Revision of Analytical Conditions for Determining Ligand Molecules

Sep 28, 2009 - Revision of Analytical Conditions for Determining Ligand Molecules Specific to Soft Metal Ions Using Dequenching of Copper(I)−Bathocu...
0 downloads 0 Views 71KB Size
Anal. Chem. 2009, 81, 9199–9200

Revision of Analytical Conditions for Determining Ligand Molecules Specific to Soft Metal Ions Using Dequenching of Copper(I)-Bathocuproine Disulfonate as a Detection System Shinya Ogawa, Rina Ichiki, Mitsuru Abo, and Etsuro Yoshimura* Department of Applied Biological Chemistry, School of Agricultural and Life Sciences, University of Tokyo, 1-1-1, Yayoi, Bunkyo, Tokyo 113-8657, Japan Reoptimization of analytical conditions was performed for a high-performance liquid chromatographic (HPLC) detection system for Cu(I) chelators based on the dequenching of Cu(I)-bathocuproine disulfonate complexes that occurs in the presence of Cu(I) chelators. The revision corrects for emission and excitation wavelengths that were in fact second-order light of the actual optimal wavelengths and for the composition of the postcolumn solution. These revisions resulted in an order of magnitude decrease in detection limits of phytochelatins, a class of cysteine-rich, heavy metal-binding peptides. The revised technique is capable of phytochelatin quantitation at femtomole quantities. In a recent technical note appearing in this journal, we reported a novel detection system for ligands specific to soft metal ions.1 The determination principle is based on the competitive complexation and fluorescence quenching of bathocuproine disulfonate (BCS) with Cu(I) ions. With the use of a postcolumn solution containing Cu(I)-BCS in high-performance liquid chromatographic (HPLC) analyses, the fluorescence intensity of BCS was proportional to the analyte molecule’s ability to bind Cu(I) ions. The method was successfully applied to quantitate phytochelatins, a family of heavy metal-binding peptides with the general structure (γ-Glu-Cys)n-Gly (n G2; PCn) produced by the primitive red alga Cyanidioschyzon merolae. The method possesses a high sensitivity with detection limits comparable to those of thiol group labeling with 3,7-dimethyl-4-bromomethyl-6-methyl-1,5-diazabicyclo-[3,3,3]-octa-3,6-diene-2,8-dione (monobromobimane; mBBr) in HPLC analyses and 1000-fold less than those obtained using a postcolumn reaction with the thiol-specific reagent 5,5′dithiobis(2-nitrobenzoic acid).2 In addition, no pretreatment is required. Relative to most current methods, which are based on a specific structural property of the target molecule (i.e., the presence of a thiol group), this detection system represents a novel analytical methodology in which detection depends solely on the ability of target molecules to bind soft metal ions. * To whom correspondence should be addressed. Phone: +81-3-5841-5153. Fax: +81-3-5841-8027. E-mail: [email protected]. (1) Shirabe, T.; Ito, K.; Yoshimura, E. Anal. Chem. 2008, 80, 9360–9362. (2) Sneller, F. E. C.; van Heerwaarden, L. M.; Koevoets, P. L. M.; Vooijs, R.; Schat, H.; Verkleij, J. A. C. J. Agric. Food Chem. 2000, 48, 4014–4019. 10.1021/ac901782d CCC: $40.75  2009 American Chemical Society Published on Web 09/28/2009

Figure 1. Fluorescence of 40 µM BCS measured at an excitation wavelength of 580 nm in the absence (A) and presence (B) of a YP46 optical filter, which cuts off light of less than 460 nm.

During reoptimization of our experimental setup, it was revealed that the excitation and emission wavelengths employed, which were adapted from reference values,3 were nonoptimal. A contour plot of excitation/emission wavelengths of BCS at pH 9.5 showed that the optimal wavelengths for excitation and emission were 280 and 395 nm, respectively (data not shown). These are far from those used by Sneller et al.2 and in our previous study1 (λex ) 580 nm, λem ) 770 nm). The emission spectrum of 40 µM BCS at an excitation wavelength of 580 nm contains a band at wavelengths greater than 700 nm (Figure 1A). Note that the fluorescence spectrometer employed (FP-6500, Jasco, Tokyo, Japan) limited measurement to wavelengths less than 750 nm. However, upon insertion of a YP-46 optical filter, which cuts off light below 460 nm, either in front of or behind the sample, the band was completely extinguished (Figure 1B). These observations demonstrated that the source light at 580 nm was the secondorder light of 290 nm and that the corresponding emission band at 770 nm was the second-order band of 385 nm. It can therefore be concluded that the nonoptimum conditions resulted from the omission of a YP-46 optical filter. (3) Rapisarda, V. A.; Volentini, S. I.; Farı´as, R. N.; Massa, E. M. Anal. Biochem. 2002, 307, 105–109.

Analytical Chemistry, Vol. 81, No. 21, November 1, 2009

9199

Table 1. Linear Regression Equations and Detection Limits for Quantitating PCs peptides

regression equation

linearity range (nM)

r2

detection limita (nM)

GSH PC2 PC3 PC4

y ) 0.076x + 1.758 y ) 1.605x + 1.887 y ) 2.457x - 41.304 y ) 2.671x + 37.312

100-1000 10-1000 10-1000 10-1000

0.9919 0.9955 0.9982 0.9932

72.6 4.15 1.74 1.27

a The detection limit was defined as the concentration of PCs yielding a peak area corresponding to 3 times the standard deviation of peak areas obtained from three replicate determinations of 500 nM (GSH), 50 nM (PC2), or 20 nM (PC3 and PC4).

On the basis of these findings, the operating conditions were reoptimized using the same HPLC apparatus as employed previously.1 The emission intensity at 395 nm was monitored at an excitation wavelength of 280 nm. The composition of the postcolumn solution was then reduced in such a way as to yield BCS, CuSO4, and ascorbate concentrations of 500 nM, 200 nM, and 5 µM, respectively, buffered at pH 10.0 by 50 mM CHES-NaOH. These solution modifications were necessary because the revised fluorophotometric conditions increased the baseline level of the sample fluorescence. Glutathione (γ-Glu-Cys-Gly) and phytochelatins (PC2, PC3, and PC4) were separated on a C18 column. Following injection of a 20 µL aliquot of sample, the peptides were separated using a 5 min isocratic elution with a mobile phase composed of 5% acetonitrile and 0.1% trifluoroacetic acid (TFA) followed by a 5-30 min concave gradient elution with 5-20% (v/v) acetonitrile in 0.1% TFA. The eluate was merged with the postcolumn solution, through which He was bubbled, passed through a mixing coil, and monitored at an emission wavelength of 395 nm. The analytical figures of merit resulting from the revised conditions are shown in Table 1. A linear response was observed

9200

Analytical Chemistry, Vol. 81, No. 21, November 1, 2009

within a concentration range of 100-1000 nM for GSH and from 10 to 1000 nM for the PCs. Steeper slopes were observed in the response toward peptides with a greater number of γ-Glu-Cys units, as previously reported.1 This is consistent with the increased ability of PCn to bind Cd(II) with increasing n.4 The optimized operating conditions improved detection limits by more than an order of magnitude for each PC peptide. However, no appreciable improvement was evident with GSH. Lowering the levels of BCS and Cu may not have been beneficial for the determination of GSH, likely due to its inherently weaker bonding with Cu(I). The present method yielded a detection limit of 35 fmol per injection for PC3 (0.1 pmol of SH). This value was much less than the reported value of 0.3 pmol of SH per injection obtained by the mBBr method after accounting for low mBBr derivatization efficiencies.2 Although an intrinsic fluorescent compound with excitation and emission wavelengths of 280 and 395 nm, respectively, may be false positive in the present analytical system, a parallel experiment using a postcolumn solution without BCS will preclude such a compound. Indeed, no such signals appeared in the extract of the primitive red alga, C. merolae grown in the presence of Cd (data not shown). Furthermore, the greatly improved detection limits would contribute to our deeper understanding of metal biochemistry of higher plants. Phytochelatin levels would be determined readily in higher plants, even in the absence of metal stresses. This may lead to the elucidation of physiological roles of the peptide in the metabolism of essential metal ions. Received for review August 6, 2009. Accepted September 17, 2009. AC901782D (4) Loeffler, S.; Hochberger, A.; Grill, E.; Winnacker, E.-L.; Zenk, M. H. FEBS Lett. 1989, 258, 42–46.