion, thus disrupting the polymer-metal complex.17,33 The adenine nucleotides (ATP, ADP, and AMP), which feature phosphate groups by analogy to PPi, are anticipated to give rise to a similar fluorescence recovery effect. To investigate the fluorescence recovery effects, a series of titrations were carried out using ATP, ADP, or AMP over a concentration range from 0 to 40 µM into a HEPES buffer (10 mM, pH 7.5) solution containing 5 µM PPECO2 and 10 µM Cu2+. Figure 2 illustrates the fluorescence recovery efficiency, Ir/Iq, as a function of the concentration of ATP, ADP, or AMP. (Iq and Ir are, respectively, the fluorescence intensity of PPECO2/Cu2+ at 525 nm before and after addition of the adenosine phosphate). At the same concentration level, the recovery efficiency increases in the order of ATP > ADP > AMP, which correlate with the decrease of negative charge density of the anions. With the largest negative charge density, ATP features the strongest electrostatic association with Cu2+, thus showing the greatest recovery efficiency among three adenosine phosphates. Similar to PPi recovery,33 the plot of Ir/Iq of ATP features a sigmoidal shape: increasing slowly at very low [ATP] and rising sharply with increase of [ATP] until reaching a plateau. In the same concentration range, ADP gives rise to a lower recovery efficiency, while AMP exhibits negligible recovery. When the concentrations of ATP, ADP, and AMP are at 40 µM, the fluorescence is recovered to 77%, 25%, and 0% of the initial intensity of the unquenched 234

Analytical Chemistry, Vol. 81, No. 1, January 1, 2009

polymer, or as a 17, 5, and 0-fold intensity increase relative to the quenched fluorescence, respectively. It is important to point out that ADP is able to substantially recover the quenched fluorescence when its concentration is very high (e.g., 600 µM ADP induces a 15-fold fluorescence intensity increase), but AMP does not give rise to noticeable recovery even at concentrations as high as 3 mM. The different response of the Cu2+-quenched fluorescence of PPECO2 to ATP, ADP, and AMP provides the basis for development of an ADK assay. In principle, both turn-off and turn-on approaches as shown in Figure 1 are applicable; however, the turnoff approach is much easier to implement in practice because of the following two factors: (1) easier control of reaction direction; (2) more sensitive signal detection. Concerning the first factor, the apparent equilibrium constant, Kapp () [AMP] [ATP]/[ADP]2) for the transphosphorylation interconversion reaction catalyzed by ADK is reported as ∼0.82,1 so it is desirable to control the reaction during an assay to ensure that it occurs predominantly in one direction. Generally this is realized by manipulating the amount of AMP in the solution, either by adding excess AMP in the forward direction as shown in Scheme 1 (turn-off approach) or by eliminating generated AMP in the reverse direction (turnon approach). Clearly it is easier to add excess AMP than to eliminate AMP as it is generated. The second factor arises from the sigmoidal shape of the recovery curve for ATP as shown in Figure 2, and the relatively stronger fluorescence recovery by ADP that occurs at higher concentrations. At early times in the turnon assay as shown in Figure 1, any florescence recovery induced by the small amount of ATP produced is obscured by the stronger recovery effect induced by the large amount of ADP substrate that is in the reaction solution. Consequently, it is difficult to monitor [ATP] at the beginning of the turn-on assay by monitoring the change in fluorescence intensity. By contrast, at early times in the turn-off assay, the solution has high ATP concentration and a low ADP concentration. Thus, the change of fluorescence intensity accurately represents the decrease of ATP concentration, without interference from ADP. Because of these two factors, we emphasize the turn-off assay for the forward reaction, providing details concerning the assay procedure, effect of Mg2+, kinetics, inhibition, and specificity. The turn-on assay for the reverse reaction is described briefly at the end of the paper. Calibration Curves of ADK Turn-off Assay. Magnesium ion, Mg2+, is a required activator for ADK activity.1 However, addition of Mg2+ into the polymer solution also induces quenching of PPECO2 but with a much smaller KSV (4.6 × 102 M-1) compared with that of Cu2+. In addition, the presence of Mg2+ changes the ionic environment of the solution, influencing the interactions between PPECO2, Cu2+, and ATP. Therefore, the effect of Mg2+ must be considered when the amount of ATP is quantified by fluorescence intensity. A series of titrations of ATP into a HEPES buffer (10 mM, pH 7.5) solution of 3 µM PPECO2 and 10 µM Cu2+ in the presence of various concentrations of Mg2+ were carried out. Each plot of the fluorescence recovery efficiency (Ir/Iq) as a function of [ATP] displays a sigmoidal curve similar to that shown Figure 2. To obtain an approximately linear correlation that can be used as a calibration curve to relate the change in fluorescence intensity to [ATP], the fluorescence intensity data was plotted as shown in

Figure 3, which shows correlations of log(Ir/Iq) versus [ATP] at different concentrations of Mg2+. Note that as the Mg2+ concentration increases, more ATP is needed to recover the fluorescence. This is consistent with the weak fluorescence quenching effect of Mg2+, as well as the fact that complexation between ATP and Cu2+ is less efficient in the presence of Mg2+ (presumably because Mg2+ competes with Cu2+ for binding to ATP). Importantly, each of the plots which correspond to different [Mg2+] exhibits a linear increase in log(Ir/Iq) with [ATP] over the concentration range from 0 to about 1.2 equiv of [Mg2+]. For example, in the presence of 300 µM Mg2+, the linear range of [ATP] is from 0 to 360 (1.2 × 300) µM. This suggests that each curve can be used as a calibration plot which linearly relates fluorescence intensity to ATP concentration. These calibration curves are used to derive eq 1 which is applied to determine [ATP]t during the ADK turn-off assay. Note that the applicability of eq 1 is independent of [Mg2+], but the range of [ATP] over which the calibration curve is linear varies with [Mg2+]. In addition, this linear range of [ATP] is also affected by the Cu2+ concentration.33 Therefore, the range of applicable substrate concentration can be adjusted by the amount of Mg2+ or Cu2+ in the system. To control the reaction to be favorable in the forward direction, excess AMP is added in the ADK turn-off assay. Control experiments reveal that the added AMP elicits little effect on the calibration curves, even when it is present at concentrations in the millimolar range. In the following discussion, we will refer to a solution containing 3 µM PPECO2, 10 µM Cu2+, 2 mM AMP, and 10 mM HEPES buffer (pH 7.5) as the “standard assay solution” used in the ADK turn-off assays. ADK Turn-off Assay. To eliminate the possibility that the fluorescence of the polymer is influenced by unexpected factors, several control experiments were conducted. The effects that addition of three adenosine phosphates, ATP, ADP, AMP, and the enzyme, ADK, have on the fluorescence of the solution of Cu2+-free PPECO2 were explored, and little effects were observed when these species were added individually to the polymer solution. Introduction of ADK into a solution containing PPECO2, Cu2+, Mg2+, ATP, and AMP initiates transphosphorylation from ATP to AMP, producing ADP as the product (forward reaction). The enzyme reaction is monitored by the decrease of the polymer’s fluorescence intensity. Figure 4a shows typical fluorescence spectroscopic changes observed during an end-point ADK turnoff assay carried out in the standard solution at 37 °C. The photograph in the inset clearly illustrates that the fluorescence changes accompanying the assay are visible by eye (under UV illumination to excite the polymer fluorescence). In particular, the quenched fluorescence from solution 1 which contains 200 µM Mg2+ in the standard solution is significantly enhanced in intensity by adding 200 µM ATP, forming solution 2 (change indicated by arrow 1 f 2). Following introduction of 500 nM ADK into the solution 2 and initiating the enzymatic reaction, consumption of ATP is signaled by a decrease of the fluorescence intensity with increasing incubation time (change indicated by arrow 2 f 3). In solution 3, the fluorescence of the polymer is quenched by the free Cu2+, which is produced as a result of the ADK-catalyzed transphosphorylation reaction. The fluorescence intensity de-

Figure 3. Calibration curves, logarithm of fluorescence recovery efficiencies (log(Ir/Iq)) as a function of [ATP], at different Mg2+ concentrations. Solution conditions: 3 µM PPECO2 and 10 µM Cu2+ in 10 mM HEPES buffer (pH 7.5) at 37 °C, λex ) 390 nm, λem ) 525 nm.

creases initially very quickly (∼50% in the first 3 min), then it decreases more gradually as the reaction is approaching equilibrium, decreasing by ∼83% relative to the initial “recovered” fluorescence intensity after 60 min of incubation time. The decrease of the fluorescence intensity is faster at the beginning of the assay because at early times the ADK-catalyzed reaction shown in Scheme 1 is dominated by the forward reaction in the presence of ATP and (excess) AMP substrates. However, as the reaction proceeds, the ADP that is generated by the forward transphosphorylation initiates the reverse reaction which begins to compete with the forward reaction. Therefore, the rate of decrease of the fluorescence intensity slows as the transphosphorylation reaction proceeds. In addition, the non-specific effect due to the addition of ADK into the assay solution also contributes to the initial decrease of fluorescence intensity during the initial period of assay. We will discuss the non-specific effects in detail below. To test the feasibility of using the PPECO2-based fluorescence assay as a real-time turn-off assay for ADK, we monitored the fluorescence intensity continuously as a function of time at varying ADK concentrations. The real time assays were carried out with 200 µM Mg2+ and 200 µM ATP in the standard assay solution at 37 °C. Figure 4b illustrates the continuous decrease of fluorescence intensity at 525 nm for solutions with [ADK] ranging from 0-500 nM, where the fluorescence intensity was measured at 1 s intervals. Note that the reaction rate increases with [ADK]. The fluorescence intensity decreases approximately linearly with time during the initial time period (∼60 s). However, the reaction rate decreases as the reaction continues, which is in good agreement with what was observed in the end-point ADK turn-off assay (Figure 4a). This effect becomes more pronounced at higher enzyme concentration. To determine the initial rate of the ADK-catalyzed reaction (v0), the concentration of ATP during the assay, [ATP]t was calculated from the time-varying fluorescence intensity in Figure 4b by using the calibration expressed by eq 1, affording plots of [ATP]t as a function of time (see Supporting Information, Figure S-1). These plots show the [ATP]t during the entire time course of the ADK turn-off assay at different concentrations of ADK. It is evident that the plots are linear, and the slopes of the plots using the data of the first 60 s of assay were calculated to afford values of v0 at Analytical Chemistry, Vol. 81, No. 1, January 1, 2009

235

Figure 4. (a) Fluorescence spectroscopic changes observed in an end-point ADK turn-off assay. (red) fluorescence of solution 1: 200 µM Mg2+in the standard assay solution at 37 °C, λex ) 390 nm; (green) fluorescence of solution 2: addition of 200 µM ATP into solution 1; fluorescence intensity as a function of time of solution 3: addition of 500 nM ADK into solution 2 and incubation for 3 (yellow), 5 (blue), 8 (magenta), 15 (light blue), 60 (red) min. Inset: Photographs of solution 1, 2, and 3 (60 min) illuminated with near UV light illustrate the polymer fluorescence under the different condition of the assay. (b) Decrease of fluorescence intensity at 525 nm recorded every 1 s during real-time ADK turn-off assays with varying concentration of ADK. Solution conditions: 200 µM Mg2+ and 200 µM ATP in the standard assay solution at 37 °C, λex ) 390 nm.

Figure 5. Dependence of initial rate of reaction (v0) on ADK concentration. Solution conditions: 200 µM Mg2+ and 200 µM ATP in the standard assay solution at 37 °C, λex ) 390 nm, λem ) 525 nm.

seven different ADK concentrations. Figure 5 illustrates a plot of v0 as a function of ADK concentration in the range of 0-500 nM. The initial rate of reaction increases approximately linearly with [ADK] when the enzyme concentration is below 300 nM, which indicates that the reaction is kinetically controlled by ADK. The slope of the linear plot affords the enzyme turnover number, kcat, to be 0.75 s-1, which is similar to the value obtained in separate experiments as described below (kcat ) 0.16 s-1). The correlation “flattens” at higher concentration of ADK because of the inhibition of the forward reaction by ADP, which is generated more quickly at high [ADK]. This experiment also demonstrates that it is possible to use the PPECO2-based assay to detect ADK activity in less than 1 min, even at low [ADK]. An analytical detection limit of ∼42 nM ADK is obtained from the linear range of the plot in Figure 5. This sensitivity is far below the concentration range of ADK which is generally assayed by commonly used coupled-enzyme assays in previous applications.5,14,35-37 Note that the fluorescence intensity values used here were not corrected for photobleaching as was done in the previous work.26,33 Because the photobleaching of PPECO2 in HEPES buffer is negligible, less than 2% of the initial polymer fluorescence intensity is lost because of bleaching upon 5 min of continuous excitation. (35) Hamada, M.; Kuby, S. A. Arch. Biochem. Biophys. 1978, 190, 772–792. (36) Zhai, R. T.; Meng, G.; Zhao, Y. M.; Liu, B.; Zhang, G. F.; Zheng, X. F. FEBS Lett. 2006, 580, 3811–3817. (37) Han, Y.; Li, X.; Pan, X. M. FEBS Lett. 2002, 528, 161–165.

236

Analytical Chemistry, Vol. 81, No. 1, January 1, 2009

Figure 6. Effect of concentration of Mg2+ on initial rate of reaction (v0) at different concentrations of ATP. Solution conditions: 200 nM ADK in the standard assay solution at 37 °C, λex ) 390 nm, λem ) 525 nm.

Effect of Mg2+ on ADK Activity in Turn-off Assay. The ADK-catalyzed reaction requires a nucleotide complexed with Mg2+ as one substrate and a free nucleotide as the second substrate.16 In particular, in the forward reaction, Mg2+ combines with ATP to form a complex (MgATP) which acts as a substrate and occupies one of two active substrate sites in the enzyme. The other site is bound with the metal-free substrate, AMP. In the reverse reaction, the two substrate sites are occupied by complex MgADP and metal-free ADP.1 Therefore, MgATP/AMP and MgADP/ADP are generally considered to represent the substrates in the ADK catalyzed interconversion reaction. To optimize the concentration of the activator, the effect of Mg2+ on the ADK activity was investigated. In particular, the dependence of the initial reaction rate, v0, on [Mg2+] was examined with the standard assay solution containing 200 nM ADK, and ATP concentration ranging from 50 to 500 µM. At a given [Mg2+], the linear [ATP] range on the calibration curve is from 0 to 1.2 equivalents of [Mg2+], therefore, the range of [Mg2+] is this experiment was chosen to begin at 0.83 equiv of the ATP concentration () [ATP]/1.2). For example, if 300 µM ATP was used in the assay, the concentration of Mg2+ started at 250 µM () 0.83 × 300). Figure 6 illustrates the effect of Mg2+ on v0 at different concentrations of ATP. It is evident that at each [ATP], v0 reaches a maximum rate when [Mg2+] equals [ATP] and decreases quickly when [Mg2+] is in excess over [ATP] even to a small extent. The observed dependence of v0 on [Mg2+] is

Figure 7. Specificity of ADK turn-off assay. (a) Changes in fluorescence emission intensity at 525 nm after 60 min of incubation of 300 nM ADK, HRP, PLD, GOx, BSA or 48 µg · mL-1 PTD. 400 µM Mg2+ and 400 µM ATP in the standard assay solution at 37 °C, λex ) 390 nm, λem ) 525 nm. Photographs of solutions illuminated with near UV light illustrate the polymer fluorescence under the ADK turn-off assay and control assays. (b) Changes in fluorescence emission intensity at 525 nm as a function of incubation time after addition of 300 nM ADK and BSA.

consistent with the previous kinetic studies of ADK activity,3,16 which showed the maximum enzyme activity when [Mg2+]/[ATP] ) 1, and strong inhibition in catalyzed reaction is observed when [Mg2+] is in excess. The result further confirms that Mg2+ combines with ATP to form the stoichiometric complex, MgATP, which acts as a substrate occupying the substrate site. Determination of ADK Catalyzed ATP Transphosphorylation Kinetic Parameters. The PPECO2-based fluorescence turn-off assay was used to determine the kinetic parameters (Km and Vmax) for the ADK catalyzed transphorphorylation. The kinetic experiments were carried out by fixing the concentration of one substrate, AMP in excess, and varying the concentration of the other, ATP. To maintain the maximum activity of ADK, the stoichiometry of Mg2+ and ATP was set at 1:1, and the Mg2+ATP complex concentration, [MgATP] was varied from 0 to 500 µM. The experiment was conducted with 300 nM ADK in the standard assay solution at 37 °C. The values of initial rate of reaction, v0, were obtained and plotted as a function of initial [MgATP] (see Supporting Information, Figure S-2). Using a nonlinear regression routine, the plot of v0 versus [ATP] was fitted with the Michaelis-Menten equation (eq 2, when v0 is initial rate of reaction and [S]0 is the initial MgATP concentration). The apparent kinetic parameters for MgATP, Kmapp. and Vmaxapp. were obtained from the fitted curve with the values of 104 ± 15 µM and 0.48 ± 0.02 µM · s-1 (4.53 ± 0.22 µmol · min-1 · mg-1), respectively (see Supporting Information, Figure S-2). In addition, the turnover number, kcat, was determined to be 0.16 s-1 () Vmaxapp./ [ADK]). The value of the apparent Km is in good agreement with the value obtained using a coupled enzyme assay (Km ) 160 ± 30 µM);35 however, the apparent Vmax is about 14-fold smaller compared with the value reported from the same coupled enzyme assay (Vmax ) 63 ± 6 µmol · min-1 · mg-1).35 It is likely that the lower Vmax arises because of the effect of Cu2+ on the sulfhydryl groups of the enzyme. As a potent sulfhydryl reagent,38 Cu2+ is capable of oxidizing sulfhydryl groups on ADK which are essential for maximum enzyme activity as they are involved in maintaining the necessary conformation of the enzyme.16 However, the concentration of Cu2+ (10 µM) used in this assay is low enough to maintain a considerable fraction of the ADK activity. (38) Zhang, G. H.; Melvin, J. E. Proc. Soc. Exp. Biol. Med. 1996, 211, 190–198.

v0 )

Vmax × [S]0 Km + [S]0

(2)

Inhibition of the ADK Catalysis in Turn-off Assay. To further demonstrate that the observed fluorescence intensity decrease that is induced by addition of ADK arises because of ADK catalyzed transphosphorylation, the effect of the known inhibitor, Ag+, on the enzyme activity was examined. Ag+ is a sulfhydryl reagent, and its inhibition toward ADK arises from its effect on two particular sulfhydryl groups on the enzyme.13 The inhibition experiments were conducted at 37 °C with the standard assay solution containing 200 µM Mg2+, 200 µM ATP, 300 nM ADK, and Ag+, the concentration of which was varied from 0 mM to 1.0 mM. From the plot of v0 versus [Ag+] (see Supporting Information Figure S-3), it is evident that Ag+ inhibits the ADK activity from 20% to 60% when its concentration varies from 10 µM to 500 µM. While the overall inhibition increases with increasing inhibitor concentration, the inhibition efficiency is largest at low inhibitor concentration. The inhibition experiments provide very strong evidence that the assay is based on the specific ADK catalyzed transphosphorylation reaction. Specificity of the ADK Turn-off Assay. To test the specificity of the ADK turn-off assay, the response of the fluorescence signal to other proteins was examined including peroxidase (HRP), peptidase (PTD), phospholipase D (PLD), glucose oxidase (GOx), and bovine serum albumin (BSA). None of the control proteins has a specific interaction with ATP. In this experiment, 400 µM Mg2+ and 400 µM ATP in the standard assay solution were assayed with 300 nM ADK or one of the control proteins (for PTD the concentration was 48 µg · mL-1). Figure 7a compares the fluorescence intensity changes at 525 nm observed after a 60 min incubation period in the presence of the different proteins, and the inset photographs illustrate the different visual fluorescence response (under UV illumination) that accompanies each control assay. The assay with ADK exhibits more than a 70% decrease in fluorescence intensity, while the assays containing the control proteins (with the exception of BSA) exhibit in the range of 8% ∼ 35% decrease in fluorescence intensity. The decrease in the intensity upon addition of the control protein to the assay solution arises from a non-specific interaction(s) between species in the solution, including the protein, PPECO2, Cu2+, and adenosine Analytical Chemistry, Vol. 81, No. 1, January 1, 2009

237

phosphates (ATP, ADP, and AMP). This effect results in a disturbance of the electrostatic balance and the equilibria which control the polymer’s fluorescence intensity.39 Note that the nonspecific effect is most pronounced when the control protein is charged, which is demonstrated by the ∼70% decrease in fluorescence intensity by negatively charged BSA. The presence of a non-specific effect is further demonstrated by the real time assay with ADK and BSA added individually as shown in Figure 7b. In contrast to the behavior seen for ADK where there is a continuous decrease in PPECO2 fluorescence following addition of the enzyme into the assay solution, addition of BSA induces a rapid decrease in the fluorescence intensity in 1 min; following this initial change, the fluorescence intensity remains constant, as a new electrostatic balance in the solution is attained. Since these control experiments clearly show that non-specific interactions can induce significant fluorescence response, it is also of interest to determine how much non-specific effects that occur in the ADK-containing solution contribute to the overall decrease in the fluorescence intensity during the ADK-catalyzed forward reaction (turn-off assay). The control assay was carried out under the same conditions used for Figure 4b except that no AMP was added. The shape of intensity decrease curve in this control experiment is very similar to that of the BSA curve shown in Figure 7b, and this decrease arises because of the non-specific effect caused by addition of ADK. The “hypothetical” initial rate of reaction, v0c, was calculated from the decrease of fluorescence intensity. The absence of AMP eliminates the catalytic effect of ADK; however, the value of v0c is ∼30% of the value of v0 obtained in Figure 5 at each concentration of ADK. The decrease in fluorescence intensity that is seen upon addition of ADK is likely due to a non-specific effect of ADK on the electrostatic balance among the polymer, metal ions, and ATP. Although the control experiment suggests that the non-specific effect accounts for 30% of fluorescence intensity decrease in the ADK-ATP solution, its contribution in the assay solution with AMP added is believed to be smaller than 30%. This is because the presence of AMP initiates the catalytic function of ADK and increases the specific affinity of ADK toward ATP. As a result, the non-specific effect is likely attenuated. In summary, it is apparent that the decrease in the fluorescence intensity noted in the ADK turn-off assay arises mainly from the specific enzymatic activity of ADK, and nonspecific interactions account for only small component of the fluorescence intensity decrease. ADK Turn-on Assay. To examine the feasibility of using the PPECO2/metal ions/adenosine phosphates complex as a platform for an ADK turn-on assay to detect the reverse reaction as shown in Scheme 1, the effect of added ADK on the polymer’s fluorescence intensity was examined in the presence of Cu2+, Mg2+, and ADP (which in this case is regarded as the substrate for ADK). The results of turn-on assay are illustrated in Figure 8. Initially, the fluorescence of 3 µM PPECO2 is quenched by 10 µM Cu2+ in the presence of 100 µM ADP and 50 µM Mg2+ in 10 mM HEPES buffer (pH 7.5) at 37 °C. (Note that [Mg2+] is 0.5 equiv of [ADP] because the two substrates that occupy the active site of ADK are the MgADP complex and metal-free ADP.1) After introduction of 300 nM ADK, the fluorescence increases as a (39) Dwight, S. J.; Gaylord, B. S.; Hong, J. W.; Bazan, G. C. J. Am. Chem. Soc. 2004, 126, 16850–16859.

238

Analytical Chemistry, Vol. 81, No. 1, January 1, 2009

Figure 8. Change of fluorescence intensity observed during ADK turn-on assay. After addition of 300 nM ADK into the solution containing 3 µM PPECO2, 10 µM Cu2+, 100 µM ADP, and 50 µM Mg2+ in 10 mM HEPES buffer (pH 7.5) at 37 °C, fluorescence intensity increases as a function of incubation time (0-180 min), λex ) 390 nm, λem ) 525 nm.

function of assay time and is doubled in fluorescence intensity after 180 min of incubation. Compared with the ADK turn-off assay, the turn-on assay is considerably less sensitive in terms of rate of reaction and magnitude of the intensity change. This low sensitivity is consistent with our expectation described in the section of “fluorescence recovery by ATP, ADP, and AMP”, where it is shown that ADP is much less effective at inducing recovery compared to ATP, and therefore the transducer system is less sensitive to changes in ADP that occur during the reverse reaction assay. General Discussion. The assay described herein takes advantage of the selective response of the PPECO2-Cu2+ complex to adenosine phosphates and affords the ability to detect ADK activity both in forward and in reverse transphosphorylation reactions. Compared with ADK enzyme coupled assays,13,14 this method offers comparable sensitivity in terms of the amount of enzyme required. However, in contrast to the most popular enzyme coupled assays, the CPE-based assay eliminates the use of other enzymes, preventing any complications arising from their activity. In addition, the CPE-based assay is superior in that considerably less substrate is needed. Specifically, the CPE assay is carried out with the substrate at hundreds of micromolar range which is much lower compared with literature reported range in the other assays.5,14,35-37 Moreover, the PPECO2 based fluorescence assay affords the ability to carry out real time detection of the enzyme activity and to measure kinetic parameters and to quantify the enzyme inhibition. Finally, the assay is relatively easy to implement, and it affords a rapid response. Because the method is based on fluorescence intensity, it can be easily adapted a highthroughput screening (HTS) format using a multi-well plate reader.40 While the PPECO2 based ADK assay offers many advantages, nonetheless this method still has some disadvantages. The existence of non-specific interaction with various proteins and other charged solutes could interfere with the sensor response, especially if quantitative (kinetic) data is need. As a result, when applying this assay with biological samples (e.g., blood serum), it would be necessary to purify the ADK to eliminate the source of non-specific interactions induced by different salts, proteins, or other biological components. The method is also sensitive to the presence of metal sequestering agents (e.g., EDTA) which will interrupt the equilibria that are necessary for fluorescence (40) Hertzberg, R. P.; Pope, A. J. Curr. Opin. Chem. Biol. 2000, 4, 445–451.

signal transduction by the system. Finally, the sulfhydryl groups on ADK may be affected by Cu2+ in the solution, which may lead to a decrease in the activity of the enzyme. Therefore, the amount of Cu2+ needs to be added at a low concentration to maintain activity of the enzyme. CONCLUSION AND OUTLOOK This paper describes the design and application of a novel conjugated polyelectrolyte-based fluorescence assay that affords rapid, sensitive, and real-time detection of ADK activity at very low substrate concentration. The assay solution contains a fluorescent ionic conjugated polymer, Cu2+, Mg2+, and adenosine phosphates (ATP, ADP, and AMP). The sensor takes advantage of the ATP selective fluorescence recovery of Cu2+-quenched conjugated polymer and operates either in a turn-off approach or in a turn-on approach. The ADK fluorescent assay allows for direct calculation of initial rate of reaction, determination the kinetic parameters, and evaluation of enzyme inhibition. The high sensitivity of the sensor arises from the amplified response of fluorescent conjugated polyelectrolyte to the presence of charged fluorescence quenchers.21,22,24,32,41,42 Even though this work displays a fluorescent assay that is specific for a particular substrate/enzyme pair, this method is easy to extend to detect many other enzymatic reactions in which the adenine nucleotides (e.g., ATP) may participate as a substrate, activator, or inhibitor. For example, a similar turn-off assay could be developed for a hexokinase which is capable of transferring a (41) Harrison, B. S.; Ramey, M. B.; Reynolds, J. R.; Schanze, K. S. J. Am. Chem. Soc. 2000, 122, 8561–8562. (42) Wang, J.; Wang, D. L.; Miller, E. K.; Moses, D.; Bazan, G. C.; Heeger, A. J. Macromolecules 2000, 33, 5153–5158.

phosphate group from ATP to a hexose (e.g., D-glucose, Dmannose, and D-fructose) producing ADP and the corresponding sugar-phosphate derivative.43 It also may be possible to utilize this assay in the study of the function of ADK in adenine nucleotide metabolism and cell metabolism and to develop a diagnostic tool for certain ADK-related diseases such as hemolytic anemia.44,45 ACKNOWLEDGMENT We thank Xiaoyong Zhao for synthesis and characterization of PPECO2. We also gratefully acknowledge financial support by the U.S. Department of Energy, Basic Energy Science (Grant DEFG02-03ER15484). SUPPORTING INFORMATION AVAILABLE Details concerning the decrease of [ATP] during the real-time ADK turn-off assays with varying concentration of ADK; nonlinear regression of v0 vs [ATP] data to derive apparent Km and Vmax; inhibition of ADK turn-off assay by Ag+. This material is available free of charge via the Internet at http://pubs.acs.org.

Received for review September 8, 2008. Accepted October 27, 2008. AC801908F (43) Mulcahy, P.; O’Flaherty, M.; Jennings, L.; Griffin, T. Anal. Biochem. 2002, 309, 279–292. (44) Abrusci, P.; Chiarelli, L. R.; Galizzi, A.; Fermo, E.; Bianchi, P.; Zanella, A.; Valentini, G. Exp. Hematol. 2007, 35, 1182–1189. (45) Fermo, E.; Bianchi, P.; Vercellati, C.; Micheli, S.; Marcello, A. P.; Portaleone, D.; Zanella, A. Blood Cells, Mol., Dis. 2004, 33, 146–149.

Analytical Chemistry, Vol. 81, No. 1, January 1, 2009

239