2CSGI, University of Firenze, Italy
3Department of Biochemistry and Molecular Biology I, Complutense University, Madrid, Spain
A coupled enzyme assay able to monitor the kinetics of reactions catalyzed by phosphate- or pyrophosphate-releasing enzymes is presented here. The assay is based on the concerted action of inorganic pyrophosphatase (PPase), purine nucleoside phosphorylase (PNPase), and xanthine oxidase (XOD). In the presence of phosphate, PNPase catalyzes the phosphorolysis of inosine, generating hypoxanthine, which is oxidized to uric acid by XOD. The uric acid accordingly formed can be spectrophotometrically monitored at 293 nm, taking advantage of a molar extinction coefficient which is independent of pH between 6 and 9. The coupled assay was tested using DNA polymerases as a model system. The activity of Klenow enzyme was quantitatively determined, and it was found in agreement with the corresponding activity determined by traditional methods. Moreover, the continuous coupled assay was used to determine Km and Vmax of Klenow enzyme, yielding values in good agreement with previous observations. Finally, the coupled assay was also used to determine the activity of partially purified DNA polymerases, revealing its potential use to monitor purification of phosphate- or pyrophosphate-releasing enzymes.
The detection of inorganic phosphates represents an analytical challenge in the environmental, agricultural, and industrial areas. Accordingly, the relevance of phosphates as analytical targets has prompted the development of chemical methods for their determination. In addition, chemosensors for phosphates were recently designed, synthesized, and tested (1, 2). Using traditional or chemosensor-assisted chemical methods, phosphates can be detected using sensitive, simple, and cost-effective procedures, some of which can also be miniaturized (3) and fully automated (4). Nevertheless, these procedures are discontinuous, rendering rather cumbersome the kinetic evaluation of reactions involving phosphates.
Over the years, different enzymatic methods have been proposed for the detection of inorganic phosphates or pyrophosphates. By coupling the reactions catalyzed by purine nucleoside phosphorylase (PNPase; EC 18.104.22.168) and xanthine oxidase (XOD; EC 22.214.171.124, formerly 126.96.36.199), de Groot et al. were able to quantitatively assay inorganic and organic phosphate (5). Including inorganic pyrophosphatase (PPase; EC 188.8.131.52) in their assay, de Groot et al. were also able to quantitate inorganic pyrophosphate (5). This method relies on the following steps (Supplementary Figure S1): (i) pyrophosphate is converted to phosphate by PPase; (ii) PNPase catalyzes the phosphorolysis of inosine to hypoxanthine and ribose-1-phosphate; and (iii) XOD oxidizes hypoxanthine to uric acid, the absorbance of which can be spectrophotometrically monitored (e.g., at 293 nm). This assay performed well to analyze total phosphate or pyrophosphate concentrations, i.e., by determining the absorbance increase at reaction completion. Unfortunately, de Groot et al. did not investigate in detail the potential use of their method to study the kinetics of reactions catalyzed by enzymes releasing phosphate or pyrophosphate (5). Moreover, they used PNPase (500 mU/mL) in excess over XOD (50 mU/mL), while the opposite should be for this enzyme-coupled assay (6, 7). Similarly, the optimal amount of auxiliary enzymes was not determined for the coupled assay of pyrophosphate relying on sulfate adenyltransferase (8).
A simple kinetic assay for phosphate-releasing enzymes was proposed by Webb (9). In this case, PNPase was used to convert phosphate and 2-amino-6-mercapto-7-methylpurine ribonucleoside (e.g., methylthioguanosine, MESG) into ribose-1-phosphate and 2-amino-6-mercapto-7-methylpurine. At pH 7.6 and 360 nm, the difference in absorbance between MESG and the corresponding base generated by PNPase yields a Δε equal to 11,000 M−1cm−1 (9). Accordingly, the assay proposed by Webb eliminates the need to couple PNPase and XOD reactions. However, this method features the disadvantage that, at pH values below 7.6, the Δε between MESG and the reaction product strongly decreases (9).
More recently, Tagiri-Endo reported on the use of hypoxanthine-guanine phosphoribosyl transferase (HGPRT; EC 184.108.40.206) to detect the pyrophosphate released during the action of DNA polymerases (10). However, this assay was specifically designed to estimate the amount of DNA amplified by PCR, and no attempts were made to evaluate the kinetics of reactions catalyzed by DNA polymerases.
A continuous enzyme-coupled assay of phosphate- or pyrophosphate-releasing enzymes (PREs) is presented here. In particular, the PPase-PNPase-XOD (PPX) system was investigated in detail to: (i) identify the concentration of each auxiliary enzyme necessary to correctly detect the kinetics of reactions catalyzed by PREs; (ii) define the pH interval where the activity of PREs can be determined; (iii) compare the PPX assay with a well known reference method (11). Finally, DNA polymerases were chosen as a model to test the PPX system.
Materials and methods
Escherichia coli PPase (recombinant), bacterial PNPase, bacterial xXOD, type XV activated calf thymus DNA (12), and analytical grade reagents were from Sigma-Aldrich. The Klenow fragment of E. coli DNA polymerase I was from New England Biolabs (Ipswich, MA, USA). Crude E. coli DNA polymerase III holoenzyme was prepared as previously described (13).
Determination of phosphate with Malachite green
The procedure described by Baykov et al. (3) was used as a reference method to detect orthophosphate in aqueous samples. Briefly, a 5x assay mixture was obtained by mixing 10 mL of Malachite green solution (1.2 g in 1 L of 3 M sulphuric acid) with 2.5 mL of a freshly-prepared ammonium molybdate solution (7.5%, w/v in H2O) and 0.2 mL of 10% (v/v) Tween 20. To detect orthophosphate, 40 μL of the 5× assay mixture were added to 160 μL of each sample; upon incubation at 37°C for 30 min, the absorbance at 600 nm was determined using a Bio-Rad (Hercules, CA, USA) 550 microplate reader. To quantitate orthophosphate, an appropriate calibration curve (ranging from 0 to 12.5 μM) was used. The activity of DNA polymerases was determined in the presence of 100 mM Tris-HCl pH 7.8, 65 μg/mL of activated DNA, 10 mM MgCl2, 200 μM dNTPs, 60 mU of PPase. The reactions were stopped by the addition of the Malachite green assay mixture. Controls were carried out in the absence of MgCl2 and in the presence of 10 mM EDTA.
PPase, PNPase, and XOD units (U) are defined as the amount of enzyme able to produce per min, at 25°C, 1 μmol of product at pH 9.0, 7.4, and 7.5, respectively. The activity of DNA polymerases was assayed in the presence of activated calf thymus DNA (65 μg/mL) or in the presence of 2.5 μM poly-dA: oligo-dT (60-mer and 20-mer, respectively). The assay mixture contained 100 mM Tris-HCl pH 7.8, 5 mM MgCl2, 0.25 mM inosine, 0.1 mM dNTPs (or 0.1 mM dTTP), and 5, 50, and 500 mU/mL of PPase, PNPase, and XOD, respectively. The reactions were started by addition of dTTP, and monitored at 293 nm using a PerkinElmer (Waltham, MA, USA) λ19 spectrophotometer. At this wavelength, the ε of uric acid was assumed equal to 12.6 × 103 M−1cm−1 (14).
DNA polymerization reactions were also assayed in the presence of 0.5 mM 2-(4-Iodophenyl)-3-(4-nitrophenyl)-5-phenyl-2H-tetrazolium chloride (INT), and monitored at 460 nm. At this wavelength, the ε of INT-formazan was assumed equal to 12.5 × 103 M−1cm−1 (15).
The E. coli DNA polymerase III catalytic core (comprising the α, ε, and θ subunits) was overexpressed in E. coli TOP10 as previously described (16). Upon overexpression, soluble proteins were extracted (using 50 mM Tris-HCl pH 8, 50 mM NaCl, 1 mM EDTA) and loaded onto a Q-Sepharose FF column. A linear NaCl gradient (from 50 to 600 mM) was then applied to the column, and 70 fractions were collected. The exonuclease activity of the ε subunit was assayed with thymidine 5′-monophosphate p-nitrophenyl ester, as described by Hamdan et al. (17). The polymerase activity of the α subunit was determined using our coupled assay in the presence of 65 μg/mL of activated calf thymus DNA and 0.5 mM INT. Both exonuclease and polymerase activities were assayed (at 420 and 450 nm, respectively) using microplates and a Bio-Rad (Hercules, CA, USA) 550 microplate reader.
Results and discussion
As a first step, we decided to evaluate the presence of phosphates in the reagents that we used to determine DNA polymerase activity. To this aim, each reagent was analyzed by the conventional Malachite green method (3), and the concentration of phosphates was quantitated using a calibration curve, ranging from 0 to 12.5 μM (Abs = 0.019 + 0.046 × [Pi], Supplementary Figure S2A). No significant contamination was detected in Tris-HCl, MgCl2, and calf thymus activated DNA (Supplementary Figure S2B). On the contrary, significant levels of phosphates were determined in dNTPs (purchased from two different suppliers), and in both DNA polymerases tested, i.e., E. coli DNA polymerase III (DNA Pol-III) and Klenow fragment of DNA polymerase I (Supplementary Figure S2C). Therefore, to avoid contamination by phosphates, we extensively dialyzed the two polymerases against Tris-HCl, and we used dNTPs from New England Biolabs. The concentration of phosphates in these dNTPs was indeed found to be very low (≤1%, data not shown). To avoid contamination by phosphates, it is however important to prepare frozen aliquots of dNTP stock solutions. Upon repeated freezing and thawing of these solutions, a significant increase in the concentration of phosphates was observed (data not shown).
To determine the appropriate amount of each enzyme to be used in the coupled assay, we first determined the relevant catalytic constants of XOD and PNPase at pH 7.8 (100 mM Tris-HCl). First, we assayed XOD activity, as a function of hypoxanthine concentration, in the presence of 25 mU/mL of enzyme. Under these conditions, we estimated Km and Vmax equal to 52.7 ± 5.7 μM and 33.5 ± 1.5 nM/s, respectively (Figure 1A). Next, we assayed PNPase activity, as a function of phosphate concentration, in the presence of 5 mM MgCl2, 0.25 mM inosine, 500 mU/mL of XOD and 10 mU/mL of PNPase. Under these conditions, Km and Vmax were estimated equal to 170 ± 8 μM and 285 ± 3 nM/s, respectively (Figure 1B). According to Lee et al. (18), the PNPase supplied by Sigma-Aldrich is isolated from Cellulomonas sp., and features a Km for phosphate equal to 167 ± 22 μM at pH 7.6. Therefore, our use of XOD to assay PNPase activity appears quantitatively reliable, and suggests the optimal concentration of these two enzymes when performing the PPX assay.