Quenching and Dequenching of Pyrene Fluorescence by Nucleotide

May 27, 2008 - Francesco Lopez,*,† Francesca Cuomo,† Andrea Ceglie,† Luigi Ambrosone,† and. Gerardo Palazzo*,†,‡. Consorzio InteruniVersit...
0 downloads 0 Views 225KB Size
Download PDF
7338

J. Phys. Chem. B 2008, 112, 7338–7344

Quenching and Dequenching of Pyrene Fluorescence by Nucleotide Monophosphates in Cationic Micelles Francesco Lopez,*,† Francesca Cuomo,† Andrea Ceglie,† Luigi Ambrosone,† and Gerardo Palazzo*,†,‡ Consorzio InteruniVersitario per lo sViluppo dei Sistemi a Grande Interfase (CSGI), c/o Department of Food Technology (DISTAAM), UniVersity of Molise, I-86100 Campobasso, Italy, and Department of Chemistry, UniVersity of Bari, I-70126, Bari, Italy ReceiVed: January 14, 2008; ReVised Manuscript ReceiVed: March 6, 2008

The fluorescence behavior of pyrene solubilized in the hexadecyltrimethylammonium bromide aqueous micellar solution in the presence of adenosine 5′-monophosphate (AMP) and uridine 5′-monophosphate (UMP) was investigated. AMP and UMP were found to influence oppositely the fluorescence of micellized pyrene. UMP acts as quencher, while AMP acts as dequencher. Both effects saturate at high nucleotide concentration (about 40 mM). Dequenching of micellized pyrene fluorescence is induced also by addition of disodium hydrogen orthophosphate (Na2HPO4), while loading with sodium bromide (NaBr) quenches the fluorescence. Furthermore, in absence of micelles, pyrene fluorescence depends on the UMP, according to the Stern-Volmer relation, but is unaffected by AMP. Dynamic light scattering experiments showed that the size and shape of aggregates is not affected by different types of nucleotide loaded into the solution; thus, we conclude that the opposite photophysical effect exploited by AMP and UMP are uncorrelated to any change in micellar microstructure. The whole fluorescence data set was successfully accounted for by assuming that the anionic nucleotides compete with the surfactant counterion (bromide) for the surface of the micelle. Accordingly, substitution of bromide with the more effective quencher UMP results in a strong decrease of the pyrene fluorescence, while the substitution of bromide with the nonquencher AMP results in an increase in the pyrene fluorescence. 1. Introduction The studies on interactions between DNA and self-assembled nanostructures, such as micelles and vesicles, have recently received growing interest, due to the possibility to use these complexes as transfection tools.1 In particular, the interaction between cationic surfactants and DNA has earned special attention.2,3 Often, these attraction interactions lead to polynucleotide compaction1,2 reminiscent of the DNA-histone packing in the nuclei of living cells.4 At a lower degree of complexity, the electrostatic interactions between the positively charged surface of micelles formed by cationic surfactant and the negatively charged mononucleotides are expected to lead to the formation of complexes. The quantification of the affinity of mononucleotides for the micellar wall could be useful for the analysis of the compaction behavior of polynucleotides and for some applications involving surfactantmononucleotide interactions (e.g., mononucleotide determination by means of capillary electrophoresis).5 Furthermore, by studying the interactions of macromolecules, such as nucleotides and cationic surfactants, significant information, as to the recognition, sensing of anionic analytes, and supramolecular chemistry could be obtained.6 To this purpose, we have undertaken a study of the interactions between positively charged hexadecyltrimethylammonium bromide (CTAB) micelles and nucleotide monophosphates (NMPs) by following the fluorescence behavior of pyrene in * To whom correspondence should be addresed. Phone: +39 0874404632. Fax: +39 0874404652. E-mail: [email protected] (F.L.); palazzo@chimica. uniba.it (G.P.). † University of Molise. ‡ University of Bari.

CTAB micelle in the presence of NMPs. Pyrene was extensively used as a fluorescent probe to investigate several aspects of micellar features such as aggregation number7,8 and physicochemical properties of its surroundings.9–12 It is easily solubilized by CTAB micelles9 so that any change in its fluorescence in the presence of water soluble mononucleotides can be reasonably correlated to the interaction of NMP with the micelles. It turned out that the two nucleotides used, viz., adenosine 5′-monophosphate (AMP) and uridine 5′-monophosphate (UMP), influenced oppositely the fluorescence of micellized pyrene. The latter (UMP) efficiently quenches the pyrene emission, while the former (AMP) enhances it. As described below, the detailed analysis of steady state fluorescence response to the presence of AMP, UMP, and other relevant ions (bromide and orthophosphate) gives insight into the competitive adsorption of these anions on the positively charged micellar surface. 2. Experimental Section AMP (g99%, disodium salt), UMP (>99%, disodium salt), CTAB (g99%), sodium bromide (NaBr), disodium hydrogen orthophosphate (Na2HPO4), Tris hydrochloride, and pyrene were from Sigma. The chemicals have been used as received. Since the self-assembly properties of the surfactant and of the NMPs are strongly dependent on the salt concentration, all the experiments were performed in a 10 mM NaCl, 1 mM Tris/HCl pH 7.2 buffer solutions, as suggested by Petrov.13 Solutions were prepared by adding suitable amounts of the components in volumetric flasks (pyrene was added as methanol solutions). For all the micellar solutions the final concentration of pyrene was 5 × 10-7 M (the aqueous solubility of pyrene is 3.9 × 10-7 M 14). All fluorescence measurements were per-

10.1021/jp8003344 CCC: $40.75  2008 American Chemical Society Published on Web 05/27/2008

Micelle-Nucleotide Interactions

J. Phys. Chem. B, Vol. 112, No. 24, 2008 7339

formed using a Varian Eclipse spectrofluorimeter in a 1-cm quartz fluorescence cuvette, at 25 °C. The measurements were performed on samples after an incubation time of 30 min. The excitation and the emission slit widths were 5 mm. The excitation and the emissions wavelengths utilized for this study were 334 and 374 nm, respectively. The hydrodynamic diameter of the micelles was determined by means of dynamic light scattering measurements using a Malvern commercial instrument Zetasizer-Nano ZS90 operating with a 4 mW He-Ne laser (633 nm wavelength) and fixed detector angle of 90°; the average micellar size was estimated by a cumulant analysis of the autocorrelation function using the software provided by the manufacturer. The critical micelle concentration was determined through conductivity measurements15,16 using a CDM230 conductivity meter (Radiometer Analytical) equipped with a two-pole conductivity cell tailored for small volumes CDC749 (cell constant 1.84 cm-1) and calibrated with KCl solution in the appropriate concentration range. The conductivity was measured at 25 °C after an equilibration time of 4 h. 3. Results 3.1. Steady State Quenching in Aqueous Buffer. Pyrene is one of the most widely used neutral fluorescence probes. An advantage of using pyrene as a fluorescence probe is its high quantum yield that allows its use at low concentrations, thus avoiding the presence of the excimer emission. The NMPs used in the present investigation are the purine-based AMP and the pyrimidine-based UMP. Typical fluorescence spectra of pyrene in aqueous buffer (10 mM NaCl, 1 mM Tris/HCl pH 7.2) upon addition of UMP and AMP are depicted in Figure 1 (panels a and b, respectively). Clearly, addition of UMP strongly decreases the fluorescence intensity of pyrene (Figure 1a), whereas addition of AMP produces only very small reduction of pyrene emission (Figure 1b). For the system reported in Figure 1 (pyrene and NMPs molecularly dissolved in aqueous buffer), we can reasonably assume that the quenching results from encounters between pyrene and nucleotides.13,17–20 In such a case, the decrease of fluorescence intensity is related to the concentration of the quencher by the Stern-Volmer equation

F0 ) 1 + K’SV[Q] F

(1)

where F and F0 stand for the fluorescence intensity in the presence and in the absence of the quencher, respectively; [Q] indicates the quencher concentration, and K′SV represents the Stern-Volmer constant, which has dimensions of reciprocal concentration. In the inset of Figure 1a F0/F is plotted as a function of nucleotide concentration. According to eq 1, the ratio F0/F is linearly dependent on the NMP concentration, but a considerable difference between the two nucleotides in terms of K′SV is evident (UMP being a quencher much more effective than AMP). Although some details are still not fully resolved (for example the role of protons) because the redox potentials of nucleosides in water are not well established since they exhibit irreversible oxidation in aqueous media, there is a general consensus on the idea that the quenching of pyrene emission by nucleosides, nucleotides, and nucleic acids occurs via a photoinduced electron transfer.19,21 According to the work of Seidel and co-workers22 (in which the redox potentials for nucleobases have been determined in aprotic solvents and corrected for an additional water-specific offset), the observed sequence of quenching efficiency is consistent with a photoinduced electron transfer in which UMP is more easily reduced by the photoexcited pyrene than AMP and confirm that the

Figure 1. Fluorescence emission spectra (excitation at 334 nm) of pyrene dissolved in 1 mM Tris/HCl pH 7.2 10 mM NaCl upon addition of sodium salt of UMP (panel a) and AMP (panel b). The arrows indicate the evolution of emission spectra upon increase in the NMP concentration. The inset shows the Stern-Volmer plots (emission at 374 nm) for both nucleotides; the lines are the linear regression according to eq 1 for UMP (K′SV ) 218 M-1) and for AMP (K′SV ) 6 M-1).

effectiveness of quenching depends on the nucleotide chemical structure. 3.2. Steady State Quenching in Micellar Solution. The effects of the addition of UMP and AMP to pyrene solubilized in CTAB micellar solution were next investigated (parts a and b of Figure 2). The representative emission spectra displayed in Figure 2 refer to 10 mM CTAB micelle solution loaded with different concentration of nucleotides (0-100 mM). The spectra of Figure 2 show that the loading of both nucleotides has a significant influence on the intensity of pyrene fluorescence in the micellar solution in opposite ways. The pyrene fluorescence dropped significantly in the presence of UMP (Figure 2a). On the contrary, the fluorescence intensity was enhanced by AMP loading (Figure 2b). In both cases, the ratio between emissions at different wavelengths remains constant, suggesting that the polarity of pyrene surroundings is unaffected by NMPs loading. The opposite effect of UMP and AMP on pyrene fluorescence is better appreciated analyzing the dependence of fluorescence emission at 374 nm on the NMP concentration, as shown in Figure 3 for two CTAB concentrations. It is evident that UMP acts as a quencher, while AMP acts as a dequencher. Furthermore, it is clear that the NMP-specific effect saturates at high nucleotide concentration. It should also be noted that the fluorescence intensity in absence of NMP decreases upon increase in the surfactant concentration. The fluorescence plateau reached at high NMP loading depends on the CTAB concentra-

7340 J. Phys. Chem. B, Vol. 112, No. 24, 2008

Figure 2. Fluorescence spectra (excitation at 334 nm) of pyrene (5 × 10-7 M) solubilized in 10 mM CTAB micellar solution upon addition of sodium salt of UMP (panel a) and AMP (panel b). The arrows indicate the evolution of emission spectra upon increase in the NMP concentration ([NMP]) were: 0, 3, 5, 10, 50, 75, and 100 mM. The insets show the plots of FAPP 0 /F vs [NMP]; see discussion for details.

tion in the case of UMP (the plateau value decreases with the surfactant concentration) but not in the case of AMP. The effect of AMP and UMP on pyrene fluorescence appears at least to be intriguing. Usually, increments of fluorescence phenomena are ascribed to a reduction of self-absorption, due to a decrease in the local probe concentration or to changes in the probe surrounding. However, these mechanisms can hardly account for the opposite effect of the two nucleotides akin. In principle, the peculiar photophysical response of pyrene to NMP loading could be related to changes in the microstructure of micellar solution. Both the NMPs are anions and are expected to interact strongly with the cationic surfactant. To clarify this point, we have investigated how the presence of relevant ions may influence the CTAB critical micelle concentration (CMC). Conductivity measurements indicated that the buffer (NaCl 10 mM, Tris/HCl 1 mM pH 7.2) by itself lowered the value of CMC significantly (from 0.9 mM in pure water to 0.22 mM in buffer; see Table 1). Addition of 10 mM AMP or UMP or hydrogen orthophosphate ions results only in a slight (20%) increase in this value. Further NMP loading (50 mM) decreases the CMC to 0.13 mM for AMP and 0.11 mM for UMP (see Table 1). The decrease of the CMC of ionic surfactants upon addition of electrolytes is a well-understood phenomenon, and the data of Table 1 indicate that under all the conditions tested more than 90% of the surfactant is in the micellized form. Since the CMC value is related to the free energy change for the surfactant micellization,23 no significant changes where detected

Lopez et al.

Figure 3. Fluorescence intensities at 374 nm as a function of NMP concentration for two CTAB concentrations (O) CTAB 2 mM, (2) CTAB 10 mM, [pyrene] ) 5 × 10-7 M. (a) Emission in presence of UMP; symbols denote experimental data and lines the nonlinear best fit according to eq 8b (best fit parameters are listed in Table 2). The inset shows the linear dependence of the reciprocal fluorescence intensity in absence of NMP (FAPP 0 , see text) on the CTAB concentration; the slope of the best-fit straight line is 15 ( 1 M-1. (b) Emission in presence of AMP; symbols denote experimental data and lines the nonlinear best fit according to eq 6 (best fit parameters are listed in Table 2). The dashed horizontal line denotes the value of fluorescence at infinite AMP concentration (F0, see discussion).

TABLE 1: CMC of CTAB in Different Solutions CMC pure water buffer buffer + 10 buffer + 10 buffer + 10 buffer + 50 buffer + 50

mM mM mM mM mM

AMP UMP Na2HPO4 AMP UMP

(mM)

0.90 ( 0.05 0.22 ( 0.02 0.27 ( 0.02 0.29 ( 0.02 0.26 ( 0.02 0.13 ( 0.02 0.10 ( 0.02

in the CTAB propensity to self-assembly in the presence of AMP or UMP. Furthermore, the free energy decrease upon micellization in the presence of orthophosphate is the same as in the presence of NMPs, suggesting that micelles counterion interactions have mainly an electrostatic origin. 3.3. Effects of Inorganic Salts on Pyrene Fluorescence in CTAB Micellar Solution. Specific surfactant-counterion interactions could also affect the surfactant packing parameter leading to changes in the micellar size and shape. This point could be relevant for fluorescence studies; for example, it was reported that an increase in ionic strength results in a bimodal size distribution of sodium dodecylsulfate (SDS) micelle containing fluorophores obfuscating the quenching data.24 We performed dynamic light scattering measurements on the micellar solutions at 10 mM CTAB concentration upon loading

Micelle-Nucleotide Interactions

J. Phys. Chem. B, Vol. 112, No. 24, 2008 7341

TABLE 2: Parameters Describing the Effect of UMP and AMP on the Fluorescence of Micellized Pyrene According to Eqs 8a and 6, Respectively UMP

AMP

CTAB (mM)

a FAPP 0

KUMP (M-1) L

APP KSV

UMPb KSV

F0a

KAMP (M-1) L

KBr[Br-]

2 5 7.5 10

221 213 205 197

310 ( 42 195 ( 21 152 ( 22 180 ( 30

0.62 ( 0.027 0.42 ( 0.013 0.30 ( 0.012 0.27 ( 0.012

1.09 ( 0.05 0.90 ( 0.03 0.81 ( 0.04 0.82 ( 0.05

284 288 283 283

213 ( 32 153 ( 20 189 ( 46 67 ( 15

0.27 ( 0.01 0.33 ( 0.01 0.36 ( 0.02 0.39 ( 0.02

Fluorescence values depend on spectrofluorimeter features; the values of FAPP and F0 are listed here only to highlight the different 0 UMP APP dependence on CTAB concentration. b Evaluated as KSV ) (1 + KSV )F0/FAPP - 1 (according to eqs 4 and 8aa) and assuming F0 ) 283 as for 0 AMP. a

Figure 4. Dependence of the average hydrodynamic diameter (from dynamic light scattering) of CTAB micelles on different solute concentrations (viz., UMP, AMP, NaBr, Na2HPO4). The CTAB concentration was held constant at 10 mM.

of relevant anions, viz., AMP, UMP, hydrogen orthophosphate, and bromide (all as sodium salts). In all cases, the autocorrelation function of the scattered light can be well accounted for by a monomodal size distribution (data not shown). As shown in Figure 4, the increase in ionic strength induces an increase in the average micellar hydrodynamic diameter from 4.4 nm in absence of exogenous salt to a plateau around 7 nm at high salt content. What matters for the purpose of the present study is that no substantial differences were found in the dependence of the micellar size on the anion concentration among the four anions investigated. Since the hydrodynamic diameter reflects both micelle’s size and shape, we conclude that the opposite photophysical effect exploited by AMP and UMP are uncorrelated to any change in micellar microstructure. To have insight on the specific role of the negatively charged moiety of the NMPs, we investigated the pyrene fluorescence in CTAB micellar system upon addition of bromide ion (NaBr) and of Na2HPO4. The latter should mimic the negatively charged group of NMPs, while the former is the native counterion of the cationic surfactant used. In Figure 5 the effects of bromide and orthophosphate on pyrene fluorescence are compared to the UMP and AMP behaviors in a 2 mM CTAB micellar solution. From these data two distinct kinds of interactions are clearly noticed. The pyrene fluorescence appeared to be dequenched by hydrogen orthophosphate ions, similarly to what it is obtained with AMP ions. On the contrary, bromide ion quenches the pyrene fluorescence like UMP. The same results have been obtained in 10 mM CTAB solution (data not shown). This suggests that the decrease in fluorescence intensity observed upon increasing the CTAB

Figure 5. Comparison among the effects of bromide, orthophosphate, AMP, and UMP loading on pyrene fluorescence in a 2 mM CTAB micellar solution. Conditions: emission wavelength ) 374 nm, [pyrene] ) 5 × 10-7 M. The fluorescence dequenching induced by AMP and orthophosphate was fitted to eq 6; the fluorescence quenching induced by UMP was fitted to eq 8a. Lines represent the corresponding nonlinear fits.

concentration is related to the parallel increase of bromide counterions. Eventually, we noted that the quenching extent induced by UMP is larger than that obtained by loading NaBr at the same surfactant concentration. 4. Discussion As to the micellar solutions, the ratio between the fluorescence intensity in absence of NMPs (hereafter FAPP 0 ) and the fluorescence in presence of NMP does not depends linearly on the NMPs concentration but follows a saturation trend (see the insets of parts a and b of Figure 2). Saturation of quenching effectiveness is usually ascribed to mechanisms that somehow restrict the encounters between the probe and the quencher. Segregation of the probe in regions inaccessible to the quencher leads to hyperbolic quenching curves;25 this is often found in biological systems, where a fraction of the probe molecules (e.g., tryptophan) is buried within the protein, yet this mechanism does not hold in the present case, because pyrene within micelles is readily available to quenchers, as demonstrated by several studies26 (note also that pyrene is believed to be localized in the “palisade” of the micelle).9 A different mechanism considers the presence of a physical limit in the number of quencher molecules that can encounter the probe. This is relevant to probes embedded in micelles where the number of quenchers surrounding a micelle is limited by steric and/or electrostatic arguments. In such a case, the micellar surface can be described in terms of a limited number of “binding sites” for the quencher

7342 J. Phys. Chem. B, Vol. 112, No. 24, 2008 leading to the saturation of the fluorescence quenching.27 For CTAB micelles, it is reasonable to assume that the anionic quencher UMP strongly interacts with the positively charged micellar surface until all the surfactant cations are neutralized by UMP anions. Indeed, the fluorescence of pyrene solubilized in micelles formed by the anionic surfactant SDS is totally unaffected by the presence of UMP or AMP (data not shown). The peculiar fluorescence response of the micellized pyrene upon addition of AMP can be rationalized, keeping in mind that the micellar solutions always contain the surfactant counterions (bromide anion). It is well-known that bromide is a good quencher for many fluorophores, and also in the present case addition of exogenous NaBr resulted in a decrease of pyrene fluorescence (see Figure 5). On the other hand, the effectiveness of AMP as a quencher of the pyrene fluorescence was negligible in absence of CTAB micelles (Figure 1). Thus, if AMP displaces the bromide ions, the photophysical response should be the dequenching of pyrene fluorescence, due to the replacement of a quencher with a nonquencher molecule. In other words, the results of previous sections can be explained tacking into account the competition between bromide and NMPs for the cationic micellar surface and the different photophysical properties of the two NMPs. We will next discuss these concepts on a quantitative ground. Let us assume the bromide and NMP ions interact with the positively charged micellar surface, as shown below

Br- + S+ a Br - S and NMP- + S+ a NMP - S where S+ indicates the adsorption sites available for the adsorbate on the micellar surface. The interaction between anions and CTAB micelles may be treated as classical Langmuir adsorption isotherms

[Br - S] [NMP - S] ) KLBr and ) KLNMP + + [Br ][S ] [NMP ][S ] where KLi stands for the Langmuir’s constant for the ith species. The competition between bromide and NMP is given by the constraint: [ST] ) [S+] + [Br-S] + [NMP-S], where [ST] denotes the concentration of total adsorption sites. This formalism is equivalent to the competitive binding of two ligands for the binding sites of a macromolecule (in such a case KBr L and KNMP stands for the respective binding constant). L Of course, the adsorption/binding of the ith species to micelles leads to observable fluorescence effects only if the micelle contains pyrene. Thus, the contribution of a species, say NMP, to the observed fluorescence quenching is given by the concentration of NMP adsorbed to the micelles (i.e., [NMP-S] ) KNMP [NMP][ST]/(1 + KNMP [NMP] + KBr L L L [Br ])) multiplied by the fraction of micelle containing pyrene, i.e., (pyrene concentration)/(micelle concentration). The concentration of total binding sites for the quencher is proportional to the micelle concentration, and in a first-order approximation, each surfactant head represents a potential binding site ([ST] ≈ Nagg(micelle concentration) ≈ [CTAB]; Nagg being the surfactant aggregation number). Therefore, the concentrations of species effectively probed by a steady state fluorescence quenching experiments are

Lopez et al.

[NMP - S] ) [Br - S] )

KLNMP[NMP][pyrene]Nagg 1 + KLNMP[NMP] + KLBr[Br-] KLNMP[Br-][pyrene]Nagg

1 + KLNMP[NMP] + KLBr[Br-]

By assumption that the above concentrations determine the fluorescence quenching through the Stern-Volmer relation, eq 1 develops into Br NMP F0 [Br-]KLBr + KSV [NMP]KLNMP KSV )1+ F 1 + KBr[Br-] + KNMP[NMP] L

(2)

L

i where the adimensional KSV constants include the pyrene i concentration and the micelle aggregation number (KSV ) i K′SV Nagg[pyrene]). It should be noted that the experimental fluorescence value in absence of NMP does not correspond to the true values of the pyrene fluorescence without quencher (F0). It corresponds to F0APP, i.e. the fluorescence of sample without the nucleotide contribution but still in the presence of Br-. In the face of it, when nucleotides are absent ([NMP] ) 0) the following relation holds

Br KSV [Br-]KLBr ) 1 + FAPP 1 + KLBr[Br-] 0

F0

(3)

The above equation describes the variation of fluorescence intensity while increasing the CTAB concentration. In the explored CTAB range, the reciprocal of FAPP increases linearly 0 with the CTAB (Br-) concentration (see inset of Figure 3a). This suggests that the term KBr L [Br ]is negligible with respect to 1 and thus eq 3 becomes

F0 FAPP 0

) 1 + KBr[Br-]

(4)

Br Br where KBr ) KSV KL . Linear regression of the data of the inset of Figure 3a gives K Br ) 15 ( 1 M-1. By consideration of the previous adjustments, eq 2 can be rewritten as

NMP F0 [NMP]KLNMP KBr[Br-] + KSV )1+ F 1 + KNMP[NMP]

(5)

L

Equation 5 well describes the effect on pyrene fluorescence of the competition between NMPs and bromide ions for the adsorption sites of the micellar surface. The different (quenching and dequenching) behavior observed for different NMPs can NMP be traced out to their different photophysical properties: KSV NMP ≈ 0 (NMP does not quench pyrene fluorescence) and KSV > Br KSV (NMP is a good quencher for pyrene). NMP The first case (KSV ) 0) applies to AMP. Note that in this case F0, i.e., the fluorescence value in absence of any quencher, is physically accessible at (nonquencher) NMP concentration high enough to displace the quencher bromide ions (strictly [NMP] ) ∞; see Figure 3b). With this in mind, eq 5 becomes

F0 KBr[Br-] )1+ F 1 + KNMP[NMP]

(6)

L

Accordingly, the fluorescence increases upon AMP loading until reaches its limit value (F ) F0), because the fluorescence unproductive AMP displaces the quencher Br- from the micellar surface with a mechanism akin to the action of a competitive inhibitor on an enzyme. Representative fit of fluorescence values

Micelle-Nucleotide Interactions

J. Phys. Chem. B, Vol. 112, No. 24, 2008 7343

FAPP [NMP]KLNMP 0 APP ) 1 + KSV F 1 + KNMP[NMP]

(8b)

L

The above equation well describes the experimental decrease and saturation of pyrene fluorescence upon addition of UMP as shown for representative CTAB concentrations in Figure 3a; NMP APP the best-fit parameters (FAPP , and KSV ) determined for 0 , KL different CTAB concentrations are listed in Table 2. Equation 8a can be easily linearized in term of the ratio FAPP 0 / ∆FAPP (∆FAPP being FAPP - F) giving 0

FAPP 0 APP

∆F

Figure 6. Linearization of fluorescence data for UMP and AMP (see discussion) for four CTAB concentrations: (9) 2 mM, (O) 5 mM, ([)7.5 mm, (4)10 mM. (a) Plots of F0APP/∆FAPP vs 1/[UMP]; lines correspond to the prediction of eq 9 (the parameters are listed in Table 2). (b) Plots of F0/∆F vs [AMP]; lines correspond to the prediction of eq 7 (the parameters are listed in Table 2).

in presence of AMP to eq 6 are shown in Figure 3b; the best-fit parameters (F0, K Br[Br-], and KNMP ) determined for different L CTAB concentrations are listed in Table 2. The above equation can be easily linearized introducing ∆F ) F0 - F

F0 KLNMP[NMP] 1 + Br - + 1 ) ∆F KBr[Br-] K [Br ]

(7)

Consistent with the above equation, plotting F0/∆F vs [AMP] results in straight lines (see Figure 6b); this representation is formally equivalent to the Dixon plot for the binding of competitive inhibitor to an enzyme.28 NMP The second case (KSV * 0) applies to nucleotides that successfully quench the pyrene fluorescence such as UMP. Note that, if the NMP is a good quencher, it is not possible to obtain experimentally F0. On the contrary, what one can easily measure is the pyrene fluorescence in absence of NMP (but still in presence of Br-). By combination of eqs 3 and 5 and introduction of the parameter APP KSV )

NMP 1 + KSV

1 + KBr[Br-]

-1

(independent of NMP concentration), we have

(8a)

)

1

1

APP NMP [NMP] KSV KL

+ (1 +

1 APP KSV

)

(9)

APP is Consistent with the above equation, the ratio FAPP 0 /∆F linearly dependent on the reciprocal UMP concentration as for classical quenching equations or for the Lineweaver-Burk plot in enzyme kinetics (see Figure 6a). Equations 6 and 8a well describe the experimental behavior of AMP and UMP, respectively, and have been used to determine the parameters listed in Table 2 by means of nonlinear best fits. On the other hand, the linear eqs 7 and 9 are useful to highlight the different response of pyrene fluorescence to the binding of AMP and UMP by a simple visual inspection of the data. In Figure 6 the fluorescence results obtained at different CTAB concentration for both nucleotides according to the formalism relevant to the eqs 7 and 9 are displayed. In the case of nucleotides preventing the quenching caused by bromide the ratio F0/∆F is linearly dependent on NMP concentration (AMP, panel b) with slopes that clearly decrease with the increase in the CTAB concentration consistent with eq 7 ([Br-] is proportional to the CTAB concentration). On the contrary, in the case of nucleotide acting as a true quencher (UMP, panel a) the ratio APP depends linearly on the reciprocal UMP concentraFAPP 0 /∆F tion with a slope that strongly increases with the CTAB concentration in agreement with the prediction of eq 9 because the parameter KAPP SV is expected to decrease with [Br ] (eq 8aa). In Figure 7a the values of the Langmuir’s constant for the adsorption of UMP, AMP, and orthophosphate obtained at different CTAB concentrations are compared. As described in the results section, the behavior of Na2HPO4 is analogous to that of AMP; thus, the pyrene fluorescence at different orthophosphate loading was fitted to eq 6 (see Figure 5 for a representative fit). AMP, UMP, and Na2HPO4 share essentially the same values of Langmuir’s constant indicating that the adsorption to the micellar surface is mainly due to the contribution of the phosphate moiety, likely through electrostatic interactions with the cationic surfactant head. The other best-fit parameters listed in Table 2 deserve some words of comment, in particular their dependence on the CTAB concentration. The parameter F0 determined in experiments performed with AMP is independent from CTAB concentration as expected for an intrinsic property of the pyrene probe. On the contrary, FAPP coming from experiments with UMP de0 creases with CTAB concentration consistent with eq 4. The term K Br[Br-] has been evaluated from the UMP data as F0/FAPP 0 1 (according to eq 4; the FAPP values have been determined by 0 fitting the fluorescence data in presence of UMP to eq 8a and assuming F0 ) 283). The same parameter was directly obtained from the fit of the AMP data to eq 6 (Table 2). The values of KBr [Br-] determined from independent experiments are compared in Figure 7b. The agreement between the K Br[Br-] values coming from the analysis of UMP-induced quenching and of AMP-induced dequenching is excellent. The parameter K Br[Br-]

7344 J. Phys. Chem. B, Vol. 112, No. 24, 2008

Lopez et al. for by assuming that the NMPs compete with the surfactant counterion (bromide) for the surface of the micelle and that the effectiveness of quenching follows the scale UMP > Br- . AMP. The reliability of this interpretation is supported by the quantitative consistency between the values of K Br estimated by analyzing the dependence of pyrene fluorescence on the concentration of UMP and AMP and the dependence of pyrene fluorescence on the CTAB concentration in absence of NMPs. The obtained results indicate that UMP, AMP, and orthophosphate have similar affinities for the cationic surface of cetyltrimethylammonium micelles. The quantitative rationalization of the effects of a switch between quencher and nonquencher micellar counterions allows a detailed study of the interactions between nucleotides and CTAB micelles and could also be highly valuable for sensing applications allowing the use of simple spectrofluorimetric assays for the determination of species that do not have any influence on the fluorescence of the micellized probe. Acknowledgment. This work was supported by the Italian Ministero Universita` Ricerca (PRIN 2006: “Self-assembling nanosystems with DNA-like addressability”), by CNR (project FUSINT CSGI-CNR 2007) and by Consorzio Interuniversitario per lo sviluppo dei Sistemi a Grande Interfase (CSGI-Firenze). References and Notes

Figure 7. (a) Langmuir’s constant for the adsorption of UMP, AMP, and orthophosphate on micellar surface at different CTAB concentrations. (b) Dependence of the product K Br[Br-] on CTAB concentration. Open circles refer to values obtained by fitting the fluorescence data in presence of AMP to eq 6 (representative fits are shown in Figure 3. Closed squares refer to values calculated according to eq 4; the FAPP 0 values have been determined by fitting the fluorescence data in presence of UMP to eq 8 (representative fits are shown in Figure 3a) and assuming F0 ) 283. The straight line is the linear regression of the whole data set; intercept ) 0.25 ( 0.01 and slope ) 17 ( 2.

increases linearly with CTAB concentration because [CTAB] ) [Br-]. The slope values of the straight line is K Br ) 17 ( 2 M-1 and indicates a good consistency with the dependence of pyrene fluorescence on CTAB concentration shown in the inset of Figure 3a. Note that the intercept at [CTAB] ) 0 is not null. We believe that this is related to the presence of a fixed chloride concentration in the buffer used (Cl- is reported to be a weak quencher of fluorescence for many fluorophores).29 Note that the use of NaCl instead of NaBr enhances the effect of UMP (data not shown) because large bromide concentration tends to saturate the quenching of pyrene emission. Finally, the knowledge of the K Br[Br-] values has been profitably used to deterUMP APP mine KSV through eq 8a (using the KSV coming from the fit UMP of UMP data to eq 8b). The dependence of KSV on CTAB concentration is very weak, if any. 5. Conclusions Although akin in chemical structure, AMP and UMP were found to influence the fluorescence of pyrene secluded in CTAB micelles oppositely. UMP acts as quencher, while AMP acts as dequencher. Both the effects saturate at high NMP concentration (about 40 mM). The whole data set was successfully accounted

(1) Dias, R. S.; Pais, A.A.C.C.; Miguel, M. G.; Lindman, B. Colloids Surf. A 2004, 250, 115. (2) Dias, R.; Mel’nikov, S.; Lindman, B.; Miguel, M. G. Langmuir 2000, 16, 9577. (3) Cardenas, M.; Braem, A.; Nylander, T.; Lindman, B. Langmuir 2003, 19, 7712. (4) Horn, P. J.; Peterson, C. L. Science 2002, 297, 1824. (5) Fu, C.; Song, L.; Fang, Y. Anal. Chim. Acta 1999, 399, 259. (6) Schmidtchen, F. P.; Berger, M. Chem. ReV. 1997, 97, 1609. (7) Lianos, P.; Zana, R. J. Phys. Chem. 1980, 84, 3339. (8) Lianos, P.; Zana, R. J. Phys. Chem. 1983, 87, 1289. (9) Lianos, P.; Viriot, M. L.; Zana, R. J. Phys. Chem. 1984, 88, 1098. (10) Gould, I. R.; Kuo, P. L.; Turro, N. J. J. Phys. Chem. 1985, 89, 3030. (11) Yihwa, C.; Kellerman, M.; Becherer, M.; Hirsh, A.; Bohne, C. Photochem. Photobiol. Sci. 2007, 6, 525. (12) Capek, I. AdV. Colloid Interface Sci. 2002, 97, 91. (13) Petrov, A. I.; Khalil, D. N.; Kazaryan, R. L.; Savintsev, I. V.; Sukhorukov, B. I. Bioelectrochemistry 2002, 58, 75. (14) Yoshida, N.; Moroi, Y.; Humphrry-Baker, R.; Gra¨tzel, M. J. Phys. Chem. A 2002, 106, 3991. (15) Evans, H. C. J. Chem. Soc. 1956, 579. (16) Shanks, P. C.; Franses, E. I. J. Phys. Chem. 1992, 96, 1794, and references therein. (17) Ohtani, M.; Masuko, Y.; Wada, T.; Kodama, T. Nucleic Acid Symp. 2000, 44, 51. (18) Lianos, P.; Georghiu, S. Photochem. Photobiol. 1979, 29, 13. (19) Shafirovich, V. Y.; Courtney, S. H.; Ya, N.; Geacintov, N. E. J. Am. Chem. Soc. 1995, 117, 4920. (20) Okamoto, A.; Saito, Y.; Saito, I. J. Photochem. Photobiol. C 2005, 6, 108. (21) Manoharan, M.; Tivel, K. L.; Zhao, M.; Nafisi, K.; Netzel, T. M. J. Phys. Chem. 1995, 99, 17461. (22) Seidel, C. A. M.; Schulz, A.; Sauer, M. H. M. J. Phys. Chem. 1996, 100, 5541. (23) Evans, D. F. ; Wennerstro¨m, H. The Colloidal Domain, 2nd ed.; Wiley-VCH: New York, 1999; Chapter 4, p 153. (24) Siemiarczuk, A.; Ware, W. R.; Dong, W J. Fluor. 1997, 7, 1955. (25) Lakowicz, J. R. Principles of Fluorescence Spectroscopy, 3rd ed.; Springer: 2006; Chapter 8, p 278. (26) Zana, R. Surfactant Solutions; Zana, R. , Ed.; New York, Chapter 5, p 241. (27) Crumb, D. T.; Beck, S. C. J. Photochem. Photobiol. A 2000, 134, 87. (28) Dixon, M. Biochem. J. 1953, 55, 170. (29) Saha, S. K.; Krishnamoorthy, G.; Dogra, S. K. J. Photochem. Photobiol. A 1999, 121, 191.

JP8003344