Langmuir 2005, 21, 10355-10361
10355
Spectroscopic Investigations of Metalloporphyrin-Oligopeptide Systems: Evidence for Peptide Aggregation M. Aoudia*,† and M. A. J. Rodgers‡ Department of Chemistry, College of Science, Sultan Qaboos University, P.O. Box 36, Al-Khodh, Muscat, Sultanate of Oman, and Center for Photochemical Science, Department of Chemistry, Bowling Green State University, Bowling Green, Ohio 43402 Received April 23, 2005. In Final Form: August 28, 2005 Anionic pentapeptides consisting of a string of four glutamic acid residues terminated by either tyrosine (Glu4Tyr) or tryptophan (Glu4Trp) were synthesized, and their aggregation properties in buffered (pH ) 7.0) aqueous solutions were investigated using two different approaches. In the first approach, the effects of the concentration of peptide used as its own probe (intrinsic probe) on its fluorescence emission, circular dichroism, surface tension, and solution pH yielded similar critical peptide concentrations of around 175 µM. This particular concentration was taken as evidence for peptide aggregation. In the second approach, peptide aggregation was investigated using cationic metalloporphyrins, tetrakis(N-methyl-4-pyridyl)porphyrin (PdIITMPyP4+ and ZnIITMPyP4+), as extrinsic probes. The effect of peptide concentration on porphyrin ground-state absorption confirmed peptide aggregation, but at a lower critical peptide concentration near 125 µM. This difference was attributed to the possible distortion introduced by the association of one (or more) large metalloporphyrin molecule with the peptide aggregates. Evidence for peptide aggregation was also demonstrated from the effect of peptide concentration on PdIITMPyP4+ tripletstate decay. The fast component (kf, associated with electron transfer from the target Tyr and Trp residues to the porphyrin triplet state) was found to be independent of peptide concentration, implying no noticeable effect of peptide aggregation on the electron-transfer event. This was attributed to the fact that species formed by excitation of porphyrin associated with ion-pair complexes or bound to peptide aggregates and the diffusion together of the separate T1 and peptide entities in the bulk phase are kinetically similar. On the other hand, the slower component (ks) of the decay, which is associated with the diffuse formation of an encounter complex between the free peptide and T1 porphyrin (bulk phase), was peptide-dependent and displayed a critical peptide concentration near 125 µM, above which it became practically independent of peptide concentration. This invariance of ks was taken as an indication that the free peptide concentration in the bulk phase remains constant above 125 µM, the concentration at which peptide molecules prefer to associate as aggregates.
Introduction Electron transfer is of vital importance in many biological processes, and in nature, it is mediated by proteins. Thus, considerable research has been carried out in an effort to understand electron-transfer reactions in proteins.1-3 However, this work has shown the electron transfer to be complex to analyze because of the presence of numerous competitive pathways, and interest has shifted to the use of model systems to simplify the kinetic.4-9 We recently reported10,11 on the photophysical processes in self-assembled ion-pair complexes between cationic * To whom correspondence should be addressed. E-mail: aoudia@ squ.edu.om. Fax: 968-24413415. † Sultan Qaboos University. ‡ Bowling Green State University. (1) Nagamine, K.; Torikai, E. J. Phys.: Condens. Matter. 2004, 16 (40), 4797-4806. (2) Di Donato, M.; Peluso, A.; Villani, G. J. Phys. Chem. B. 2004, 108 (9), 3068-3077. (3) Di Donato, M.; Peluso, A. Theor. Chem. Acc. 2004, 111 (2-6), 303-310. (4) Shin, Y. K.; Newton, M. D.; Isied, S. S. J. Am. Chem. Soc. 2003, 125 (13), 3722-3732. (5) Serron, S. A.; Aldridge, W. S.; Fleming, C. N.; Danell, R. M.; Baik, M.; Sykora, M.; Dattelbaum, D. M.; Meyer, T. J. J. Am. Chem. Soc. 2004, 126 (44), 14506-14514. (6) Malk, R. A.; Gao, Z.; Wishart, J. F.; Isied, S. S. J. Am. Chem. Soc. 2004, 126 (43), 13888-13889. (7) Marme, N.; Knemeyer, J. P.; Wolfrum, J.; Sauer, M. Angew. Chem., Int. Ed. 2004, 43 (29), 3798-3801.
metalloporphyrins, tetrakis(N-methyl-4-pyridyl)porphyrin (PdIITMPyP4+ and ZnIITMPyP4+), and anionic pentapeptides consisting of a string of four glutamic acid (Glu) residues terminated by either tyrosine (Glu4Tyr) or tryptophan (Glu4Trp) moieties. The deprotonated carboxylic acid residues of the four Glus were postulated to act as an ion-pairing site for a tetracationic porphyrin, thus bringing the target (Tyr and Trp) moiety in proximity to the porphyrin and setting up the possibility of an intracomplex electron-transfer event after photoexcitation of the porphyrin into its triplet (T1) excited state. Indeed, evidence for ion association between porphyrin and peptide ground states was provided by the effect of peptide concentration on the porphyrin ground-state absorption. In addition, our results10 clearly indicated a photoinduced electron transfer from the ground state of the target aromatic amino acid (Tyr or Trp) to the metalloporphyrin triplet state and elucidated the role played by the thermodynamic driving force (∆G°), the electronic coupling energy between the donor and acceptor (HAD), the solvent reorganization energy (λ), and the local dynamics in the metalloporphyrin-peptide complex. In particular, the (8) Liu, L.; Hong, J.; Ogawa, M. Y. J. Am. Chem. Soc. 2004, 126 (1), 50-51. (9) Lasey, R. C.; Liu, L.; Zang, L.; Ogawa, M. Biochemistry 2003, 42 (13), 3904-3910. (10) Aoudia, M.; Rodgers, M. A. J. J. Am. Chem. Soc. 1997, 118 (52), 12859. (11) Aoudia, M.; Guliaev, A. B.; Leontis, N. B.; Rodgers, M. A. J. Biophys. Chem. 2000, 83, 121-140.
10.1021/la051085q CCC: $30.25 © 2005 American Chemical Society Published on Web 09/30/2005
10356
Langmuir, Vol. 21, No. 23, 2005
Aoudia and Rodgers
quenching of the metalloporphyrin triplet state was described by the sum of two exponential terms: A(t) ) a1 exp(-kst) + a2 exp(-kft). The fast decay component kf [(6.6 ( 0.2) × 106 s-1 for the Tyr variant and (1.4 ( 0.15) × 107 s-1 for the Trp analog) was found to be independent of peptide concentration. The slow decay component ks showed a linear increase with peptide concentration, with a bimolecular rate constant of 6.2 × 109 M-1 s-1 for both peptides (Glu4Tyr and Glu4Trp). The fast contribution to the decay was associated with intracomplex electron transfer, whereas the slow contribution was attributed to diffusive formation of an encounter complex between free peptide and porphyrin (T1) molecules in the bulk phase, followed by electron transfer.10 The above investigation was conducted at relatively low peptide concentrations (up to 120 µM). However, numerous studies have provided evidence for peptide aggregation in solution as the concentration of peptide is increased. Thus, the ion mobility/time-of-flight technique has been used12 to examine the onset of aggregation in model dipeptide systems of Gly-X (where X ) Ala, Asn, Asp, Glu, His, Leu, Ser, Thr, and Trp). Under the experimental conditions used, it was found that simple binary and quaternary mixtures of these dipeptides produce clusters containing as many as 16-75 peptide units (and 1-7 charges). The aggregation of N-methylacetamide by interpeptide hydrogen bonding was investigated by polarized Raman and FT-IR spectroscopies.13 In water, effective aggregation of N-methylacetamide was found to require quite high concentrations of more than 6 M. On the other hand, N-methylacetamide in acetonitrile starts to form oligomers at concentrations of 0.1 M. Spectroscopically, three different species were inferred for both solvents investigated, namely, monomers (C1) and two different types of peptide oligomers (C2 and C3) that differ in terms of chain length and stability. Diffusion coefficients determined with pulsed-field gradient NMR spectroscopy were used to evaluate the aggregation of a family of model peptides based on a β-amyloid β (12-28) peptide fragment.14 Hydrophobic interactions were shown to play a major role in the stabilization of peptide aggregates. In a study reported by Nagaraj and Balaram,15 fluorescent aminoterminal fragments of emerimicin were synthesized and used as models for the study of peptide aggregation and interaction with lipids and proteins. The nonapeptide esters aggregated at concentrations higher than 8 µM, whereas tri- and pentapeptide esters did not. The peptides bound to lipid dispersions, with the largest changes in fluorescence observed for the nonapeptide. Optical rotatory dispersion (ORD) and circular dichroism (CD) were used to investigate the interactions between the polyacids poly (R-L-glutamic acid) and poly(R-L-aspartic acid) and the polybases poly-l-lysine and poly-l-ornithine in a variety of solvents.16 The data suggest that extreme conformation changes can occur upon aggregation of polypeptides. The nature of the aggregate is a sensitive function of the polypeptide side chains as well as of solvent composition. Insight into the environment of tryptophan in a hydrophobic model peptide upon aggregation and interaction with lipid vesicles was obtained by steady-state and time-
resolved fluorescence.17 This study was carried out on a model hydrophobic peptide and its fatty acylated derivative. Steady-state fluorescence measurements suggest that the fatty acyl chain attached to an amino acid associates with the peptide chain in aqueous environments. Aggregation kinetics for a tetrapeptide analogue of cholecystokinin (A-71623) has been studied by quasi-elastic light scattering18 and was treated by a kinetic model. This model predicted an increase of the average molecular weight of peptide aggregates with time. Evidence for peptide aggregation was also provided by Kastin et al.19 Steady-state spectroscopy (ground-state absorption and singlet-excited-state emission) is extensively used to monitor the aggregation of peptides in aqueous solutions. This is because the absorption and fluorescence characteristics of optically active probes are very sensitive to the local environment. In general, there are two approaches for investigating aggregation formation by the means of light-absorbing and -emitting probes. The first approach is based on the use of an extrinsic probe and consists of the utilization of aromatic hydrocarbons or their derivatives as the optically active probes. One drawback of this approach is that the association of one (or more) relatively large aromatic molecules with the aggregate species surely leads to distortions of the normal aggregate structure, which, in turn, can affect the size and shape of the aggregates. The second approach is based on the use of an intrinsic probe and consists of the utilization of a molecule having an aromatic chromophore as a part of its structure; this introduces no distortions into the peptide aggregate. In most previous studies, extrinsic probes have been used, and only few studies have reported on the use of intrinsic probes, whereas investigations using both intrinsic and extrinsic probes for a similar system are rather rare.20,21 Thus, in the present study there has been an attempt to use both approaches (intrinsic and extrinsic) to monitor peptide aggregation in porphyrin-oligopeptide systems. One set of experiments was carried out using the amino acid residues (Tyr and Trp) as the intrinsic probes to examine the effect of peptide concentration on the aromatic amino acid fluorescence and circular dichroism spectra. In addition, the pentapeptide aggregation was investigated using surface tension and pH techniques. In another set of experiments, PdIITMPyP4+ and ZnIITMPyP4+ were used as the extrinsic probes. This possibility arises from the observed formation of ion-pair complexes between porphyrin and peptide molecules.10 As a result, the effect on the ground-state absorption of PdIITMPyP4+ and ZnIITMPyP4+ of increasing the peptide concentration was examined to assess a plausible peptide aggregation. Furthermore, an additional factor of interest was that the peptide concentration surely perturbs the porphyrin triplet state, as reflected by the dramatic change of the porphyrin triplet-state kinetics from a singleexponential T1 decay with a lifetime of 125 µs without peptide into a biexponential decay that removes all T1 from the system within 10 µs in the presence of peptide.10 Therefore, another aim of this study was to investigate peptide aggregation via the effect of peptide concentration
(12) Countermane, A. E.; Hilderbrand, A. E.; Srebalus Barnes, C. E.; Clemmer, D. E. J. Am. Soc. Mass Spectrom. 2001, 12 (9), 1020-1035. (13) Schweitzer-Stenner, R.; Sieler, G.; Christiansen, H. Asian J. Chem. 1998, 7 (2), 287-312. (14) Mansfield, S. L.; Jayawickrama, D. A.; Timmons, J. S.; Larive, C. K. Biochim. Biophys. Acta 1998, 1382 (2), 257-265. (15) Nagaraj, R.; Balaram, P. Biochem. Biophys. Res. Commun. 1979, 89 (4), 1041-1049. (16) Hames, G. G.; Schullery, S. E. Biochemistry 1968, 7 (11), 38823887.
(17) Joseph, M.; Nagaraj, R. Indian J. Biochem. Biophys. 1998, 35 (2), 67-75. (18) Silvestri, S.; Lu, M. Y.; Johnson, H. J. Pharm. Sci. 1993, 82 (7), 689-93. (19) Kastin, A. J.; Castellanos, P. F.; Fischman, A. J.; Alan, J.; Proffitt, J. K.; Graj, M. V. Pharmacol., Biochem. Behav. 1984, 21 (6), 969-73. (20) Aoudia, M.; Rodgers, M. A. J.; Wade, W. H. J. Colloid Interface Sci. 1984, 145 (2), 493-501. (21) Aoudia, M.; Rodgers, M. A. J.; Wade, W. H. J. Phys. Chem. 1984, 88 (21), 5008-5012.
Spectroscopy of Metalloporphyrin-Oligopeptide Systems
Langmuir, Vol. 21, No. 23, 2005 10357
on the porphyrin triplet-state decay. Such a study was previously performed,10 but only at peptide concentrations up to 120 µM, where no evidence for peptide aggregation was noted. Experimental Section Material and Procedures. The solid-phase synthesis of the different peptides (Glu4Tyr and Glu4Trp) and the purification and characterization were carried out using procedures that have been fully described elsewhere.10 Potassium dihydrogen phosphate and sodium hydrogen phosphate in equal proportions were used to make up the aqueous buffer (10 mM). Fluorescence spectra were obtained with a Perkin-Elmer LS-5B luminescence spectrometer. The pH was measured with a calibrated Corning ion analyzer (model 250). Circular dichroism (CD) experiments were carried out with a Jasco spectropolarimeter. Absorption spectra were recorded on a GBC 918 UV-visible rapid-scan spectrophotometer. Surface tensions measurements were obtained with a spinning drop tensiometer (model 500). Flash photolysis experiments were carried out with the second harmonic (532 nm) of a Continuum Surelie I Q-switched Nd: YAG laser that provides 6-ns pulses. Transient absorbance was monitored at right angles to the laser excitation beam, using a computer-controlled kinetic spectrometer, which has been described elsewhere.22 Solutions were saturated with argon prior to each measurement.
Results and Discussion Intrinsic Probe Measurements. Peptide Fluorescence Spectra. Fluorescence emission from Glu4Tyr in buffered (pH ) 7.0) aqueous solutions was measured at different peptide concentrations in the range 28-310 µM. The fluorescence spectra were taken at λexc ) 230 nm. All spectra consist of a single, unstructured band around 307 nm.10 This is in excellent agreement with the results reported elsewhere,23 in which the fluorescence spectra of tyrosine, both in aqueous solutions and when bound to polypeptides and proteins, also show a single and unstructured band with a maximum around 305 nm, implying that the fluorescence maximum emission wavelength (λem,max) of the tyrosyl residue is insensitive to the local environment. On the other hand, the fluorescence emission maxima (If,max at 307 nm) of the tyrosine were peptide concentration-dependent. Thus, fluorescence maxima were extracted from the peptide fluorescence spectra and plotted versus Glu4Tyr concentration (Figure 1). Clearly, a transition concentration is observed, and the intersection of the two lines drawn between points at the extreme ends of the concentration range used yielded a critical peptide concentration around 175 µM. In accord with many reported studies,24-26 the quantum yield of the fluorescence of the tyrosyl residue is extremely sensitive to the nature and structure of its molecular environment; thus, the observed critical peptide concentration can be attributed to a change in the tyrosyl residue environment upon aggregation, compared to its environment in bulk solution as free monomeric Glu4Tyr species. In fact, other studies27-29 have also reported evidence for peptide (22) Rihter, B. D.; Kenney, M. E.; Ford, W. E.; Rodgers, M. A. J. J. Am. Chem. Soc. 1993, 115, 8146-8152. (23) Lakowicz, J. R. Topics in Fluorescence Spectroscopy; Plenum Press: New York, 1992; Volume 3, Biochemica. (24) Chen, R. F.; Edelhoch, H. Biochemical Fluorescence: Concepts; Marcel Dekker: New York, 1976. (25) Cogwill, R. W. Biochim. Biophys. Acta 1967, 133 (1), 6-18. (26) Cogwill, R. W. Biochim. Biophys. Acta 1970, 200 (1), 18-25. (27) Jayakumar, R.; Mandal, A. B.; Manoharan, P. T. J. Chem. Soc., Chem. Commun. 1993, 10, 853-855. (28) Jakayumar, R.; Jayanthy, C.; Gomathy, L. Int. J. Pept. Res. Ther. 1995, 45, 129-137. (29) Venkatesh, B.; Jayakumar, R.; Manoharan, P. T. Biochem. Biophys. Res. Commun. 1996, 223, 390-396.
Figure 1. Variation of the fluorescence intensity at the maximum emission wavelength (λmax ) 307 nm) with peptide concentration.
aggregation in aqueous solutions, and the peptide critical concentration was defined as the peptide critical micelle concentration (cmc). Although this concept of cmc is generally used to define the concentration at which surfactant molecules self-associate to form aggregates called micelles, the analogy between peptides and surfactants is somehow legitimate because peptides have an amphipathic structure that consists of two antagonist groups, a hydrophobic group (water-insoluble) and a hydrophilic group (water-soluble). This particular structure might explain the possibility of peptide aggregation with increasing peptide concentration, as it is generally observed with surfactants. Peptide Circular Dichroism Spectra. The evidence for peptide aggregation in aqueous media with increasing system concentration was also investigated by circular dichroism, using the peptide as its own probe. Circular dichroism (CD) curves were measured at different Glu4Tyr concentrations in the range 20-300 µM in buffered aqueous solutions (pH ) 7.0). All spectra showed a single cotton effect with a maximum at λmax ) 225 nm. Maximum signals were extracted from CD curves and plotted versus the reciprocal of peptide concentration (Figure 2). As clearly seen from this figure, a transition over a narrow range of peptide concentration was observed. Below and above that transition, the maximum CD signal varied linearly with peptide concentration. Extrapolations of the two lines intersect at 178 µM, the concentration at which a change of tyrosine moiety microenvironment is believed to occur upon peptide aggregation. Circular dichroism (CD) is concerned with the change in the index of refraction, as well as with the differential absorption by the chromophore (tyrosine residue) of one of the two circulary polarized waves. It is therefore an extremely useful technique for monitoring the formation of organized aggregates of hydrophobic model peptides containing aromatic amino acid residues such as tyrosine, tryptophan, and phenylalanine.30-32 Interestingly, the particular peptide concentration observed from the CD absorption spectra is in good agreement with that determined from fluorescence emission characteristics (Figure 1). (30) Broughman, J. R.; Shank, L. P.; Takegushi, W.; Schultz, B. D.; Iwamoto, T.; Mitchell, K. E.; Tomich, J. M. Biochemistry 2002, 41 (23), 7350-7358. (31) Li, L. K.; Spector, A. Exp. Eye. Res. 1974, 19 (1), 49-57. (32) Raj, P. A.; Balaram, P. Biopolymers 1985, 24 (7), 1131-1146.
10358
Langmuir, Vol. 21, No. 23, 2005
Aoudia and Rodgers
Figure 4. Variation of the surface tension with Glu4Tyr concentration, pH ) 7.0. Figure 2. Variation of the single cotton effect at λmax ) 225 nm with Glu4Tyr concentration, pH ) 7.0.
Figure 3. Variation of the solution pH with Glu4Tyr concentration.
Effect of Glu4Tyr Concentration on Its Solution pH. The pH of Glu4Tyr aqueous solutions was measured at different peptide concentrations in the range 25-500 µM. A plot of pH versus Glu4Tyr concentration is shown in Figure 3, in which the pH of the aqueous peptide solution first decreases rapidly with Glu4Tyr concentration up to a particular peptide concentration near 172 µM and then continues to decrease gradually as the peptide concentration is increased. Similar pH-peptide concentration behavior was observed for a pentapeptide containing tyrosine at the end of its backbone,29 and the authors reported a critical peptide concentration value of 90 µM. Thus, the breakpoint in the pH-peptide concentration curve can be attributed to peptide aggregation. Again, the critical Glu4Tyr concentration (172 µM) is in good agreement with the values derived from fluorescence emission (175 µM) and circular diachroism (178 µM). Surface Tension of Glu4Tyr Aqueous Solutions. The variation of the surface tension of a buffered (pH ) 7.0) Glu4Tyr aqueous solution was investigated at different peptide concentrations in the range 25-500 µM. A plot of surface tension versus Glu4Tyr concentration is shown in Figure 4, where a critical peptide concentration is observed near 172 µM. The variation of the surface tension with concentration is generally used to measure the critical
micelle concentration (cmc) of surfactants, the concentration at which the surfactant molecules start to form micelles.33 Thus, the observed break in the variation of the surface tension with peptide concentration around 172 µM also reflects a self-association of peptide molecules and the formation of aggregates. Furthermore, it is relevant to note the interesting similarity between the peptide critical concentration values (174-178 µM) derived from the different and independent experimental techniques used in our investigation. This consistency between fluorescence emission, circular diachroism, solution pH, and surface tension measurements appears to provide strong evidence for peptide aggregation around 174 µM. Extrinsic Probe Measurements. Effect of Peptide Concentration on PdIITMPyP4+ and ZnIITMPyP4+ GroundState Absorption. Evidence for peptide aggregation was also obtained using metalloporphyrins (PdIITMPyP4+ and ZnIITMPyP4+) as extrinsic probes. Thus, the effect of the peptide (Glu4Tyr) concentration on PdIITMPyP4+ and ZnIITMPyP4+ ground-state absorption was investigated. In accord with our results reported elsewhere,10 the addition of peptide caused the Soret band to undergo a red shift, with a lessening of the absorption intensity at the band maximum. Thus, maximum absorbencies were extracted from the ground-state absorption spectra of PdIITMPyP4+ and ZnIITMPyP4+ at different peptide concentrations and were plotted versus peptide concentration (Figures 5 and 6). Both figures showed two linear regimes. The breakpoint between the two occurred at a critical peptide concentration near 125 µM. In metalloporphyrin-peptide systems at low peptide concentration (up to 125 µM), the presence of two different species of PdIITMPyP4+ has been unequivocally demonstrated: free metalloporphyrin monomers and metalloporphyrin-peptide ion pair complexes formed by Coulombic interactions between the tetracationic palladium(II) metalloporphyrin and the deprotonated carboxylic acid residues for the four Glus.10 The observed break near 125 µM (Figures 5 and 6) therefore reflects the appearance of a third kind of metalloporphyrin species. In line with the above results (intrinsic probe approach), these new species are probably metalloporphyrin molecules bound to aggregates. These aggregates form at a critical peptide concentration near 125 µM, somewhat lower than the corresponding values deter(33) Rosen, M. J. Surfactants and Interfacial Phenomena, 2nd ed.; John Wiley and Sons: New York, 1989.
Spectroscopy of Metalloporphyrin-Oligopeptide Systems
Figure 5. Variation of the absorption intensity at the Soret band maximum for PdIITMPyP4+ (12 µM) in buffered aqueous solutions (pH ) 7.0) with Glu4Tyr concentration.
Figure 6. Variation of the absorption intensity at the Soret band maximum for ZnIITMPyP4+ (5 µM) in buffered aqueous solutions (pH ) 7.0) with Glu4Tyr concentration.
mined from measurements based on the use of the peptide as its own probe (∼175 µM). Also, it is worth noting that the critical peptide concentrations derived from PdIITMPyP4+ and ZnIITMPyP4+ are in good agreement with each other, probably because of the similar formal charges of the metalloporphyrin peripherical groups in both porphyrins, which might result in similar types of porphyrin-peptide interactions. At this juncture, it is appropriate to point out the difference between the critical peptide concentrations derived from measurements using the peptide as the intrinsic probe (∼175 µM) and those derived using the porphyrin molecules (PdIITMPyP4+ and ZnIITMPyP4+) as the extrinsic probes (∼125 µM). The first approach consists of the utilization of the peptide molecule having an aromatic chromophore (tyrosyl residue) as a part of its structure to monitor the changes of fluorescence and circular dichroism with peptide concentration. Clearly, this introduces no distortion into the peptide aggregates. This same situation also prevails in surface tension and pH measurements where no external perturbation is introduced in the system. All of the above experiments yielded similar peptide concentrations of ∼175 µM. On
Langmuir, Vol. 21, No. 23, 2005 10359
the other hand, in the second approach, porphyrin molecules (PdIITMPyP4+ and ZnIITMPyP4+) were used as the light-absorbing probes. An extrinsic probe having a considerable molecular size, such as the porphyrin molecules used in this study, is likely to cause significant perturbations in the region being explored. Further complications would arise if a solvent reorientation process and peptide-probe interactions were involved. In fact, and as discussed above, peptide-porphyrin interactions between the components of the systems investigated in this study are quite significant, as reflected by the strong effect of the Glu4Tyr concentration on the PdIITMPyP4+ and ZnIITMPyP4+ ground-state absorption spectra. As a result, the average size (and shape) of the aggregates might be influenced by the incorporation of the extrinsic probe molecule, which might explain the lower critical peptide concentration (∼125 µM) obtained with the extensive probe approach. In most previous work, either intrinsic or extrinsic probes have been used. Only very few studies20,21 have reported on the use of both intrinsic and extrinsic probes for the same system. Interestingly, our results show that the two approaches indeed lead to different values for the critical peptide concentration at which aggregates start to form, thereby confirming the general anticipated hypothesis that aggregate-extrinsic probe interactions can lead to distortion of the normal peptide aggregates. Indeed, this situation is clearly occurring in the metalloporphyrin-peptide systems investigated in this study, as shown by the observed difference between the critical peptide concentration determined from intrinsic probe measurements (∼175 µM) and that derived from extrinsic probe measurements (∼125 µM). To the best of our knowledge, this is the first instance where such a difference has reported for a single system using both intrinsic and extrinsic approaches. Effect of Peptide Concentration on PdIITMPyP4+ Triplet-State Kinetics. As previously mentioned, the porphyrin triplet-state decay is seriously perturbed by peptide concentration,10 because of the strong interaction between the porphyrin triplet and oligopeptide ground states. However, this effect was previously investigated only with dilute peptide buffered aqueous solutions (up to 120 µM). We therefore extended this study to porphyrin-oligopeptide systems at higher peptide concentrations. Our hypothesis is that the observed peptide (S0)porphyrin (T1) interaction at low peptide concentration10 might provide an appropriate means to monitor eventual peptide aggregation at higher concentrations. Thus, argonsaturated aqueous solutions of PdIITMPyP4+ in the absence and in the presence of peptide at pH ) 7.0 were irradiated with 6-ns pulses of 532-nm light. In the absence of peptide, the decay showed in-pulse formation of a transient absorption with λmax ) 460 nm, the decay of which was strictly monoexponential with a decay rate of ∼7.0 × 103 s-1, in good agreement with our previously reported value of 8.0 × 10-3 s-1.10 In the presence of peptide (Glu4Tyr and Glu4Trp), the kinetic profile changed significantly, and the decay was best fitted by the sum of two exponential terms, also in accord with our study reported elsewhere.10 PdIITMPyP4+ triplet-state (T1) decay was measured at different aqueous buffered (pH ) 7.0) Glu4Tyr and Glu4Trp concentrations in the range 20-800 µM. The fast decay component (kf) and the slow decay component (ks) were extracted and plotted against peptide concentration (Figures 7 and 8). As shown in Figure 7, the fast triplet decay component has a peptide-concentrationindependent rate constant (6.4 × 106 s-1 for Glu4Tyr and 1.3 × 107 s-1 for Glu4Trp) that compares well with the results derived in the low peptide concentration range
10360
Langmuir, Vol. 21, No. 23, 2005
Figure 7. Variation of the fast component kf of PdIITMPyP4+ (12 µM) triplet-state decay in argon-saturated buffered solutions (pH ) 7.0) in the presence of Glu4Tyr (O) and Glu4Trp (b).
Figure 8. Variation of the slow component ks of PdIITMPyP4+ (12 µM) triplet-state decay in argon-saturated buffered solutions (pH ) 7.0) in the presence of Glu4Tyr (O) and Glu4Trp (b).
(up to 120 µM), namely, 6.2 × 106 s-1 for Glu4Tyr and 1.4 × 107 s-1 for Glu4Trp.10 On the other hand, the slow decay component ks is peptide-concentration-dependent (Figure 8). A plot of ks versus peptide concentration displays two linear regimes. The linear regime at low peptide concentration exhibits a first-order increase in ks with peptide concentration, with a bimolecular rate constant of 6.4 × 109 M-1 s-1, close to that previously reported.10 The linear regime at higher peptide concentration is practically independent of peptide concentration. The intersection of the two regimes occurred at around 125 µM. The slower contribution to the T1 decay (ks) arises from the fraction of the equilibrium ground-state population that is not involved in complex formation (free metalloporphyrin monomers) prior to the excitation pulse but that undergoes diffusion formation of an encounter complex with free peptide molecules (bulk phase).10 The fact that ks is first-order in peptide concentration (up to ∼125 µM) and then becomes concentration-independent is a clear indication that the free peptide concentration in the bulk phase remains constant above 125 µM. This appears to suggest that any further addition of peptide molecules above this particular concentration will not affect the free peptide concentration in the bulk phase as a consequence of peptide aggregation. Thus, additional
Aoudia and Rodgers
evidence for peptide aggregation is obtained from the effect of the peptide concentration (Glu4Tyr and Glu4Trp) on the porphyrin triplet-state decay. Furthermore, the critical peptide concentration derived from this approach based on the use of the porphyrin triplet state as the extrinsic probe is in good agreement with the values derived from a similar approach based on the use of porphyrin (PdIITMPyP4+ and ZnIITMPyP4+) ground state as the extrinsic probe. Having shown that the invariance of ks with peptide concentration above ∼125 µM is probably related to the aggregation of peptide molecules and reflects a constant free-peptide concentration in the bulk phase, it is appropriate to turn to the understanding of the variation of PdIITMPyP4+ triplet-state fast decay component with peptide concentration. According to Figure 7, kf is not sensitive to the peptide aggregation, as it remains constant over the whole range of peptide concentration investigated (20-800 µM). kf is attributed to intracomplex electron transfer from the aromatic amino acid residue (Tyr and Trp) to the porphyrin triplet state.10 Electron transfer depends on several parameters,34 including (among others) the driving force of the reaction (∆G°), the solvent reorganization energy (λ), the electronic coupling energy between the electron donor and acceptor (HDA), and the relative donor-acceptor orientation. The reduction potential of Tyr and Trp is pH-dependent,35 and one can assume that ∆G° remains invariant at the fixed pH of 7.0 used in our investigation. In addition, the solvent reorganization energy is probably unchanged, as water is the solvent used in all measurements. Thus, among the abovecited factors influencing the electron-transfer process, only the donor-acceptor separation distance and mutual orientation (local dynamics) can be considered to have an effect on the rate constant kf. However, other observations10 can also be invoked to explain the invariance of kf with peptide concentration even at concentrations higher than the critical peptide concentration. Above this particular concentration, the conditions are such that the equilibrium contains significant mole fractions of free metalloporphyrin monomers (bulk phase), metalloporphyrin-peptide ion-pair complexes, and metalloporphyrin bound to peptide aggregates. Therefore, photoexcitation of the tetrapyrrole π system will produce three populations of porphyrin S1 and then T1 states: one that is localized within the ion-pair complexes; one that is bound to peptide aggregates and thus, in both cases, proximate to a tyrosine (or tryptophan) moiety of the oligopeptide; and one that has no such preset organization. Accordingly, electron transfer can occur via three channels. In the first two, the metalloporphyrin associated with ion-pair complexes or bound to peptide aggregates is quenched by electron transfer from tyrosine (or tryptophan) to the nearby metalloporphyrin T1 state within the ion-pair and aggregate assemblies. In the third, the free metalloporphyrin monomer first undergoes photoexcitation into the S1 state, followed by a rapid intersystem crossing to the triplet state T1. During its triplet lifetime, T1 and a free peptide molecule diffuse together in the bulk phase to form an encounter complex and then undergo electron transfer. Most likely, the T1 species formed by the photoexcitation of the ground-state metalloporphyrin complexed as an ion-pair or bound to the peptide aggregates and the diffusion together of the separate T1 (formed in the bulk phase) and peptide entities might not be of identical configuration, but at least they appear to be kineticallyz (34) Marcus, R. A.; Sutin, N. Biochim. Biophys. Acta 1985, 811, 265322. (35) Harriman, A. J. Phys. Chem. 1987, 91 (24), 6102-6104.
Spectroscopy of Metalloporphyrin-Oligopeptide Systems
similar, which explains the absence of any effect of peptide aggregation on the electron-transfer event from Tyr (or Trp) to the metalloporphyrin triplet state. Finally, without delving into the specific nature of the interactions between peptide molecules that results in the formation of aggregates, one can assume that hydrophobic interactions probably play a key role in the stabilization of these aggregates. Also, although the results from measurements using the peptide as the intrinsic probe (fluorescence, circular diachronic, surface tension, and pH) and porphyrin as the extrinsic probe (groundstate absorption and triplet-state decay) provided evidence for the formation of aggregates at a particular peptide concentration, the nature and structure of the aggregates cannot be established from these measurements. NMR studies might elucidate this issue. Conclusions Evidence for peptide (Glu4Tyr and Glu4Trp) aggregation was provided by the effect of peptide concentration (used at its own probe) on its fluorescence emission, circular dichroism, pH, and surface tension techniques. All measurements were in good agreement with each other, as shown by the observed similar critical peptide concentrations near 175 µM. In a second approach, peptide aggregation was investigated using cationic metalloporphyrins (PdIITMPyP4+ and ZnIITMPyP4+) as extrinsic probes. The effect of peptide concentration on the metalloporphyrin ground-state absorption also revealed peptide aggregation, but at lower critical peptide con-
Langmuir, Vol. 21, No. 23, 2005 10361
centration (∼125 µM). A similar critical peptide concentration was derived from the effect of peptide concentration on PdIITMPyP4+ triplet-state (T1) decay. The slower component (ks) of the triplet-state decay switched from a first-order dependence on peptide concentration below and up to 125 µM to a zeroth-order dependence for peptide concentrations above 125 µM. This particular concentration was taken as additional evidence for peptide aggregation. The faster component (kf) of T1 decay was insensitive to peptide aggregation in the whole concentration range investigated (20-800 µM). Such an observation was explained by the assumption that the porphyrin triplet-state species formed by photoexcitation of metalloporphyrin associated with ion-pair complexes or bound to peptide and those formed in the bulk phase by diffusion together of free T1 and peptide molecule are kinetically similar. Peptide critical concentrations derived from the intrinsic approach (175 µM) were higher than those derived from the extrinsic approach (125 µM). This difference was attributed to the possible perturbation introduced by the association of porphyrin molecules with peptide aggregates. Acknowledgment. Support for this work was provided by a grant from Sultan Qaboos University (IG/ SCI/ CHEM/03/03) and by the Center for Photochemical Sciences, Bowling Green State University (Bowling Green, OH). LA051085Q
