14402
J. Phys. Chem. B 2008, 112, 14402–14408
Porphyrin J-Aggregates Stabilized by Ferric Myoglobin in Neutral Aqueous Solution† Koji Kano,* Kenji Watanabe, and Yoshiyuki Ishida Department of Molecular Chemistry and Biochemistry, Doshisha UniVersity, Kyotanabe, Kyoto 610-0321, Japan ReceiVed: March 25, 2008; ReVised Manuscript ReceiVed: May 20, 2008
J-Aggregates of a diacid form (H4TPPS2-) of 5,10,15,20-tetrakis(4-sulfonatophenyl)porphyrin (H2TPPS4-) were stabilized by binding with ferric myoglobin (metMb) in aqueous solution at neutral pH. The J-aggregates gradually dissociated to monomeric H2TPPS4-. The average half-lifetime (t1/2) of the J-aggregates in the presence of sufficient amounts of metMb was ca. 3 h in phosphate buffer at pH 7.0 and 25 °C. The stabilization of the J-aggregate by metMb is ascribed to encapsulation and fixation of an edge-to-edge structure of the J-aggregate by the relatively rigid protein molecules. The secondary structure of metMb was altered in some extent in the presence of an excess amount of the J-aggregates while no effect on denaturation of metMb was observed with the H2TPPS4- monomer or polyacrylate. The hydrophobic nature of the J-aggregate seems to play an important role for denaturation of metMb. The ability of denatured metMb to bind the azide anion was higher than that of natural metMb. The denaturation of metMb by the J-aggregates seems to induce surfacing of hemin leading to an entropy gain in coordination of the N3- anion to the iron(III) center. 1. Introduction Coulomb interaction is a very convenient force to promote formation of molecular association complexes.1 The Coulomb force (F) between two charges, q1 and q2, is represented by eq 1:
F ) (q1q2) ⁄ (4πε0εrr2)
(1)
where ε0 and εr are the electric constant of vacuum and the dielectric constant of the medium, respectively, and r is the distance between the charges. Equation 1 indicates that the interaction between the charges is relatively insensitive to the distance r and that the Coulomb interaction is weakened in a polar solvent such as water. Even though the Coulomb force is weak when a univalent cation attractively interacts with a univalent anion, the sum of the Coulomb forces (ΣFki) would become strong if a host having several charged points (qk, the number of the charged points ) n) interacts with a guest having oppositely charged points (qi, the number of the charged points ) m) as predicted by eq 2:
∑
n
f Fki )
m
∑∑
k)1 i)1
(
qkqi 4πε0εrrki2
)( ) rfki rki
(2)
where rki is the distance between the charged point k of the host and the oppositely charge point i of the guest. Such multipoint Coulomb interactions are utilized in nature as well as artificial systems for realizing strong complexation of protein-protein pairs and protein-artificial effector pairs. For example, cytochrome c (cyt c), a cationic electron-transport protein, forms a complex with cytochrome bc1 (cyt c reductase) through Coulomb interactions, the Michaelis constant for this system being reported to be 10-5 M1.2 Similar interaction has been known in complexation of cyt c with cyt c peroxidase.3–5 Meanwhile, cationic proteins have been known to complex with † Part of the “Janos H. Fendler Memorial Issue”. * To whom correspondence should be addressed. E-mail: kkanomail.doshisha.ac.jp.
various polyanionic effectors, which alter the functions of the proteins, such as a calixarene having four cyclic peptide loops,6–8 modified porphyrins,9–15 a fullerene,16 a ruthenium trisbipyridine complex,17 peptides,18,19 dendrimers,20,21 carboxylate-embedding gold nanoparticles,22–28 and an O-methylated β-cyclodextrin.29 These polyanionic effectors commonly have complicated structures. More simple polyvalent porphyrins can also be utilized as the effectors of the proteins. It has been reported that 5,10,15,20-tetrakis(4-carboxyphenyl)porphyrin (H2TCPP4-) binds to cyt c with a dissociation constant (Kd) of 5 × 10-6 M.30 Hematoporphyrin31 and protoporphyrin IX32 interact with myoglobin (Mb) and hemoglobin (Hb) resulting in the reduction in the abilities of these hemoproteins to bind dioxygen. We found a fairly strong complexation of 5,10,15,20-tetrakis(4-sulfonatophenyl)porphinatoiron(III) (FeIIITPPS3-) with R-chymotrypsin (ChT), the binding constant (Kb) being 4.8 × 105 M-1 in 5 × 10-3 M phosphate buffer at pH 7.4.33 FeIIITPPS3- definitely inhibits the catalysis of ChT in hydrolysis of a model substrate (N-succinylL-phenylalanine p-nitroanilide) by covering the active site of ChT. FeIIITPPS3- is a three-valent anionic porphyrin. Sevenvalent 5,10,15,20-tetrakis[4-(3,5-dicarboxyphenylmethoxy)phenyl]porphinatoiron(III) forms a very stable 1:2 porphyrin-ChT complex (K1 ) 2 × 108 M-1, K2 ) 1 × 106 M-1).33 Increase in the charge number of an effector generally yields a more stable association complex. However, it might be difficult to prepare a multivalent effector whose charged points are located at a narrow area that corresponds to a protein surface area. We noticed a J-aggregate of the diacid (H4TPPS2-) of 5,10,15,20tetrakis(4-sulfonatophenyl)porphyrin (H2TPPS4-). The pKa value of H4TPPS2- is 4.8 and two protons almost simultaneously bind to the two pyrrole nitrogens of H2TPPS4- in acidic medium to form H4TPPS2-. The positive charges at the center of H4TPPS2intermolecularly interact with the SO3- groups of other H4TPPS2- molecules to form an edge-to-edge J-aggregate that is a polymer whose monomer units are bound to each other by Coulomb interactions (Figure 1).34–39 Since each porphyrin unit in the J-aggregate still contains two negative charges, the
10.1021/jp802567b CCC: $40.75 2008 American Chemical Society Published on Web 07/18/2008
Porphyrin J-Aggregates Stabilized by Ferric Myoglobin
J. Phys. Chem. B, Vol. 112, No. 46, 2008 14403
Figure 1. Energy-minimized structure of a part of the J-aggregate of H4TPPS2- obtained from MM2 calculation using BioMedCAChe 6.0. The effects of the solvent were not considered in the calculation.
J-aggregate of H4TPPS2- is a polyvalent anionic species whose charge density must be high. Interactions of the J-aggregates of H4TPPS2- with proteins have been studied in aqueous acidic solutions. It has been reported that the J-aggregates are formed on the surfaces of the proteins such as bovine serum albumin (BSA), γ-globulin, and ChT in aqueous solution at pH 1.86.40 Human serum albumin (HSA) also promotes the formation of the J-aggregates in an aqueous solution at pH 2.0.41 The previous studies on the formation of the J-aggregates on the protein surfaces have been carried out in acidic solutions where the J-aggregates of H4TPPS2- are stable but the nature of the protein differs from the wild conditions. In the present study, we found that the J-aggregates formed in an acidic solution are stabilized by binding with ferric myoglobin (metMb) in aqueous solution at pH 7.0. The J-aggregates bound to metMb spontaneously dissociated with time into the monomeric H2TPPS4-. The results suggest a possibility of “self-consuming effector”. 2. Experimental Section Materials. H2TPPS · nH2O in an acidic form (>98%) was purchased from Tokyo Chem. Ind. Co. and used without purification. The concentration of H2TPPS4- was determined from the absorbance of the solution in 5 × 10-3 M phosphate buffer at pH 7.0 using the extinction coefficient at 414 nm of 5.33 × 105 M-1 cm-1.34 MetMb from horse heart (Sigma,>90%) was used. To determine the concentration of metMb, the aqueous solution of metMb was reduced by Na2S2O4 to yield ferrous Mb and air was bubbled into the ferrous Mb solution to obtain oxyMb whose concentration could be determined from the extinction coefficient at 410 nm of 1.28 × 105 M-1 cm-1.42 Polyacrylic acid (Aldrich, average mol. wt. ) 1800, mean degree of polymerization ) 25) was used as received. Other reagents and solvents (Wako Chem. Co.) were commercially available and used without purification. Water was purified using a Millipore Simpak 1. Instruments. UV-vis spectra were measured by a Shimadzu UV-2100 spectrophotometer. Circular dichroism (cd) spectra
were taken using a Jasco J820 spectropolarimeter. A quartz cell with a 1 mm optical length was used for measuring cd spectra. Microcalorimetric measurements were performed with a Microcal isothermal titration calorimeter VP-ITC and the data were analyzed by the ORIGIN software program. Sample Preparation. H2TPPS in the acidic form (9.3 mg) was dissolved into 5 mL of 5 × 10-3 M phosphate buffer at pH 7.0 and allowed to stand for a night. Since the buffer action of the system was not enough, the pH value of the resulting solution decreased to 2.8 ( 0.3 due to acid-dissociation of the SO3H groups of the porphyrin. The main species of the solution were the J-aggregates of H4TPPS2-, and we used this solution as the stock solution of the J-aggregates of H4TPPS2-. The concentration of the stock solution was 2.0 × 10-3 M as the H2TPPS4- monomer. 3. Results and Discussion Complexation and Stabilization of J-Aggregates in Neutral Ferric Mb Solution. The solution of 1 × 10-5 M H2TPPS4- in phosphate buffer (5 × 10-3 M) at pH 7.0, which was prepared from the 2 × 10-3 M H4TPPS2- stock solution at pH 2.8 where H4TPPS2- predominantly exists as the J-aggregates, showed the absorption maxima at 413, 515, 551, 579, and 633 nm due to the monomer of the porphyrin. This means that the J-aggregates of H4TPPS2- in an acidic solution easily dissociate to the monomeric free base (H2TPPS4-) when they undergo a pH jump. Meanwhile, a 200× dilution of the acidic porphyrin stock solution (2 × 10-3 M, 15 µL) by injecting it into the 1 × 10-5 M metMb solution in phosphate buffer at pH 7.0 (3 mL) led to a quite different UV-vis spectrum as shown in Figure 2. The absorption maxima were observed at 491 and 705 nm that are ascribed to the J-aggregates of H4TPPS2-.34–39 Judging from the absorbance at 413 nm due to the coexisting monomer (H2TPPS4-) in the differential spectrum (cancelation of the contribution of metMb at 413 nm), at least 80% of the porphyrin molecules exist in the J-aggregate form immediately after the sample was prepared. These results definitely indicate that the J-aggregates of H4TPPS2- are stabilized by metMb even in a neutral solution. The final concentration of the porphyrin (1 ×
14404 J. Phys. Chem. B, Vol. 112, No. 46, 2008
Figure 2. Progressive absorption spectral changes of the sample that was prepared by injecting 15 µL of the stock solution of H4TPPS2- (2 × 10-3 M) at pH 2.8 into 3 mL of 1 × 10-5 M metMb in 5 × 10-3 M phosphate buffer at pH 7.0 and 25 °C. The inset is the time course of the absorbance at 491 nm due to the J-aggregates.
Figure 3. Absorption spectra of the J-aggregates of H4TPPS2- in 5 × 10-3 M phosphate buffer at pH 7.0 containing 1 × 10-5 M metMb: [H4TPPS2-]0 ) 1 × 10-5 (black), 1 × 10-6 (red), and 1 × 10-7 M (blue). A cell with a 10-mm optical length was used.
10-5 M) was set at a relatively high value because of the requirement for the following experiments. We confirmed that the J-aggregates of H4TPPS2- were stabilized by metMb in aqueous solution at pH 7.0 even when the initial concentration of H4TPPS2- was reduced to 1 × 10-7 M (Figure 3). All samples whose initial concentrations of H4TPPS2- were 10-5∼10-7 M showed almost the same apparent extinction coefficients ((absorbance at 491 nm)/[H4TPPS2-]0), indicating that the Jaggregate/metMb complex is very stable and an apparent Kb value should be larger than 107 M-1 as an independent bindingsite mode. The J-aggregates in the metMb solution dissociated with time to the monomer (H2TPPS4-) as shown in Figure 2. Figure 4 shows the effect of the concentration of metMb on the stabilization of the J-aggregates. At first, the half-lifetime of the J-aggregate (t1/2) linearly increases with increasing [metMb] and then saturated at [metMb] ) ca. 2 × 10-5 M where [H4TPPS2-]/[metMb] ) ∼0.5 and t1/2 ) ∼3 h. Such a clear saturation suggests the strong complexation of the J-aggregates with metMb. Figure 5 shows a quite hypothetical image of a J-aggregate/metMb complex where the positively charged residues of metMb interact with the negatively charged SO3-
Kano et al.
Figure 4. Half-lifetimes (t1/2) for dissociation of the J-aggregates of H4TPPS2- (1.0 × 10-5 M) in the presence of various amounts of metMb in phosphate buffer (5.0 × 10-3 M) at pH 7.0 and 25 °C.
groups of the J-aggregate. If the interaction between metMb and the J-aggregates is very strong and only Coulomb interaction participates in complexation as shown in Figure 5, the mole ratio, [H4TPPS2-]/[metMb], is expected to be 1.5∼2.5, which depends on the direction of metMb to the rod of the J-aggregate, at the saturation point. However, the experiment shows the molar ratio of ∼0.5 at the saturation point. These findings might be interpreted in terms of more complicated structures of the J-aggregate/metMb complex than that shown in Figure 5. On the basis of the results of small-angle X-ray scattering, a unique hollow cylindrical structure was considered for the J-aggregate of H4TPPS2-.43 Meanwhile, a rodlike J-aggregate of H4TPPS2with ca. 104 molecules along its length and ca. 20 molecules across its diameter was assumed.44 These proposed structures, however, cannot reasonably explain the present results showing that two metMb molecules interact with one H4TPPS2- molecule. Hydrophobic and/or van der Waals interactions between metMb and the porphyrin planes seem to participate in more complicated complexation. Further studies need to clarify the structure of the metMb/J-aggregate complex. Since the isoelectric point of metMb is ∼7, the total net charge of metMb in aqueous solution at pH 7.0 is almost zero. Therefore, it might be assumed that Coulomb interactions between the J-aggregates and metMb are significantly weak. To know the participation of Coulomb interactions, the effect of inorganic salt on the stabilization of the J-aggregates by metMb was examined by measuring the rates of dissociation of the J-aggregates (Figure 6). In 5 × 10-3 M phosphate buffer in the absence of NaCl, the t1/2 was 2 h. However, the t1/2 values became 20 and
