J-Aggregation of Anionic Ethyl meso-Thiacarbocyanine Dyes Induced

Publication Date (Web): August 2, 2007 ... New insights into the dimerization and site-specific cooperative interaction of Azure B with model transpor...
0 downloads 0 Views 180KB Size
Download PDF
J. Phys. Chem. B 2007, 111, 10023-10031

10023

J-Aggregation of Anionic Ethyl meso-Thiacarbocyanine Dyes Induced by Binding to Proteins Tatyana D. Slavnova,† Helmut Go1 rner,*,‡ and Alexander K. Chibisov† Center of Photochemistry, Russian Academy of Sciences, 119421 Moscow, Russia, and Max-Planck-Institut fu¨r Bioanorganische Chemie, D-45413 Mu¨lheim an der Ruhr, Germany ReceiVed: March 30, 2007; In Final Form: June 18, 2007

The effects of ribonuclease A (RNase), lysozyme, trypsin, and bovine serum albumin (BSA) on the J-aggregation behavior of 3,3′-bis[sulfopropyl]-5-methoxy-4′,5′-benzo-9-ethylthiacarbocyanine (1), 3,3′-bis[sulfopropyl]-4,5,4′,5′-dibenzo-9-ethylthiacarbocyanine (2), and 3,3′-bis[sulfopropyl]-5,5′-dimethoxy-9-ethylthiacarbocyanine (3) were studied in aqueous solution. The formation of J-aggregates at pH 6 is induced by RNase for 1-3, by lysozyme for 1 and 2, and by trypsin for 2. The formation of J-aggregates correlates with decay of the dimers and is supported by induced circular dichroism spectra. The concentration of J-aggregates for lysozyme/1 increases with an increase in the protein/dye concentration ratio, reaches a plateau, and then gradually decreases. J-aggregates are characterized by relatively weak fluorescence; e.g., Φf ) 0.01 for lysozyme/1, and by a small Stokes shift of 6-8 nm, indicating almost resonance fluorescence. J-aggregation proceeds in the range of seconds to minutes with sigmoidal type kinetic curves for trypsin/2 and nonsigmoidal kinetics in the other cases. The presence of BSA, in contrast to RNase, lysozyme, and trypsin, results in deaggregation of dimers of 1-3 and formation of bound monomers and exhibits intense fluorescence from the trans-monomer; e.g., Φf ) 0.22 for BSA/1. Generally, the binding of 1-3 to the proteins is a cooperative process, where the number of binding sites changes from n ) 15 for lysozyme/1 to n ) 6 for trypsin/2 and n ) 0.3 and 1 for BSA/3.

Introduction In recent years much attention has been focused on finding suitable conditions for the formation of nanostructured selfassociated assemblies which are expected to have important applications for the development of new technologies. Among the variety of nanosized systems, J-aggregates of cyanine dyes are the subject of keen interest and importance. These J-aggregates are formed as a result of the adsorption of dyes onto silver halide microcrystals,1 due to self-association in aqueous solution at high dye concentrations,2-4 and due to the presence of either metal ions3-10 or polymers.11-16 It was established that ion pairs between monomeric or/and dimeric cyanine dyes and mono-, di-, and trivalent metal ions serve as reactants.9,10 Another possibility is the use of gelatin as a template for inducing J-aggregation of cyanine dyes in aqueous solution.17,18 Recently, we have found that for 3,3′-bis[sulfopropyl]-5-methoxy-4′,5′dibenzo-9-ethylthiacarbocyanine (1) and 3,3′-bis[sulfopropyl]4,5,4′,5′-dibenzo-9-ethylthiacarbocyanine (2) the kinetics of J-aggregation in the presence of gelatin follow a sigmoidal curve and that for a sulfopropyl meso-oxacarbocyanine the rate of J-aggregation reaches the maximum value at pH 3.3-4.3.19 The process of J-aggregation can be started by injection of H+, yielding a pH value around the isoelectric point of gelatin, and then blocked by addition of OH- in appropriate concentration. The peculiarity of J-aggregation in the presence of gelatin is that the binding of the dye polypeptide molecules is a cooperative process. The Scatchard analysis of the binding allows evaluation of the number of binding sites and hence estimation of the number of dye molecules involved in J-aggregation.19 A few proteins, such as bovine serum albumin (BSA) or human † ‡

serum albumin (HSA), have been studied.20-25 For noncovalently linked complexes of HSA with a series of thiacarbocyanine dyes the number of binding sites (n) of 0.13-0.75 has been determined.22 Biotemplating, where polypeptides are widely used to produce functional nanostructures, is important for nanotechnology.26 One could expect that J-aggregation of cyanine dyes is enhanced by proteins, but surprisingly, to the best of our knowledge, no data are available in the literature. In this paper we extend our study of the formation of J-aggregates in aqueous solution to proteins, such as ribonuclease A (RNase), lysozyme, trypsin, and BSA. We aim to elucidate the conditions for three cyanine dyes: 1, 2, and 3,3′bis[sulfopropyl]-5,5′-dimethoxy-9-ethylthiacarbocyanine (3).27 J-aggregation of 1-3 occurs in the presence of RNase, whereas lysozyme induces formation of J-aggregates for 1, 2, and trypsin only for 2. The rate of J-aggregation is varied in the time range of minutes, depending on the properties of the protein-dye couple and influenced by concentration and pH. The binding of 1-3 to BSA results in deaggregation of the dimers and conversion of a cis-monomer into the trans-isomer. Generally, the binding of 1-3 to the proteins is a cooperative process, where n changes from 15 for lysozyme/1 to 0.3 for BSA/1, 3. One requirement of J-aggregation is a large n.

Russian Academy of Sciences. Max-Planck-Institut.

10.1021/jp072503y CCC: $37.00 © 2007 American Chemical Society Published on Web 08/02/2007

10024 J. Phys. Chem. B, Vol. 111, No. 33, 2007

Slavnova et al.

TABLE 1: Absorption Maximum and Extinction Coefficient of Monomer, Dimer, and J-Aggregatea dye λM (nm) λD (nm) D ( 104 M-1 cm-1) 1 2 3

572 579 560

530 536 523

7.9 8.1 10

λJb (nm) (646)c

639 648 (659)c 627

J (104 M-1 cm-1) 7.3 11 (13)d

a In water at pH 6. b [RNase]/[D] ) 0.1. c Using trypsin. d Using trypsin, pH 4.4.

Experimental Section The dyes were the same as those used previously.8-10,27 Lysozyme was from egg white (Fluka), and the other proteins were purchased from Sigma. The molecular weights of RNase, lysozyme, trypsin, and BSA are 13.6, 14.3, 24, and 66.5 kD, respectively. Water was from a Millipore (Milli Q) system. A diode array spectrophotometer (HP 8453) was employed to measure absorption spectra at discrete time intervals. The dye samples were freshly prepared by diluting (100-1000-fold) dimethyl sulfoxide stock solutions in water. The dye concentration was in the range of 2-10 µM. For kinetic measurements a few microliters of the appropriate protein concentration were injected into a 1 cm cell with the dye followed by intense stirring for 1-2 s prior to starting the measurement. Further stirring was made between each of two consecutive runs. The rate of J-aggregation was evaluated from the spectra.10 The steadystate absorption and fluorescence measurements were used to determine the concentration of free and bound dyes as well as to analyze the concentration and pH dependences. The extinction coefficients of the free dyes and J-aggregates are compiled in Table 1. The values were evaluated for a complete conversion of the dye molecules into J-aggregates bound to the protein. Note that AJ is a reliable measure of the concentration of J-aggregates.8-10,19 The pH was varied by addition of HClO4 or NaOH. The circular dichroism (CD) spectra were recorded by a Jasco J-715 spectrometer with Hamamatsu R376 photomultiplier and Spectra Manager software. To increase the signalto-noise ratio eight repetitive scans were made. The fluorescence spectra were recorded on a spectrofluorimeter (Cary, eclipse). As a reference, rhodamine 101 in air-saturated methanol, λexc ) 500-560 nm, was used.28 The measurements refer to steadystate conditions at 24 °C in air-saturated (buffer-free) aqueous solution at pH 5.9-6.2 unless indicated otherwise. Results Absorption Properties. In neat aqueous solution, dyes 1-3 are present as an equilibrated mixture of a monomer (M) and a dimer (D). The absorption spectrum of 3 consists of a D-band peaking at λD ) 523 nm and a red-shifted M-band at λM ) 560 nm which is much less intense. For 1 and 2 the M-band looks rather like a shoulder and is clearly pronounced in ethanol, where 1-3 exist as monomers.27 The values of λM and λD are compiled in Table 1. The absorbencies of 1 (6 µM) at λM and λD remain constant on going from pH 6 to either 4 or 12. When RNase was added to an aqueous solution of 1-3 the D-band decreased, accompanied by the appearance of absorption (AJ) with a maximum at λJ (Table 1) which is attributed to a J-aggregate. Figure 1 shows the changes in absorbance for 3, where the J-band is developed with time and matches the decay of the D-band (Figure 1, inset). The ratio of two characteristic absorbances AJ∞/AD∞ (AJ∞/AD0) with RNase changes from 0.3 (0.3) to 5.5 (1) for 3 and 2, respectively (Table 2), where AJ∞, AD∞, and AD0 are absorbancies of J-aggregates and dimers after completion of J-aggregation (∞) and of dimers immediately after mixing (0).

Figure 1. Absorption spectra of 3 (4.3 µM) in the presence of 0.8 µM RNase at pH 6 and at 1, 7, 12, and 70 s. Inset: Absorption at 523 (O) and 629 (2) nm vs time.

TABLE 2: Absorption Ratios Characterizing J-Aggregation: ΑJ∞/ΑD∞ (ΑJ∞/AD0)a protein

dye 1

dye 2

dye 3

RNase lysozyme trypsin BSAc

1.6 (0.71) 2.1 (0.72) 2.6b (0.88b)

5.5 (1.07) 5.2 (1.12) 1.9 (1) 0.2 (13)

0.3 (0.3)

a For [dye] ) 5 µM, [P]/[D] ) 0.1 at pH 6. b For [1] ) 10 µM at pH 4. c [P]/[D] ) 0.04.

TABLE 3: Absorption Maximum and Coherence Size of J-Aggregatea system

[P]/[D]

λJ (nm)

N

lysozyme/1b

0.01 0.1 0.5 2 6 1.2 1.8 3.6 7.1

641 638 633 615 612 647 643 632 622

15 9 5.5 2.2 1.9 4.3 3.6 2.5 1.9

RNase/2c

a

At pH 6. b [1] ) 4 µM. c [RNase] ) 25 µM.

The increase of the protein/dye concentration ratio ([P]/[D]) from 0.01 to 7 results in a hypsochromic shift of λJ (Table 3). However, a bathochromic shift of λJ was observed on a prolonged time scale for ([P]/[D]) > 0.5; an example for RNase/1 is shown in Figure 2. Under these conditions the formation of J-aggregates with λJ ) 624 nm is completed after mixing (Figure 2a). Figure 2b shows difference spectra between those measured at different times relative to the spectrum of the initially formed J-aggregate. It can be seen that ∆AJ has the peak at 659 nm and its value increases with time. J-aggregation takes place for lysozyme/1 (Figure 3) and lysozyme/2 but not for lysozyme/3 under similar conditions. Formation of J-aggregates correlates with the decay of dimers (Figure 3 inset). The relative yield of the J-aggregate of the lysozyme/1 system was found to depend on pH and [P]/[D]. The lysozyme-induced J-aggregation occurs in neutral and slightly alkaline solution with a midpoint at pH 9.4 and practically does not proceed at pH > 11 (Figure 4). AJ increases with [P]/[D], remains constant between [P]/[D] ) 0.2 and 2, and decreases above [P]/[D] ) 2 (Figure 5). The decrease of AJ is accompanied by the appearance of the absorbance at λbM ) 582 nm, which is assigned to the monomer bound to lysozyme. Moreover, the dimer band disappears at [P]/[D] )

J-Aggregation of Ethyl meso-Thiacarbocyanine Dyes

J. Phys. Chem. B, Vol. 111, No. 33, 2007 10025

Figure 5. Plot of AJ vs the lysozyme/1 concentration ratio at pH 6. Inset: Absorption spectra of 0.1, 1, and 50 µM lysozyme, lines 1-3, respectively. Figure 2. (a) Absorption spectra of 1 (7 µM) in the presence of RNase (3 µM), prior to (solid line) and 2 s after mixing (dashed line) at pH 6 and (b) differences at 250, 500, and 1600 s, lines 1-3, respectively.

Figure 6. Absorption spectra of 2 (6.5 µM) in the presence of 1 µM trypsin at pH 6 and at 2, 200, 600, and 1200 s. Inset: Absorption at 536 (O) and 656 (2) nm vs time. Figure 3. Absorption spectra of 1 (3.7 µM) in the presence of 0.2 µM lysozyme at pH 8.4 and at 2, 30, and 400 s. Inset: Absorption at 530 (O) and 646 (2) nm vs time.

Figure 4. pH dependence of AJ for 1 (2.3 µM) in the presence of 0.2 µM lysozyme (O) and 1 µM trypsin (2).

0.2-2 and appears again at [P]/[D] ) 2-100, where the peak is red-shifted to 539 nm. Therefore, the free dimers at high loading are converted into bound dimers at low-loading conditions. Similar to the RNase/2 system, λJ is blue-shifted by 29 nm, when for lysozyme/1 [P]/[D] is increased from 0.01 to 6 (Table 3, Figure 5 inset). In the case of trypsin at pH 5.9-6.2 J-aggregation was observed only for 2; an example is presented in Figure 6. At

Figure 7. Plots of AbM (O) and If at 607 nm (2) vs the BSA concentration for 1.4 µM 1 at pH 6. Inset: Absorption spectra of 1 (5 µM) with 0, 8, 15, and 75 µM BSA, lines 1-4, respectively.

pH < 6 the J-aggregates were also found for 1 with a midpoint at pH 4.5 (Figure 4) and with λJ ) 642-653 nm depending on pH. No trypsin-induced J-aggregation was found for 3. The presence of BSA does not generally result in Jaggregation. Addition of BSA to 1 in aqueous solution results in a decrease of the D-band and the appearance of an absorption with maximum at λbM ) 587 nm (Figure 7 inset), which is assigned to the monomer bound to the protein. The absorbance (AbM) for [BSA] ) 75 µM almost levels off, and the D-band is

10026 J. Phys. Chem. B, Vol. 111, No. 33, 2007

Slavnova et al. TABLE 5: Rate of J-Aggregation in the Presence of Proteins

Figure 8. Normalized fluorescence spectra of 1 (2 µM) at pH 6: (a) emission (right, λexc ) 560 nm) and excitation (left) in the presence of 15 µM BSA (dotted lines, λem ) 620 nm) and 0.5 µM lysozyme (solid line, λem ) 670 nm) and (b) emission (λexc ) 560 nm) for 30 (solid line) and 90 µM (dashed line) lysozyme.

TABLE 4: Fluorescence Data of J-Aggregate and Bound Monomera protein RNase lysozyme trypsin BSA

a

dye

λfj (nm)

ΦfJ

1 2 3 1 2 1 2 1 2 3

657 659 650 657 659 657 663

0.04 0.06

652

0.01 0.06

λfbM (nm)

ΦfbM

585 606

0.07 607 617 600

0.22 0.03 0.13

At pH 6, trypsin/1, pH 3.

absent. For 8 µM BSA the AD value is increased (2.4-fold) and AbM decreases (0.7-fold) on going from pH 6 to 12. This indicates that the 0.01 M hydroxide ions release the bound dyes. It is noteworthy that the formation of bound monomer occurs immediately after injection of BSA into the dye solution and AbM remains constant with time. The results are similar for BSA/ 3. BSA/2, however, is a special case since a J-aggregate with maximum AJ∞/AD∞ ) 0.2 was observed at [P]/[D] ) 0.04 (Table 2). Fluorescence Properties. RNase, lysozyme, and trypsin were found to enhance the fluorescence of the dyes. As an example, the fluorescence spectra of 1 in the presence of lysozyme are shown in Figure 8. For 2 and 3 the results are similar (Table 4). It is noteworthy that 1-3 in neat aqueous solution do not exhibit fluorescence (Φf < 0.001). Fluorescence of J-aggregates was observed for all protein/dye systems under conditions of J-aggregate formation. The J-aggregate as the origin of fluorescence is supported by the emission maximum at λfJ ) 657 nm and the excitation spectrum of lysozyme/1 (Figure 8b), showing that the maximum is positioned at 647 nm which is close to λJ. The lysozyme-enhanced fluorescence quantum yield of 1 was determined to be ΦfJ ) 0.01. Low ΦfJ values were also found for both RNase/1 and RNase/2 (Table 4). The fluorescence properties depend also on [protein]. With an increase of the protein concentration the intensity initially increases, followed by its decrease and the appearance of a new band at λfM ) 606 nm for lysozyme/1. This band is attributed to monomers bound to the protein. At high protein concentration (0.5 mM, [P]/[D] > 100) only the bound monomer fluorescence was observed.

protein

dye

[P]/[D]

pH

(dAJ/dt)max (s-1)

lysozyme

1

trypsin

2

0.02 0.02 0.02 0.04 0.14 0.20

5.5 7.4 10 6.0 6.0 6.0

0.015 0.009 0.0034 0.018 0.0005 0.0007

An enhanced fluorescence was observed for all BSA/dye systems at [P]/[D] > 0.04 (Table 4). From a comparison of excitation and emission spectra of 1 (Figure 8b), it follows that the fluorescence origin is the bound monomer. The highest intensity takes place for excitation at 596 nm, which is close to λbM. The quantum yields of fluorescence of 1, 3, and 2 were determined to be Φf ) 0.22, 0.13, and 0.03, respectively. The fluorescence intensity is increased with increasing [BSA] in a way similar to the absorbance AbM (Figure 7). An enhancement in Φf of cyanine dyes not related with J-aggregation has already been reported in other serum albumin studies.22-25 A special case is the BSA/2 system, where J-aggregate-induced fluorescence with λfJ ) 652 nm (Table 4) was recorded at [P]/[D] ) 0.01-0.04, in contrast to the findings for the other proteins. For BSA/2 AJ disappears at higher [P]/[D] ratios. Kinetics of J-Aggregation. The curves for both the formation of J-aggregates and the decay of dimers as a function of time are of a nonsigmoidal type with the exception of trypsin/2, where the plots are of sigmoidal form. Examples are shown in Figures 1, 3, and 6 (insets) for RNase/3, lysozyme/1, and trypsin/2, respectively. As a characteristic measure of the kinetics of J-aggregation, the maximum derivative, (dAJ/dt)max, was used. This was obtained from the differentiated kinetic curves and taken as the maximum rate, assuming that AJ is proportional to the concentration of J-aggregate, as previously applied.8-10,19 The (dAJ/dt)max value increases nonlinearly with an increase of the dye concentration and in particular when 1 grows from 1.4 to 5 µM, (dAJ/dt)max increases 36 times, respectively (Table 5). The rate also depends on pH; for 1 and [lysozyme] ) 0.2 µM it is reduced by 40 times with increasing pH from 5.5 to 10. For an analysis of the J-aggregation kinetics we use eqs 1 and 2. They describe the time course of formation of Jaggregates and the decay of dimers by means of stretched exponential functions.29

AJ ) AJ0 + (AJ∞ - AJ0){1 - exp(-kt)m}

(1)

AD ) AD∞ + (AD0 - AD∞){exp(-kt)m}

(2)

Here, AJ0 is the absorbance of the J-aggregate immediately after the mixing (0). The results of the fitting procedure for three characteristic cases, the apparent rate constant k, and the parameter m for J-aggregation and for the decay of dimers are presented in Table 6. Circular Dichroism. The binding of 1-3 to proteins is supported by the CD spectra. For all protein/dye systems which reveal J-aggregation the doublet (bisignate) signal coinciding with the J-band was observed. Figures 9b and 10b show induced CD spectra of J-aggregates for 2 with lysozyme, trypsin, and RNase, whereby the ellipticity (amplitude of the signal, Θ) of RNase is the largest. The position of the subbands, Θ, and their signs for the systems under study are compiled in Table 7. It should be noted that lysozyme and trypsin demonstrate the opposite chirality (Figure 9b). The CD spectra for RNase/2 and lysozyme/2 underline the left-handedness of these systems. The

J-Aggregation of Ethyl meso-Thiacarbocyanine Dyes TABLE 6: Parameters Used for Analysis of the Time Dependence of J-Aggregationa [P]/[D] AD0 AD AJ0 AJ k (s-1) m a

lysozyme/1

trypsin/2

RNase/3

0.05 0.32 0.23 0.013 0.15 0.025 ( 0.005 0.52 ( 0.02

0.1 0.52 0.24 0.001 0.45 0.0022 ( 0.0003 1.24 ( 0.04

0.2 0.41 0.33 0.013 0.23 0.025 ( 0.005 0.52 ( 0.02

See Figures 1, 3, and 6.

Figure 9. (a) Absorption and (b) induced CD spectra for 2 (5 µM) in the presence of lysozyme (1 µM, line 1) and trypsin (1 µM, line 2) at pH 6.

Figure 10. Effect of EtOH (none, line 1; 25 vol %, line 2): (a) absorption and (b) induced CD spectra of RNase (1 µM)/2 (10 µM) at pH 6.

chirality is left-handed for all dyes with RNase and lysozyme, whereas it is right-handed for trypsin/1, 2. For BSA/1 a small ( 1 and K .1 µM, free monomers are assumed to interact with the protein. For another class (II), where n is in the range of 0.3-0.4 and K e 1 µM, the dimers are suggested to bind with BSA, which dissociates into monomers bound to the protein. A specific case is BSA/2 which shows the formation of J-aggregates in a small concentration range at low [BSA] e 0.4 µM. This restricts the construction of the Scatchard plot in a wide ν range. However, the analysis of the plot implies that J-aggregation for the BSA/2 case is a

cooperative process. A low number of binding sites in BSA is in agreement with n ) 0.13-0.75, which has been determined for non-covalently linked complexes of HSA with a series of thiacarbocyanine dyes.22,25 It has been pointed out that the interaction of thiacarbocyanines with HSA might result in dye aggregation.22 Interaction of Cyanine Dyes with Proteins. The results obtained with 1-3 and RNase, lysozyme, and trypsin are summarized in Scheme 1. As was shown above for lysozyme/1 (Figure 5), the concentration of J-aggregates increases with increasing protein concentration ([P]/[D] e 0.2, high loading), reaches a plateau, and decreases ([P]/[D] > 2.5, low loading). In the course of decreasing the concentration of J-aggregates, both monomers and dimers are formed bound to the protein as trans- and cis-isomers, respectively. This is probably due to a different nature of the binding sites. For the BSA/1, 3 cases there are two pathways leading to the bound trans-monomer in accordance with the two classes of binding sites (Scheme 2). For pathway I with the monomers K is in the range of 7-27 µM, whereas for pathway II with the dimers K ) 0.3-1.0 µM. For the special case of BSA/2 pathway II leads mostly to J-aggregation at high loading, whereas only trans-monomers are formed via I and II at low loading. An increase of protein concentration in the [P]/[D] range of 0.2-2.5 (Figure 5 inset) might result in a splitting of the large J-aggregates into smaller assemblies. That is reflected in a blue shift of λJ and points to a decrease of the coherence size (N). N is defined by eq 7.41

N - 1 ∆νN ) N ∆ν∞

(7)

Here ∆νN is the spectral shift (cm-1) of any N-mer and ∆ν∞ that of the ∞-mer (1/λJ for [D]/[P] f ∞) with respect to the monomer. N was calculated using eq 8 which follows from a rearrangement of eq 7.

N)

n∆νn n∆νn - (n - 1)∆νN

(8)

The peak of the monomer band for lysozyme/1 is at λbM ) 582 nm. For high loading the number of dye molecules incorporated in the J-aggregate should not exceed the number of binding sites (n ) 15). With this in mind λJ ) 641 nm for [P]/[D] ) 0.01 characterizes the 15-mer (Table 3). Similar calculations were carried out for RNase/2, where n ) 12 and λJ ) 661 nm. It is noteworthy that a decrease of the coherence size from 9 to N ) 2 takes place in the [P]/[D] range of 0.1-2, where AJ is kept constant. It indicates that an increase of the protein concentration does not change the total concentration of dye molecules involved in J-aggregation but results in a reduction of N. Another case results for RNase/1, where a bathochromic shift of λJ by 25 nm occurs (Figure 2). This leads to a growing of the J-aggregate during the 25 min after mixing. A plausible

10030 J. Phys. Chem. B, Vol. 111, No. 33, 2007

Slavnova et al.

SCHEME 2

Figure 13. Time dependences of AJ (O) for trypsin (1 µM)/1 (6 µM) prior (pH ) 6.2) to and after injection of H+ at 100 s (pH ) 3.3), of OH- at 300 s (pH ) 10.9), of H+ at 950 s (pH ) 3.4), and of OH- at 1200 s (pH ) 10.7). Inset: CD spectra at pH 6.2 (line 1) and 3.3 (line 2).

reason is conformational changes of the protein molecules induced by the J-aggregate. Such changes might result in a reduction of the distance between the dye molecules and an increase of the J-aggregate size. This is supported by timedependent CD spectra which reveal a bathochromic shift and a 2-fold shortening of the difference in energy between the positive and negative CD subbands (not shown). The latter indicates an increase of the exciton interaction in the J-aggregate. The lysozyme-induced J-aggregation of 1 occurs in neutral and slightly alkaline solution with a midpoint at pH 9.4 and is practically absent at pH >11 (Figure 4). The reason is the Coulombic repulsive interaction between negatively charged amino acid residues (at high pH) and the anionic dye. The pH dependence of J-aggregation of trypsin/1 is shifted toward lower pH, and, due to a smaller pI of 7 for trypsin, the inflection point is at pH 4.5. The trypsin/1 system was used to control J-aggregation and deaggregation by variation of the pH value. The time dependence of AJ prior to and after injection of HClO4, corresponding to a jump from pH 6.2 to 3.3, is shown in Figure 13. Note that AD is of mirror image type. The pH change results in both fast formation of J-aggregates and the decay of dimers which are blocked by injection of NaOH after 300 s, yielding pH 10.9. A second injection of H+ after 950 s resulting in pH 3.4 leads to a further increase of J-aggregation, which is again blocked by injection of OH- after 1200 s (pH 10.7). The results obtained from absorption measurements are supported by CD spectra (Figure 13 inset). At pH 6.2 there is no CD signal, which, however, appears at pH 3.3. The formation of J-aggregates and their properties are governed by several factors, and among them electrostatic and hydrophobic interactions are the most important. The Coulomb

attractive forces for anionic dyes and proteins with positive charges takes place for RNase/1-3 and lysozyme/1, 2, resulting in a high yield of J-aggregate. For trypsin/1 J-aggregation occurs in acidic solution, where the total charge becomes positive. The change of the total charge of the protein by varying pH provides the possiblity to control the yield. The hydrophobic interactions also play an important role in J-aggregation. Hydrophobicity is increased in the order 3, 1, 2, reflecting in both the J-aggregate yield (Table 2) and the binding constant (Table 9). Hydrophobic interaction prevails over the Coulomb repulsive forces for BSA/2 and trypsin/2, where J-aggregation takes place for negatively charged proteins. No J-aggregation occurs for 1 and 3 in the presence of BSA and trypsin. J-aggregation of 1-3 on RNase, lysozyme, and trypsin is a cooperative process, as it has been established for gelatin.19 The number of the binding sites depends on protein and dye properties and is assumed to depend strongly on the presence of periodic elements such as β-sheets and β-bends in the secondary structure. As it is shown above, the J-aggregates of RNase and lysozyme are formed on extensive β-sheets which contain a high amount of the binding sites (Table 9). The interaction with β-sheets is more efficient than with R-helices11 probably due to the geometry of the peptide backbone in the β-sheets which approaches the most extended chain conformation. The presence of the specific elements in the secondary structure, such as a positively charged groove in RNase and a prominent cleft which traverses all three β-sheets in lysozyme, promotes efficient J-aggregation. The formation of J-aggregates and their absorption and CD spectra might serve as a test for the presence of periodic elements in the protein structure. It should be noted that the binding of an oppositely charged dye results in neutralization of the protein and alters the protein structure. This results in a change of the size of the J-aggregate with time as indicated in the absorption spectra. The existence of a protein-induced CD signal of J-aggregates implies that this system may be considered a superhelix. Conclusion The presence of RNase, lysozyme, trypsin, and BSA as templates induces J-aggregation of three ethyl meso-thiacarbocyanine dyes in aqueous solution, where the dye dimers appear as building blocks. The yield of J-aggregate depends essentially upon the protein/dye concentration ratio. At high loading the equilibrium between free dimers and J-aggregates is established and shifted with increasing protein concentration toward bound dimers and monomers. Electrostatic and hydrophobic interactions are major driving forces in J-aggregation. The pH is a decisive factor which governs the rate and the yield of J-aggregates. J-aggregation is a cooperative process, and the number of the binding sites is one of the major factors controlling the J-aggregate size. The ability of a protein to induce J-aggregation is increased in the order BSA, trypsin, lysozyme,

J-Aggregation of Ethyl meso-Thiacarbocyanine Dyes and RNase, and 2 as the most hydrophobic dye shows the highest aggregation efficiency. Acknowledgment. We thank Professor Wolfgang Lubitz for his support and Mr. Leslie J. Currell and Mr. Andreas Go¨bels for technical assistance. A.K.C. is grateful to the Deutsche Forschungsgemeinschaft and the Russian Fund of Basic Researches (Grant 06-03-32298) for financial support. References and Notes (1) Herz, A. H. In The Theory of the Photographic Processes, 4th ed.; James, T. H., Ed.; Macmillan: New York, 1977; p 235. (2) Herz, A. H. Photogr. Sci. Eng. 1974, 18, 323. (3) Daltrozzo, E.; Scheibe, G.; Gschwind, K.; Haimerl, F. Photogr. Sci. Eng. 1974, 18, 441. (4) De Rossi, U.; Da¨hne, S.; Meskers, S. C. J.; Dekkers, H. P. J. M. Angew. Chem. 1996, 108, 827. (5) von Berlepsch, H.; Kirstein, S.; Bo¨ttcher, C. Langmuir 2002, 18, 7699. (6) (a) von Berlepsch, H.; Bo¨ttcher, C.; Quart, A.; Burger, C.; Da¨hne, S.; Kirstein, S. J. Phys. Chem. B 2000, 104, 5255. (b) von Berlepsch, H.; Kirstein, S.; Bo¨ttcher, C. J. Phys. Chem. B 2003, 107, 9646. (c) von Berlepsch, H.; Kirstein, S.; Bo¨ttcher, C. J. Phys. Chem. B 2004, 108, 18725. (7) (a) Tanaka, T.; Iwasaki, M. J. Imag. Sci. 1989, 33, 78. (b) Struganova, I. A.; Lim, H.; Morgan, S. A. J. Phys. Chem. B 2002, 106, 11047. (c) Xiang, J.; Yang, X.; Chen, C.; Tang, Y.; Yan, W.; Xu, G. J. Colloid Interface Sci. 2003, 258, 198. (d) Zakharova, G. V.; Kombaev, A. P.; Chibisov, A. K., Khim. Vys. Energ. 2004, 38, 211; High Energy Chem. (Engl. Transl.) 2004, 38, 180. (e) Ikegami, K., Chem. Phys. Lett. 2004, 401, 590. (f) Alfimov, M. V.; Shtykova, A. A.; Razumov, V. F. Khim. Vysok. Energ. 2006, 40, 21; High Energy Chem. (Engl. Transl.) 2006, 40, 18. (8) Chibisov, A. K.; Go¨rner, H.; Slavnova, T. D. Chem. Phys. Lett. 2004, 390, 240. (9) Slavnova, T. D.; Chibisov, A. K.; Go¨rner, H. J. Phys. Chem. A 2005, 109, 4758. (10) Chibisov, A. K.; Slavnova, T. D.; Go¨rner, H. Chem. Phys. Lett. 2006, 424, 307. (11) Hermel, H.; Ho¨ltje, H.-D.; Bergemann. S.; De Rossi, U.; Kriwanek. J. Biochim. Biophys. Acta 1995, 1252, 79. (12) (a) Lu, L.; Helgeson, R.; Jones, R. M.; McBranch, D.; Whitten, D. J. Am. Chem. Soc. 2002, 124, 483. (b) Whitten, D.; Achyuthan, K. E.; Lopez, G. P.; Kim, O.-K. Pure Appl. Chem. 2006, 78, 2313. (13) Pasternack, R. F.; Fleming, C.; Herring, S.; Collings, P. J.; dePaula, J.; DeCastro, G.; Gibbs, E. J. Biophys. J. 2000, 79, 550. (14) von Berlepsch, H.; Kirstein, S.; Hania, R.; Didraga, C.; Pugzˇlys. A.; Bo¨ttcher, C. J. Phys. Chem. B 2003, 107, 14176. (15) (a) Kimura, T.; Arimori, S.; Takeuchi, M.; Nagasaki T.; Shinkai, S. J. Chem. Soc., Perkin. Trans. 2 1995, 1889. (b) Seifert, J. L.; Connor, R. E.; Kushon, S. A.; Wang, M.; Arimitage, B. A. J. Am. Chem. Soc. 1999, 121, 2987. (c) Wang, M.; Silva, G. L.; Arimitage, B. A. J. Am. Chem. Soc. 2000, 122, 9977. (d) Garoff, R. A.; Litzinger, E. A.; Connor, R. E.; Fishman, I.; Arimitage, B. A. Langmuir 2002, 18, 6330. (e) Wang, M.; Dilek, I.; Arimitage, B. A. Langmuir 2003, 19, 6449. (f) Hannah, K. C.; Arimitage, B. A. Acc. Chem. Res. 2004, 37, 845. (g) Mikelsons, L.; Carra, C.; Shaw,

J. Phys. Chem. B, Vol. 111, No. 33, 2007 10031 M.; Schweitzer, C.; Scaiano, J. C. Photochem. Photobiol. Sci. 2005, 4, 798. (h) Sforza, S.; Scaravelli, E.; Corradini, R.; Marchelli, R. Chirality 2005, 17, 515. (16) Kim, O.-K.; Je, J.; Jernigan, G.; Buckley, L.; Whitten, D. J. Am. Chem. Soc. 2006, 128, 510. (17) Dickinson, H. O. Nature 1949, 26, 485. (18) (a) Avdeeva, V. I.; Shapiro, B. I. Dokl. Akad. Nauk 2003, 389, 62; Dokl. Phys. Chem. (Engl. Transl.) 2003, 389, 77. (b) Avdeeva, V. I.; Shapiro, B. I. Khim. Vys. Energ. 2005, 39, 467; High Energy Chem. (Engl. Transl.) 2005, 39, 408. (19) Go¨rner, H.; Chibisov, A. K.; Slavnova, T. D. J. Phys. Chem. B 2006, 110, 3917. (20) Williams, R. J.; Lipowska, M.; Patonay, G.; Strehkowski, L. Anal. Chem. 1993, 65, 601. (21) Zheng, H.; Mao, Y. X.; Li, D. H. Zhu, C. Q. Anal. Biochem. 2003, 318, 86. (22) Tatikolov, A. S.; Costa, S. M. B. Biophys. Chem. 2004, 107, 33. (23) (a) Losytskyy, M. Yu.; Volkova, K. D.; Kovalska, V. B.; Makovenko, I. E.; Slominskii, Yu. L.; Tolmachev, O. I.; Yarmoluk, S. M. J. Fluoresc. 2005, 15, 849. (b) Yarmoluk, S. M.; Kryvorotenko, D. V.; Balanda, A. O.; Losytskyy, M. Yu.; Kovalska, V. B. Dyes Pigm. 2001, 51, 41. (24) Kim, J. S.; Kodagahally, R.; Strehkowski, L.; Patonay, G. Talanta 2005, 67, 947. (25) (a) Zhang, Y.; Du, H.; Tang, Y.; Xu G.; Yan, W. Biophys. Chem. 2007, 128, 197. (b) Zhang. Y.; Xiang, J.; Tang, Y.; Xu, G.; Yan, W. ChemPhysChem. 2007, 8, 224. (c) Panova, I. G.; Sharova, N. P.; Dmitrieva, S. B.; Poltavtseva, R. A.; Sukhikh, G. T.; Tatikolov, A. S. Anal. Biochem. 2007, 361, 183. (26) Prasad, N. Nanophotonics; Wiley: New York, 2004. (27) Chibisov, A. K.; Slavnova, T. D; Go¨rner, H. Chem. Phys. Lett. 2004, 386, 301. (28) Karstens, T.; Kobs, K. J. Phys. Chem. 1980, 84, 1871. (29) (a) Micali, N.; Monsu` Scolaro, L.; Romeo, A.; Mallamace, F. Physica A 1998, 249, 501. (b) Mallamace, F.; Monsu` Scolaro, L.; Romeo, A.; Micali, N. Phys. ReV. Lett. 1999, 82, 3480. (c) Mallamace, F.; Micali, N.; Romeo, A.; Monsu` Scolaro, L. Curr. Opin. Colloid Interface Sci. 2000, 5, 49. (30) Waugh, D. W. AdV. Protein Chem. 1954, 9, 325. (31) Laskowski, M.; Laskowski, M., Jr. AdV. Protein Chem. 1954, 9, 203. (32) Peters, Th., Jr. AdV. Protein Chem. 1985, 37, 161. (33) Carter, D. C.; Ho, J. X. AdV. Protein Chem. 1994, 45, 153. (34) McRae, E. G.; Kasha, M. In Physical Processes in Radiation Biology; Augenstein, I., Mason, R., Rosenberg, B., Eds.; Academic Press: New York, 1964; p 23. (35) Khimenko, V.; Chibisov, A. K.; Go¨rner, H. J. Phys. Chem. A 1997, 101, 7304. (36) Karapetyan, N. H.; Madakyan, V. N. Russ. J. Bioorg. Chem. 2004, 30, 172. (37) Andrade, S. M.; Costa, S. M. B. Biophys. J. 2002, 82, 1607. (38) Cantor, C. R.; Schimmel, P. R. Biophysical Chemistry. Parts I and III. The Behavior of Biological Macromolecules; Freeman: San Francisco, CA, 1980. (39) Scatchard, G. Ann. N. Y. Acad. Sci. 1949, 51, 660. (40) Kodaka, M. Biophys. Chem. 2004, 107, 243. (41) Muenter, A. A.; Brumbaugh, D. V.; Apolito, J.; Horn, L. A.; Spano, F. C.; Mukamel, S. J. Phys. Chem. 1992, 96, 2783.