J. Phys. Chem. B 2008, 112, 3451-3461
3451
Energy Transfer Photophysics from Serum Albumins to Sequestered 3-Hydroxy-2-Naphthoic Acid, an Excited State Intramolecular Proton-Transfer Probe Pinki Saha Sardar,† Subhodip Samanta,† Shyam Sundar Maity,† Swagata Dasgupta,‡ and Sanjib Ghosh*,† Department of Chemistry, Presidency College, Calcutta 700 073, India, and Department of Chemistry, Indian Institute of Technology, Kharagpur 721302, India ReceiVed: June 14, 2007; In Final Form: December 11, 2007
The steady-state and time-resolved studies of the sensitized emission of the excited-state proton transfer (ESIPT) probe 3-hydroxy-2-naphthoic acid (3HNA) when bound to bovine serum albumin (BSA) and human serum albumin (HSA) indicate that the nonradiative dipole-dipole Fo¨rster type energy transfer from Trp singlet state of proteins to the ESIPT singlet state of 3HNA is greater in the case of HSA. This is supported by the distance and the orientation of the donor-acceptor pair obtained from the protein-ligand docking studies. The docking studies of the complex of BSA-3HNA also indicate that Trp 134 rather than Trp 213 is involved in the energy transfer process. The local environment of Trp 134 in BSA rather than that of Trp 213 is perturbed because of interaction with 3HNA as revealed by the optical resolution of Trp 134 phosphorescence in the complex at 77 K. Docking studies support the larger rotational correlation time, θc (≈ 50 ns), observed for Trp residue/residues in the complexes of HSA and BSA compared with that in the free proteins.
1. Introduction Interactions of the excited-state intramolecular proton transfer (ESIPT) probe, flavonols,1 4-hydroxy-5-azaphenanthrene2 with proteins, had been initiated by Kasha et al. 3-Hydroxy-2naphthoic acid (3HNA) is known for its local 1(π-π*) emission as well as the ESIPT emission from a rotamer.3 Our recent observation of the significant enhancement of the ESIPT state emission of 3HNA in the presence of bovine serum albumin (BSA) and human serum albumin (HSA) compared with that observed in aqueous buffer of pH 73,4 has been ascribed to a motional restriction of the probe imposed because of the interaction with the proteins4 rather than the effect of change of micropolarity and microviscosity of the immediate environment of the probe. Tryptophan residues in proteins and protein-ligand complexes act as an intrinsic probe and are often characterized by steady-state and time-resolved fluorescence, steady-state and time-resolved anisotropy monitoring fluorescence.5-20 The fluorescence spectra of tryptophan residues are usually broad, and the fluorescence lifetime even in a protein having single tryptophan exhibits multiexponential decay.6 However, low temperature (77 K) phosphorescence (LTP) spectra of tryptophan residues in proteins in a suitable cryosolvent always give structured spectra with a definite (0,0) band, characteristic of the tryptophan environment.21-25 The position and the width of the (0,0) band along with the overall structural features of the spectra provide definitive information regarding the polarity, nature of solvent exposure, and hydrophobicity and the homogeneity of the immediate environment of the tryptophan residue. LTP along with conventional steady-state and time-resolved * Corresponding author. Phone: 033-2241 3893. E-mail: sanjibg@ cal2.vsnl.net.in. † Presidency College. ‡ Indian Institute of Technology.
fluorescence spectra can provide meaningful understanding of protein-protein interaction26 and protein-ligand18,27 interactions. In the present work, we explored the interaction between the probe 3HNA and both the serum albumins by steady-state and time-resolved studies of Trp emission of the proteins HSA and BSA and the sensitized ESIPT emission of the probe. Serum albumins, the most abundant protein in the blood stream, account for about 60% of the total plasma protein. The most important physiological function of these proteins is to bind and transport several exogenous and endogenous molecules like fatty acids, nutrients, steroids, and sparingly soluble drugs. HSA is a globular protein consisting of 585 amino acids and is cross linked by seventeen disulfide bonds. It is considered to have three specific domains, I, II, and III, each of which consists of two subdomains a and b possessing common structural motifs.28 HSA contains a single tryptophan residue at 214. A comparative study of the amino acid sequences of BSA and HSA by Brown shows that they have similar general structural features, the difference in sequence being generally conservative.29 The BSA molecule is made up of 3 homologous domains which are divided into 9 loops by 17 disulfide bonds. BSA has two tryptophan residues, Trp 134 in the first domain and Trp 213 in the second domain.30 BSA and HSA have 20 and 18 tyrosine (Tyr) residues, respectively. However, the exact crystal structure of BSA is unknown till date. In this article, the efficiency and the rate constant of the photoinduced energy transfer (ET) from the excited Trp residues to the 3HNA for both of the albumins have been investigated monitoring the Trp emission and the ESIPT emission of 3HNA. We also present the rotational correlation time of Trp residue/ residues in HSA and BSA induced by binding of 3HNA using time-resolved anisotropy decay studies. Phosphorescence of Trp/ Tyr residues in the wild type BSA and HSA and in the complexes with 3HNA in 40% EG matrix at 77 K have been used to find any perturbation of the Trp/Tyr environment due
10.1021/jp074598+ CCC: $40.75 © 2008 American Chemical Society Published on Web 02/23/2008
3452 J. Phys. Chem. B, Vol. 112, No. 11, 2008
Sardar et al.
Figure 1. Structural representation of 3HNA in its neutral and anionic form in the ground and excited states.
to binding with 3HNA. Protein-probe docking studies is employed to find the probable location of the probe molecule bound to the proteins, to observe any change in the immediate environment of Trps due to binding, and to find the perturbation of any other residue due to binding. 2. Experimental Section 2.1. Materials. BSA, HSA, and 3-hydroxy-2-naphthoic acid (3HNA) were purchased from Sigma-Aldrich, U.S.A. Ethylene glycol was purchased from Alfa Aeiser. The 3HNA was purified by repeated crystallization from EtOH. Phosphate buffer of pH 7 was prepared in triply distilled water and used for making all experimental solutions. 2.2 Instrumentation. UV-vis absorption spectra were recorded on a Hitachi U-3210 spectrophotometer at 298 K. The steady-state emission measurements were carried out using a Hitachi Model 4010 spectrofluorimeter (equipped with a 150 W xenon lamp) using a 1 cm path length quartz cuvette. All of the measurements at 298 K were made by exciting the samples at 280 nm using 5 nm band passes for excitation and emission using the correct mode of the instrument. Inner filter effects have been eliminated in all of the emission spectra. Emission studies at 77 K were made using a Dewar system having a 5 mm o.d. quartz tube. The freezing of the samples at 77 K was done at the same rate for all the samples. Triplet state emissions were measured in a Hitachi F-4010 spectrofluorimeter equipped with phosphorescence accessories at 77 K. All the samples were made in 40% ethylene glycol for lowtemperature measurements. The samples were excited at 280 nm using a 10 nm band pass, and the emission band pass was 1.5 nm. Singlet state lifetime was measured by Time Master fluorimeter from Photon Technology International (PTI). The system measures the fluorescence lifetime using PTI’s patented strobe technique and gated detection. The software Felix 32 controls all acquisition modes and data analysis of the Time Master system. The sample was excited using a thyratron gated nitrogen flash lamp (full width at half-maximum 1.2 ns) that is capable of measuring fluorescence time-resolved acquisition at a flash rate of 25 kHz as well as with NanoLED (295 nm, full width at half-maximum 1.0 ns). Lamp profiles were measured at the respective excitation wavelengths, namely, 297 and 356 nm (in case of nitrogen flash lamp) and 295 nm (in case of LED) using slits with a band pass of 3 nm using Ludox as the scatterer. The decay of protein samples were recovered using nano LED
(295 nm) and for the ligand emission the 337 nm nitrogen flash lamp were used by nonlinear iterative fitting procedure based on the Marquardt algorithm. Deconvolution technique used can determine the lifetime up to 400 ps with nitrogen flash lamp while the time resolution with nano LED is 300 ps. The quality of fit has been assessed over the entire decay, including the rising edge, and tested with a plot of weighted residuals and other statistical parameters, for example, the reduced χ2 ratio and the Durbin-Watson (DW) parameters. Anisotropy decay measurements are also carried out in Time Master fluorimeter (PTI) using a motorized Glen Thompson polarizer. The anisotropy, r(t), is defined as:
r(t) ) [IVV(t) - G × IVH(t)]/[IVV(t)+2G × IVH(t)] where I(t) terms are defined as intensity decay of ESIPT emission of 3HNA with excitation polarizer orientated vertically and the emission polarizer oriented vertically and horizontally respectively.
G ) IHV(t)/IHH(t) G is the correction term for the relative throughput of each polarization through the emission optics. The entire data analysis is done with the software Felix 32 which analyses the raw data IVV and IVH simultaneously by global multiexponential program and then the deconvolved curves (IDVV and IDVH) are used to construct r(t),31 and from the fitted curve, the correlation time (θ) can be recovered. 2.3. Docking Studies. The crystal structure of HSA (PDB entry 1AO6) was downloaded from the Protein Data Bank.32 Since the structure of bovine serum albumin (BSA) is unavailable in the PDB, a homology model was used for the docking studies. A BLAST search in the PDB with a bovine serum albumin sequence [Swissprot sequence ALBU_BOVIN (P02769)] revealed 75% identity with HSA. This implies that homology modeling approaches could give a reasonably good backbone and buried residue side chains even though surface side chains and loops may be less accurate. The SAM_T06 server was used to make the model structure, which has been used for the docking studies presented here.33 The three-dimensional (3D) structure of the rotamer of 3HNA responsible for the ESIPT emission was generated by Sybyl 6.92 (Tripos Inc., St. Louis, U.S.A.) and the energy-minimized conformation obtained with the help of the TRIPOS force field using Gasteiger-Hu¨ckel charges with a gradient of 0.005 kcal/mol. The FlexX software
Energy Transfer Photophysics from Serum Albumins
J. Phys. Chem. B, Vol. 112, No. 11, 2008 3453 TABLE 1: Singlet State Lifetime of Serum Albumins and Their Complexes with 3HNA at 298 Ka rotational correlation time monitoring 〈τav〉 Trp emission (ns) θc (ns)
singlet state lifetime monitoring Trp emission system
τ1 (ns)
τ2 (ns)
free BSA (10 µM) BSA + 3HNA (1:0.5) BSA + 3HNA (1:1) BSA + 3HNA (1:1.5) BSA + 3HNA (1:2) Free HSA (10 µM) HSA + 3HNA (1:0.5) HSA + 3HNA (1:1) HSA + 3HNA (1:1.5) HSA + 3HNA (1:2)
6.6 (64%) 6.6 (38%) 5.2 (53%) 4.9 (52%) 4.4 (57%) 6.2 (57%) 5.9 (25%) 4.1 (45%) 4.6 (27%) 5.5 (17%)
3.0 (36%) 2.9 (62%) 1.8 (46%) 1.8 (48%) 1.3 (43%) 1.7 (43%) 2.1 (75%) 0.9 (55%) 1.1 (73%) 1.3 (83%)
as part of the Sybyl suite was used for docking of 3HNA to HSA and BSA. The ranking of the generated solutions is performed using a scoring function that estimates the free energy of binding ∆G of the protein-ligand complex.34 The structures corresponding to the minimum score as obtained from the FlexX analysis of the protein-ligand docked structures were chosen in each case. Composite coordinates were generated to form the docked complex. PyMol35 was used for visualization of the docked conformations. The docked structures were generated using d.c values 1, 4, and 10. The structures and structural parameters for d.c ) 10 were presented since there is no further change in the structural parameters with change in the d.c values. 2.4. Accessible Surface Area Calculations. The accessible surface area (ASA) of HSA and BSA (uncomplexed) and their docked complexes with 3HNA was calculated using NACCESS.36 The structures corresponding to the best docked structure as determined from the minimum score were chosen in each case. Composite coordinates of 3HNA and the proteins were generated to form the docked complex. The change in accessible surface area for residue i was calculated using the following equation i i ∆ASAi ) ASAHSA/BSA - ASAHSA/BSA - 3HNAcomplex
If a residue lost more than 10 Å2 ASA when going from the uncomplexed to the complexed state, it was considered as being involved in the interaction. 3. Results and Discussion 3.1. Emission of 3HNA at 298 K. The molecule 3HNA is well-known in literature for its dual emissions (normal and ESIPT emissions peaking around 418 nm and 495-515 nm region respectively) in polar solvents [e.g., EtOH, ACN, water]. It shows normal emission only in nonpolar solvents like toluene and DCM.3 In polar solvents, the emission at 418 nm has been assigned as normal 1(π-π*) emission from the naphthalene moiety of 3HNA (anionic P and R forms, Figure 1) in its keto form and the broad Stokes shifted emis-
53.0 41.0b
48.0
Error in the measurements are ) ( 0.4 ns. Taken from refs 40-41. a
Figure 2. (A) Corrected fluorescence spectra of BSA (10 µM) at 298 K with varying concentration of 3HNA; curves (a-i) represents 0, 5, 7, 10, 13, 15, 17, 18, and 20 µM of 3HNA respectively. Excitation wavelength ) 280 nm; excitation and emission band pass ) 10 nm each. (B) The corrected fluorescence spectra of HSA (10 µM) at 298 K with varying concentration of 3HNA; curves (a-j) represents 0, 5, 7, 10, 11, 13, 15, 17, 18, and 20 µM of 3HNA respectively. Excitation wavelength ) 280 nm; excitation and emission band pass ) 10 nm each.
41.0b
5.3 4.3 3.6 3.4 3.1 4.3 3.2 2.3 2.1 2.0 b
sion at ≈500 nm as 1(π-π*) emission from an enol tautomer (anionic T form, Figure 1) generated by excitedstate intramolecular proton transfer (ESIPT) from hydroxyl groups to the carbonyl oxygen of the 3HNA.3 The quantum yield of 3HNA in phosphate buffer at pH )7 is 0.08.4 3.2. Steady-State and Time-Resolved Fluorescence Studies Monitoring Tryptophan Emission of Proteins at 298 K. Figure 2A shows the room temperature (298 K) emission spectra of free BSA (10 µM) in aqueous phosphate buffer of pH ) 7 and the change in the emission spectra with varying concentration of added 3HNA (5-20 µM) with excitation at 280 nm. Figure 2B represents a similar set of measurements for HSA. The quenching of both the BSA and HSA emissions are observed with the increasing concentration of added 3HNA in the serum albumin-3HNA complex. The emission maxima of HSA shifts toward blue (Figure 2B) with the gradual addition of 3HNA in the medium. However, there is no shift in the emission maxima in the case of complex of BSA (Figure 2A). This suggests that the emitting Trp residue in the complex of HSA is in less polar or in a more hydrophobic environment compared with that in free HSA while the environment of the emitting Trp residues in the complex of BSA remains unaltered. Figure 2A,B also represents the concomitant enhancement of the ESIPT emission of the probe 3HNA peaking at 515 nm in the complexes of serum albumins with excitation at 280 nm. The ESIPT emission of free 3HNA is negligible with excitation at 280 nm. This indicates that the enhancement of the ESIPT emission of 3HNA in the complexes of serum albumin is due to energy transfer from the emitting Trp/ Tyr residues of the protein to 3HNA. Binding Constant of Serum Albumin-3HNA Complex. The steady-state fluorescence quenching of the serum albumins with 3HNA shows a nonlinear Stern-Volmer plot with an upward curvature in both proteins (Figure 3A,B), indicating that both the static and the dynamic quenching of emitting Trp residues of the protein are taking place6 (F0 and F in Figure 3A,B being calculated considering the area under the emission curve of the corrected fluorescence spectra in the absence and in the presence of the quencher 3HNA, respectively). The lifetime measurements monitoring the Trp emission have been carried out in the free proteins and in the protein-3HNA complexes. The aqueous buffer solutions (pH 7) of native BSA and HSA show biexponential decays with average lifetimes
3454 J. Phys. Chem. B, Vol. 112, No. 11, 2008
Sardar et al.
Figure 3. (A) fluorescence quenching study of the tryptophan residues in BSA by 3HNA. (i) Plot of F0/ F against [3HNA], (ii) plot of τ0 /τ against [3HNA], τ represents average lifetime of proteins, and (iii) plot of (F0/ F -1)/[3HNA] against [3HNA] in aqueous buffer at λexc) 280 nm. In each case, concentration of BSA is 10 µM. Excitation and emission band pass ) 10 nm each. The data points in Figure 3Ai represent the experimental points, and the curve shown is obtained by fitting eq 4A. (B) The fluorescence quenching study of the tryptophan residues in HSA by 3HNA (i) Plot of F0/ F against [3HNA], (ii) plot of τ0 /τ against [3HNA], τ represents average lifetime of proteins, and (iii) plot of (F0/ F -1)/ [3HNA] against [3HNA] in aqueous buffer at λexc) 280 nm. In each case, concentration of HSA is 10 µM. Excitation and emission band pass ) 10 nm each. The data points in Figure 3Bi represent the experimental points and the curve shown is obtained by fitting eq 4A.
recovered 5.3 ns (for BSA) and 4.3 ns (for HSA) monitored at 340 nm (Table 1). These values agree well with the lifetime reported earlier37-38 for these proteins. Figure 4A,B represents the decay of the Trp emission of free serum albumins and their respective complexes (1:1) with 3HNA. On gradual addition of 3HNA (5-20 µM) the average fluorescence lifetimes of serum albumins were found to decrease (Table 1). Linear plot of 〈τ0〉/〈τ〉 versus [L] (Figure 3A,B) support dynamic quenching6 according to the eq 1
where 〈τ0〉 and 〈τ〉 are the average lifetimes of serum albumins in the absence and in the presence of 3HNA, KSV is the SternVolmer constant, and kq is the bimolecular quenching constant. By using the slope and the average lifetimes of free BSA and free HSA (Table 1), the values KSV and kq are obtained (Table 2.) Considering both the static and the dynamic quenching and assuming that there are “n” same and independent binding sites on the albumins, according to the following equilibrium:
〈τ0〉/〈τ〉 ) 1 + KSV[L] ) 1 + kqτ0[L]
P + nL ) PLn
(1)
(2)
Energy Transfer Photophysics from Serum Albumins
J. Phys. Chem. B, Vol. 112, No. 11, 2008 3455
TABLE 2: Binding Constants of the Serum Albumins and the 3HNA Complexes at 298 K binding constant (K) at 298 K experimental technique used
a
BSA-3HNA
HSA-3HNA
modified Benesi-Hilderbrand equationa (monitoring ESIPT emission of 3HNA)
5.3 × 105 M-1
2.2 × 105 M-1
steady-state fluorescence anisotropya (monitoring ESIPT emission of 3HNA)
3.5 × 105 M-1
2.2 × 105 M-1
from eq 4 (monitoring the emission of serum albumins)
0.7 × 105 M-1 (kq ) 7.2 × 1012 M-1 s-1)
0.5 × 105 M-1 (kq ) 1.5 × 1013 M-1 s-1)
from eq 4A (monitoring the emission of serum albumins)
1.1 × 105 M-1
0.6 × 105 M-1
From ref 4.
where P is the protein; serum albumins, L is the ligand 3HNA, one can write:
F/F0 ) 1/{(1+ KSV[L]) (1+ K[L]n)}
(3)
where K0 represents the equilibrium constant for the formation of the ground state “dark” (nonemitting) complex. Now, the modified Stern-Volmer equation takes the form:
(F0/F - 1)/[L] ) (KSV + K) + KSVK[L]n
(4)
A linear plot of (F0/ F - 1)/[L] versus [L] (Figure 3A,B) clearly indicates n ) 1. From the slope and the intercept, one obtains K0 for both proteins (Table 2). The protein-ligand fluorescence quenching data were also analyzed according to the following equation39
F ) F0
∑ fi /((1 + KSV,i[L]) exp(Ki[L]))
(4A)
where F and F0 are the corrected fluorescence intensities in the presence and in the absence of quencher 3HNA, respectively, fi is the fractional intensity corresponding to quenching site component i, Ksv,i, is the dynamic quenching constant for the component i, Ki is the static quenching constant for component i, and [L] is the total quencher concentration.39 Inclusion of the exp(Ki[L]) term in the above equation applies in cases where there is an upward curvature in a Stern-Volmer (F0/F vs [L]) plot, due to static quenching, (Figure 3Ai,Bi). The fluorescence quenching data for both the protein-ligand complexes were fitted to eq 4A (taking i ) 1 and fi ) 1 for 1:1 complexes and assuming single site quenching) with inclusion of the static quenching term. The fitted curves are shown in Figure 3Ai,Bi. The values obtained are in close agreement with those calculated using eq 4A and are given in Table 2. The equilibrium constants for the complexes of 3HNA with serum albumins determined from eqs 4 and 4A are somewhat less than those obtained from the modified Benesi-Hildebrand equation and steady-state fluorescence anisotropy monitoring the ESIPT emission of 3HNA in both of the cases4 (Table 2). The singlet state lifetime of the probe 3HNA has also been measured in both the complexes of BSA and HSA with excitation at 297 nm monitoring the ESIPT emission at 515 nm. The decays are best fitted with a single-exponential function and the recovered lifetimes are given in Table 3 (Figure 5). These lifetimes are the same as those observed using excitation at 356 nm. The order of the lifetimes of ESIPT state and those observed monitoring the Trp fluorescence are the same. This indicates that the ET is taking place from the singlet state of the Trp residues.
3.3. Time-Resolved Anisotropy Decays Monitoring Trp Emission. The rotational correlation time, θc, computed from the time-resolved anisotropy decays (Figure 6) monitoring BSA and HSA emissions at 298 K are found to be 53 and 48 ns, respectively (Table 1). The value of θc for free serum albumins was found to be 41 ns.40-41 (The literature values have been checked by our set up and the error is within (1 ns.) The increase in θc in the complexes implies the binding of 3HNA leads to some rigidity of Trp residue/residues in HSA and BSA, respectively. This could be explained from the results obtained from docking studies discussed in the next section. 3.4. Docking Studies. Docking studies reveal the distances of the 3HNA molecule from Trp 134 and Trp 213 in BSA (Figures 7 and 8, Table 4). The docked pose indicates a possible hydrogen bond with Tyr 149. For HSA, the 3HNA molecule is nearer to the sole Trp residue (Figures 7 and 8, Table 4) than the Trps in BSA. In addition to possible hydrogen bonding with Tyr 150, there exists a similar interaction with His 242 and Arg 257. The ASA results given in Table 5A,B indicate that hydrophobic residues near the binding site lose a substantial amount of accessible area on ligand binding. The H bonding with Tyr 149 in BSA and Tyr 150, His 242, and Arg 257 in HSA and the loss of ASA for other nearby hydrophobic residues clearly impose a restricted motion of the 3HNA tautomer. The results thus corroborate our steady-state and time-resolved anisotropy of the ESIPT emission of the probe.4 The larger θc (5.2 ns) observed for HSA-3HNA ESIPT state compared to that in BSA-3HNA (2.4 ns)4 correlate well with the docking results where we find His 242 and Arg 257 are also involved in H bonding apart from Tyr 150 in the complex of HSA. The interactions noted above are probably also responsible for the increased θc values obtained by monitoring Trp emissions of BSA and HSA in the complexes compared with that observed in the respective free proteins (Table 1). This is also supported by the change of the ASA values of several residues in the complexes (Table 5A,B). 3.5. Energy Transfer Study. The steady-state and timeresolved quenching of serum albumins by 3HNA are utilized to determine energy transfer efficiency as well as the rate constants of the energy transfer in both complexes of BSA and HSA. The energy transfer efficiency (ET) is calculated using6
ET ) (1 - FD-A/FD)
(5)
ET ) (1 - τD-A/τD)
(6)
and
where, FD and FD-A are the fluorescence intensities of the serum albumins in the absence and in the presence of quencher 3HNA; τD and τD-A are the average lifetime of the serum albumins in the absence and in the presence of quencher 3HNA.
3456 J. Phys. Chem. B, Vol. 112, No. 11, 2008
Sardar et al. TABLE 3: Lifetime of 3HNA (10 µM) Monitoring ESIPT Emission in Serum Albumins (10 µM) with λexc ) 297 nm at 298 K systems
lifetime of 3HNA [τav (ns)]
free 3HNA in aqueous buffer (pH)7)a BSA:3HNA (1:1) HSA:3HNA (1:1)
1.5 2.8 3.6
a λexc ) 356 nm (the intensity of the emission with λexc ) 297 nm in aqueous buffer is very weak and the decay is not measured).
where FA and FAD are the fluorescence intensities of the acceptor 3HNA in the absence and in the presence of donor serum albumins with excitation at 280 nm. A and AD are the extinction coefficient of the acceptor 3HNA at 280 nm in the absence and in the presence of donor serum albumins. The energy transfer efficiencies calculated from the eqs 5, 6, and 7 are given in Table 6. The greater value of energy transfer efficiency in HSA-3HNA complex reflects that the probe molecule resides nearer to the Trp residue in HSA than that in BSA. Docking studies discussed in section 3.4 also support this view. The rate constant of the energy transfer has been evaluated using6
kET ) 〈τD〉-1 (ET /1 - ET)
(8)
kET ) ET〈τDA〉-1
(9)
and
The kET values thus obtained are provided in Table 6. It is to be noted that 〈τDA〉 in eq 9 represents the average lifetime for the complex where the ratio of the protein and ligand is 1:1. The distances (>10 Å) of Trp residues from 3HNA in BSA and HSA obtained from docking studies indicate singlet-singlet (Trp singlet to ESIPT singlet state of 3HNA) nonradiative energy transfer is responsible for sensitized ESIPT emission of 3HNA. According to Fo¨rster theory, the nonradiative singletsinglet energy transfer efficiency mainly depends on the extent of overlap between the donor emission and the acceptor absorption,42 the distance, and the orientation of the donor and acceptor molecules. The rate constants kET, in s-1, for energy exchange between two species coupled by the Fo¨rster transfer mechanism42 is given by
kET = (8.71 × 1023) k0qr-6κ2Jn-4
Figure 4. (A) Fluorescence decay of free BSA (10 µM, blue), complex of BSA (10 µM) with 3HNA (10 µM, black) in aqueous buffer, λexc ) 295 nm (LED), λmonitor ) 340 nm (the excitation and emission band passes are 10 nm each). The best fitted curve of BSA-3HNA complex is given in red color. (B) Fluorescence decay of free HSA (10 µM), complex of HSA (10 µM) with 3HNA (10 µM) in aqueous buffer, λexc ) 295 nm (LED), λmonitor ) 340 nm (the excitation and emission band passes are 10 nm each).
Since there is no overlap of the donor (Trp) emission and the ESIPT emission of 3HNA (Figure 2A,B), ET is also determined monitoring the enhancement of the ESIPT emission of the acceptor 3HNA using the following equation:6
ET ) (FAD/FA- 1) A (λD )/AD (λD ) ex
ex
(7)
(10)
where k0 is the decay constant for donor in the absence of acceptor, q is the radiative quantum yield in the absence of energy transfer, r is the distance between the centers of the donor and acceptor (in Å), n is the refractive index of the medium, J is the overlap integral between the donor luminescence and the acceptor absorption spectrum, and κ is the orientation factor for the dipole-dipole interaction. Assuming J, n, q, and κ2 to be same for the system BSA3HNA and HSA-3HNA, one can calculate the ratio of the kET in the two systems using the donor-acceptor distance (r) obtained from the docking studies.
(kET)HSA-3HNA (kET)BSA-3HNA
)
[〈τD〉]BSA[(r)6]BSA-3HNA [〈τD〉]HSA[(r)6]HSA-3HNA
(11)
The results calculated are compared with the experimental ratio of the kET (assuming that Trp 134 being nearer to 3HNA in the BSA-3HNA complex is mainly involved in the ET
Energy Transfer Photophysics from Serum Albumins
Figure 5. Fluorescence decay of 3HNA (10 µM) monitoring the ESIPT emission with serum albumins (10 µm) at 298 K in aqueous buffer with λexc ) 297 nm; excitation and emission band-pass ) 10 nm each.
J. Phys. Chem. B, Vol. 112, No. 11, 2008 3457
Figure 7. Docked poses of 3HNA with (A) BSA and (B) HSA. Certain parts of the protein have been removed for clarity. Distances are in angstroms.
Figure 6. Fluorescence anisotropy decay of BSA-3HNA complex (2:1) monitoring the tryptophan emission of BSA at 298 K; inset: IVV and IVH represent decays of tryptophan emission in BSA-3HNA complex with excitation polarizer at vertical position and emission polarizer at vertical and horizontal position respectively. λexc ) 297 nm; excitation and emission band pass ) 15 nm each.
process, Table 6). Although the transition dipole moment direction of Trp is known,43 the direction of the transition dipole moments of the ESIPT emission is not known. Thus, it is difficult to estimate actual r and κ2 values of the donor-acceptor pair in the complexes. We presented the theoretical ratio of kET in the complexes using the distance between the oxygen atom of the -OH group of 3HNA and the center of the indole unit of Trp residues in proteins (r3 in Table 4) as donor-acceptor distance. Since the angles between the indole plane of Trp 134 in BSA/Trp 214 in HSA and the naphthoic acid plane obtained
Figure 8. Distances (in Å) obtained from docked poses of 3HNA with (A) BSA and (B) HSA.
by docking studies are quite different (Table 4), the κ2 values for the Trp 134-3HNA pair (in BSA) and Trp 214-3HNA pair (in HSA) will be different. While taking this into account, the agreement of the experimental ratio of the kET with that calculated is quite well and supports our contention that the Trp134 in BSA is mainly involved in ET (Figure 7A and Table 6).
3458 J. Phys. Chem. B, Vol. 112, No. 11, 2008
Sardar et al.
TABLE 4: Distances of Tryptophan Residues of Serum Albumins from 3HNA and Angles Between the Indole Plane of Tryptophan and the Naphthoic Acid Plane Obtained by the Docking Studies distance
BSA Trp 213
BSA Trp 134
HSA Trp 214
from the center of the naphthalene ring of 3HNA to the center of the indole moiety of Trp residues in proteins. (r1)
22.79 Å
16.60 Å
10.94Å
from the -OH group of 3HNA to indole nitrogen of Trp residues in proteins (r2)
20.98 Å
15.63 Å
13.74 Å
from the oxygen atom of the -OH group of 3HNA to the center of the indole moiety of Trp residues in proteins (r3)
19.80 Å
16.19Å
12.28 Å
angles 129.180
3HNA
TABLE 5: (A) Accessible Surface Area (ASA) (in Å2) for the Docked Complexes of BSA with 3HNA and (B) Accessible Surface Area (ASA) (in Å2) for the Docked Complexes of HSA with 3HNA (A)
amino acid ASA for ASA for change in residues of BSA uncomplexed (Å2) complexed (Å2) ASA (Å2) ARG 10 ASP 13 LEU 14 LEU 22 TRP 134 TYR 149 PRO 151 TRP 213 LEU 250 ASP 254 ALA 253 ARG 256 ALA 257 ALA 260 LEU 282 LEU 283 LYS 285 SER 286 HIS 287
(B)
53.44 65.33 21.91 5.5 53.3 47.04 16.24 65.37 38.65 67.66 27.36 72.46 19.03 1.47 76.07 33.63 38.42 28.94 52.78
39.83 64.62 9.74 1.11 53.3 32.97 10.79 65.37 36.12 58.5 4.38 70.72 8.73 1.40 47.48 15.05 38.13 19.31 50.32
13.61 0.71 12.17 4.39 0 14.07 5.45 0 2.53 9.16 22.98 1.74 10.3 0.07 28.59 18.58 0.29 9.63 2.46
amino acid ASA for ASA for change in residues of HSA uncomplexed (Å2) complexed (Å2) ASA (Å2) TYR 150 GLU 153 GLN 196 LYS 199 TRP 214 LEU 219 ARG 222 PHE 223 LEU 224 LEU 238 VAL 241 HIS 242 ARG 257 LEU 260 ALA 261 ILE 264 SER 287 ILE 290 ALA 291
17.3 11.65 8.5 32.01 61.74 13.13 37.12 3.73 2.49 30.64 0.71 5.23 19.65 17.72 4.27 10.89 8.88 13.69 36.99
6.41 11.15 8.15 22.92 61.74 0.69 27.23 0.22 0.02 1.34 0.62 0 6.42 5.16 3.72 0.11 6.4 1.18 8.54
10.89 0.5 0.35 9.09 0 12.44 9.89 3.51 2.47 29.30 0.09 5.23 13.23 12.56 0.55 10.78 2.48 12.51 28.45
3.6. Phosphorescence Studies. The phosphorescence spectra of free BSA and its complex with 3HNA in 40% ethylene glycol at 77 K are compared in Figure 9A,B with λexc ) 280 nm and λexc ) 295 nm. Free BSA exhibits a single (0,0) band at 412.6 nm (Table 7) while the BSA-3HNA complex shows a shoulder at 405.6 nm along with a distinct (0,0) band at 412.8 nm with λexc ) 280 nm (Table 7, Figure 9A: inset). Figure 10A,B shows the phosphorescence spectra of HSA and HSA-3HNA complex
19.320
107.570
under similar experimental conditions as used for BSA. The position of the (0,0) band along with their band widths for λex ) 280 nm in the different systems are given in Table 7. The phosphorescence spectra in both the complexes are quenched compared with free proteins. Phosphorescence spectra with λexc ) 295 nm for free BSA, free HSA, and their complexes (Figures 9B and 10B) are similar to that observed with λexc ) 280 nm, the contribution of tyrosine phosphorescence appearing below 400 nm being diminished. Comparing the phosphorescence spectra of BSA-3HNA and HSA-3HNA complexes, we can attribute the shoulder at 405.6 nm in the BSA-3HNA spectrum to a particular Trp residue in BSA. Proteins containing more than one Trp residue generally exhibit a single (0,0) band in their phosphorescence spectra. The absence of multiple bands may be due to (i) energy transfer to another residue, (ii) interaction with neighboring residue like formation of charge-transfer complex, or (iii) electron transfer from the excited state. However, several proteins containing more than one Trp residue are known to exhibit multiple (0,0) bands.23-26,39,44-45 The observation of multiple (0,0) bands is usually possible when the Trp residues in a protein are in widely different environments and photoinduced energy transfer among the emitting Trp residues is prevented. Previous studies on lysozyme from bacteriophage T4 and its several mutants show a correlation between the position of the (0,0) band and the solvent exposure of the tryptophan residues.24-25 The three Trp residues 126, 138, and 158 exhibit phosphorescence (0,0) bands at 404.6, 413.6, and 407.7 nm respectively. The solvent accessible surface area (ASA) values are found to be least for Trp 138 which exhibits the most redshifted (0, 0) band and highest for Trp 126. The correlation of the position of the (0,0) band and the ASA is supported by crystal structure data46 which shows Trp 126 to be exposed, Trp 158 to be partially exposed, and Trp 138 to be buried. Similar correlation has been observed in the case of horse liver alcohol dehydrogenase,44 E. coli alkaline phosphatase,23,25 and human placental ribonuclease inhibitor.26 Apart from solvent exposure, rigidity of the environment and immediate local charges also control the position of the (0,0) band. Although BSA possesses two tryptophan residues, Trp 213 and Trp 134, the (0,0) band appears (at 412.6 nm) as a single peak, and the two tryptophans are not optically resolved. This (0,0) band is found to split upon binding to 3HNA producing two distinct (0,0) bands centered at 405.6 nm (very weak) and 412.8 nm (Table 7). The blue-shifted phosphorescence, typical of free Trp in a polar solvent, can be attributed to the lower polarizability of the environment and to the poor stabilization of the triplet state by rigid solvation geometry. The red-shifted (0,0) band on the other hand, is characteristic of a Trp residue
Energy Transfer Photophysics from Serum Albumins
J. Phys. Chem. B, Vol. 112, No. 11, 2008 3459
TABLE 6: Energy Transfer Efficiency and Rate Constant of Serum Albumin-3HNA Complexes at 298 K rate constant (kET) × 10 -7s-1
energy transfer efficiency (E)
from eq 8 experimental technique from eq 5 from eq 6 from eq 7 from phosphorescence quenching with λexc ) 295 nm
from eq 9
BSA3HNA
HSA3HNA
BSA3HNA
HSA3HNA
BSA3HNA
HSA3HNA
(kET)HSA/(kET)BSA experimental
0.32 0.34 0.26 0.24
0.47 0.49 0.36 0.47
9.0 9.7 7.0
20.6 22.3 13.1
8.9 9.5 7.2
20.4 21.3 15.7
2.3 2.3 2.2
(kET)HSA/(kET)BSA theoretical (with the help of eq 11) 6.5a
a Considering distances from the oxygen atom of the -OH group of 3HNA to the center of the indole moiety of Trp 134 in BSA and Trp 214 in HSA.
TABLE 7: Phosphorescence Data for Wild-Type Serum Albumins and Their Complexes with 3HNA in a 40% Ethylene Glycol Matrix at 77 K
system
λexc (nm)
free BSA (10 µM) BSA-3HNA (1:1)
280
free HSA (10 µM) HSA-3HNA (1:1) a
position of the phosphorescence (0,0) band (nm)a 412.6 405.6 (sh) 412.8 409.8 410.6
the complex is responsible for the red shift in the phosphorescence and the blue shift in the fluorescence.
width of the phosphorescence (0,0) band at half-maxima (cm-1) 330 300 330 330
Errors in the measurements are ) (0.2 nm.
located in a buried polarizable environment that stabilizes the triplet state more than the ground state.47 The results of docking studies indicate that the two tryptophans of wild type BSA have different ASAs, Trp 134 being less exposed to solvent as compared with Trp 213. This corroborates the optically detected magnetic resonance in zero magnetic field (ODMR) and LTP studies at 4.2 K of BSA and its two fragments (1-183 and 184-582) obtained by cyanogen bromide cleavage.48 These results indicate that Trp 134 is located inside the globular structure in a hydrophobic and relatively homogeneous environment, while Trp 213 is located in a partially buried inhomogeneous environment. The above results indicate that upon 3HNA binding at least one Trp residue of BSA is strongly perturbed and has environment no longer similar to that in the free BSA. An insight into the local environment of Trp of BSA-3HNA complex is obtained from the docking studies. Calculation of the solvent accessible surface area suggests that both Trp 134 and Trp 213 show no change in ASA (Table 5A). This indicates that possibly it is the Trp 134 which is located near the binding site of 3HNA obtained from docking studies becomes perturbed upon binding. Thus, the shoulder at 405.6 nm in the BSA-3HNA complex is attributed to Trp 134. The blue-shifted position of the band is indicative of a polar environment of Trp 134 upon binding to 3HNA. This finding is similar to that observed by the ODMR study in zero magnetic field of BSA-oleic acid complex, which showed that Trp 134 local environment is polar but not exposed to solvent.49 HSA contains a lone tryptophan residue at the position 214, which shows a phosphorescence (0,0) band at 409.8 nm. Upon binding to 3HNA, there is small but definitive red shift of the (0,0) band (Table 7). This is indicative of more hydrophobic environment experienced by Trp 214 upon binding. This is supported by the blue shift of the λmax of the fluorescence of Trp 214 upon binding as mentioned in section 3.2. Docking studies indicate that binding does not produce any change in the solvent accessible surface area of Trp 214 (Table 5B). This corroborates that more hydrophobic environment of Trp 214 in
Figure 9. (A) Phosphorescence spectra of (a) 10 µM BSA and (b) BSA-3HNA (1:1) complexes, λexc ) 280 nm. (The insets show the initial expanded portion for the BSA-3HNA complex.) (B) The phosphorescence spectra of (a) 10 µM BSA and (b) BSA-3HNA (1:1) complexes, λexc ) 295 nm. The excitation and emission band passes are 10 and 1.5 nm respectively in each case.
3460 J. Phys. Chem. B, Vol. 112, No. 11, 2008
Sardar et al. of ASA of the hydrophobic residues (Table 5A) near the binding site support the above conclusions. The ET efficiency calculated using the total area under the Trp phosphorescence spectra with λexc ) 295 nm (where Tyr is not excited) is found to be 0.47 for HSA and 0.24 for BSA. It is to be noted that although ET is at the singlet level ET values obtained from phosphorescence data at 77 K are in excellent agreement with those calculated from the fluorescence data at room temperature (Table 6) and thus support our earlier conclusions. Quantitative agreement should not be expected, since ET may differ for fluid and rigid solvent media. 4. Conclusion
Figure 10. (A) Phosphorescence spectra of (a) 10 µM HSA and (b) HSA-3HNA (1:1) complexes, λexc ) 280 nm. (The insets show the initial expanded portion for the HSA-3HNA complex.) (B) The Phosphorescence spectra of (a) 10 µM HSA and (b) HSA-3HNA (1: 1) complexes, λexc ) 295 nm. The excitation and emission band passes are 10 and 1.5 nm respectively in each case.
The band widths of the (0,0) bands (Table 7) indicate that the Trp environments in HSA and in its complex are similar in heterogeneity. It is interesting to note that in case of BSA we observed a slight decrease in the (0,0) bandwidth of the Trp 213 of BSA upon binding. This is indicative of a slightly more homogeneous environment of Trp 213 in the complex compared with that in free BSA. Docking studies showing substantial loss
The steady-state and time-resolved studies monitoring the Trp emission of the protein, the sensitized ESIPT emission of the tautomer of 3HNA at room temperature and 77 K and docking studies are employed to probe the interaction between 3HNA and the serum albumins. The studies reveal the following: (i) The number of the binding site is one for both HSA and BSA. (ii) Both the static and the dynamic quenching of Trp fluorescence occur in the presence of 3HNA. The analysis of the steady-state and the time-resolved quenching provides the binding constant of the order of ≈105. (iii) The ESIPT emission of 3HNA is sensitized in both complexes of HSA and BSA. The nonradiative singlet (Trp singlet) to the ESIPT singlet state energy transfer is found to be greater in the case of HSA. It is shown that Trp 134 rather than Trp 213 is involved in the ET process in the complex of BSA from the docking studies. (iv) The ET rate constants were determined by time-resolved technique and the experimental ratio of the ET rate constants [(kET)HSA-3HNA/(kET)BSA-3HNA] agree well with the ratio calculated using the distance and the orientation of the donoracceptor obtained from the docking results. (v) Phosphorescence spectra suggest that the local environment of Trp 134 rather than of Trp 213 in BSA is perturbed due to interaction with 3HNA, indicating its proximity to the binding site. The perturbation leads to optical resolution of Trp 134 phosphorescence in the complex. The quenching of phosphorescence in the complexes agrees well with the results of the ET efficiency. (vi) Docking studies clearly indicate H bonding of the probe with Tyr 149 in BSA and with Tyr 150, His 242, and Arg 257 in HSA. The specific H bonding and substantial loss of ASA for several hydrophobic residues near the binding site support the observation of larger rotational correlation time θc, observed for 3HNA. The motional restriction of 3HNA in the binding site has been found to be responsible for greater quantum yield of the ESIPT emission of 3HNA compared with that in aqueous buffer medium. The larger θc (≈50 ns) for Trp residues/residue observed for the complexes of BSA and HSA compared with that in the free proteins also confirm that the binding of 3HNA imposes motional restriction of the Trp residue/residues of the proteins in the complexes. Acknowledgment. S.G. gratefully acknowledges DST (SP/ S1/PC-34/2002) and CSIR (01(2142)/07/EMR-II) for financially supporting this work. P.S.S. thanks DST for a SRF fellowship (No. SR/S5/NM-14/2003). S.S. thanks CSIR for a SRF fellowship (No. 8/155/(12)/2002-EMR-I). S.M. acknowledges DST for a SRF fellowship (SP/S1/PC-34/2002). We are grateful to the reviewers for their helpful suggestions.
Energy Transfer Photophysics from Serum Albumins References and Notes (1) Sytnik, A.; Gormin, D.; Kasha, M. Proc. Natl. Acad. Sci. U.S.A. 1994, 91, 11968-11972. (2) Sytnik, A.; Kasha, M. Proc. Natl. Acad. Sci. U.S.A. 1994, 91, 86278630. (3) Mishra, H.; Joshi, H. C.; Tripathi, H. B.; Maheshwary, S.; Sathyamurthy, N.; Panda, M.; Chandrasekhar, J. J. Photochem. Photobiol. A: Chem. 2001, 139, 23-36. (4) Samanta, S.; Saha Sardar, P.; Maity, S. S.; Pal, A.; Ghosh, S. communicated to Chemical Physics. (5) Brand, L., Johnson, M. L., Eds. Fluorescence Spectroscopy, Methods in Enzymology, Vol. 278; Academic Press: New York. (6) Lakowicz, J. R. Principles of Fluorescence Spectroscopy; Kluwer/ Plenum Press: New York, 1999; Chapter 17. (7) Lakowicz, J. R.; Cherek, H. J. Biol. Chem. 1981, 256, 6348-6353. (8) Royer, C. A.; Gardner, J. A.; Beechem, J. M.; Brochon, J. C.; Matthews, K. S. Biophys. J. 1990, 58, 363-378. (9) Lakowicz, J. R.; Balter, A. Photochem. Photobiol. 1982, 36, 125132. (10) Lakowicz, J. R.; Balter, A. Biophys. Chem. 1982, 15, 353-360. (11) Vivian, J. T.; Callis, P. R. Biophys. J. 2001, 80, 2093-2109. (12) Liu, T.; Callis, P. R.; Hesp, B. H.; de Groot, M.; Buma, W. J.; Broos, J. J. Am. Chem. Soc. 2005, 127, 4104-4113. (13) Kurz, L. C.; Fite, B.; Jean, J.; Park, J.; Erpelding, T.; Callis, P. R. Biochemistry 2005, 44, 1394-1413. (14) Callis, P. R.; Liu, T. J. Phys. Chem. B 2004, 108, 4248-4259. (15) Hahn, D. K.; Callis, P. R. J. Phys. Chem. A 1997, 101, 26862691. (16) Callis, P. R.; Vivian, J. T. Chem. Phys. Lett. 2003, 369, 409-414. (17) Willets, K. A.; Callis, P. R.; Moerner, W. E. J. Phys. Chem. B 2004, 108, 10465-10473. (18) Eftink, M. R. Fluorescence Spectroscopy, Methods in Enzymology, Vol. 278; Brand, L., Johnson, m. L., Eds.; Academic Press: New York; pp 221-257. (19) Loewenthal, R.; Sancho, J.; Fersht, A. R. Biochemistry 1991, 30, 6775-6779. (20) Harris, D. L.; Hudson, B. S. Biochemistry 1990, 29, 5276-5285. (21) Maki, A. H. In Biological Magnetic Resonance; Berliner, L. J., Reuben, J., Eds.; Plenum Press: New York, 1984; Vol. 6, Chapter 17, pp 187-294. (22) Ghosh, S.; Zang, L. H.; Maki, A. H. J. Chem. Phys. 1988, 88, 2769-2775. (23) Ghosh, S.; Misra, A.; Ozarowski, A.; Maki, A. H. J. Phys. Chem. B 2003, 107, 11520-11526. (24) Zang, L. H.; Ghosh, S.; Maki, A. H. Biochemistry 1988, 27, 78167820. (25) Ghosh, S.; Misra, A.; Ozarowski, A.; Stuart, C.; Maki, A. H. Biochemistry 2001, 40, 15024-15030. (26) Saha Sardar, P.; Maity, S. S.; Ghosh, S.; Chatterjee, J.; Maiti, T. K.; Dasgupta, S. J. Phys. Chem. B 2006, 110, 21349-21356.
J. Phys. Chem. B, Vol. 112, No. 11, 2008 3461 (27) Maiti, T. K.; Ghosh, K. S.; Dasgupta, S. Proteins Struct. Funct. Bioinf. 2006, 64, 355-362. (28) He, X. M.; Carter, D. C. Nature 1992, 358, 209-215. (29) Brown, J. R. In Albumin: Structure, Function, and Uses; Rosenoer, V. M., Oratz, M., Rothschild, M. A., Eds.; Pergamon Press: Oxford, 1977; pp 27-51. (30) Moriyama, Y.; Ohta, D.; Hachiya, K.; Mitsui, Y.; Takeda, K. J. Protein Chem. 1996, 15, 265-272. (31) FELIX 32, Operation Manual, Version 1.1, 2003, Photon Technology International, Inc, 1009 Lenox Drive, Suite 104, Lawrenceville, NJ 08645. (32) Berman, H. M.; Westbrook, J.; Feng, Z.; Gilliland, G.; Bhat, T. N.; Weissig, H.; Shindyalov, I. N. Bourne P. E. Nucleic Acids Res. 2000, 28, 235-242. (33) Karplus, K.; Katzman, S.; Shackleford, G.; Koeva, M.; Draper, J.; Barnes, B.; Soriano, M.; Hughey, R. Proteins: Struct. Funct. Bioinf. 2005, 61 (S7), 135-142. (34) Rarey, M.; Kramer, B.; Lengauer, T.; Klebe, G. J. Mol. Biol. 1996, 261, 470-489. (35) DeLano, W. L. The PyMOL Molecular Graphics System; DeLano Scientific: San Carlos, CA, 2004, http://pymol.sourceforge.net/. (36) Hubbard, S. J.; Thornton, J. M. ‘NACCESS’. Computer Program; Department of Biochemistry and Molecular Biology, University College London, 1993. (37) Sytnik, A.; Litvinyuk, I. Proc. Natl. Acad. Sci, U.S.A. 1996, 93, 12959-12963. (38) Beechem, J. M.; Brand, L. Ann. ReV. Biochem. 1985, 54, 43-71. (39) Eftink, M. R.; Ramsay, G. D.; Burns, L.; Maki, A. H.; Mann, C. J.; Matthews, C. R.; Ghiron, C. A. Biochemistry 1993, 32, 9189-9198. (40) Yguerabide, J.; Epstein, H. F.; Stryer, L. J. Mol. Biol. 1970, 51, 573-590. (41) Helms, M. K.; Petersen, C. E.; Bhagavan, N. V.; Jameson, D. M. FEBS Lett. 1997, 408, 67-70. (42) (a) Fo¨rster, T. Zwischenmolekulare energiewanderung und fluoreszenz. Ann. Phys. (Leipzig) 1948, 2, 55-75. (b) Fo¨rster, T. Discuss. Faraday Soc. 1959, 27, 7-17. (43) Philips, L. A.; Levy, D. H. J. Chem. Phys. 1986, 85, 1327-1332. (44) Eftink, M. R. In AdVances in Biophysical Chemistry; Bush, C. A., Ed.; JAI Press: Greenwich, CT, 1992; Vol. 2, pp 81-114. (45) Zang, L. H.; Ghosh, S.; Maki, A. H. Biochemistry 1989, 28, 22452251. (46) Remington, S. J.; Anderson, W. F.; Owen, J.; Ten Eyck, L. F.; Grainger, C. T.; Matthews, B. W. J. Mol. Biol. 1978, 118, 81-98. (47) Ozarowski, A.; Barry, J. K.; Matthews, K. S.; Maki, A. H. Biochemistry 1999, 38, 6715-6722. (48) Mao, S.-Y.; Maki, A. H. Biochemistry 1987, 26, 3106-3114. (49) Mao, S.-Y.; Maki, A. H. Biochemistry 1987, 26, 3576-3582.