J. Phys. Chem. B 2005, 109, 3319-3323
3319
Solvation Dynamics of DCM in a DPPC Vesicle Entrapped in a Sodium Silicate Derived Sol-Gel Matrix Pratik Sen,† Saptarshi Mukherjee,† Amitava Patra,*,‡ and Kankan Bhattacharyya*,† Physical Chemistry Department, Indian Association for the CultiVation of Science, JadaVpur, Kolkata 700 032, India, and Sol-Gel DiVision, Central Glass & Ceramic Research Institute, JadaVpur, Kolkata 700 032, India ReceiVed: October 1, 2004; In Final Form: NoVember 24, 2004
Solvation dynamics of 4-(dicyanomethylene)-2-methyl-6(p-dimethylaminostyryl) 4H-pyran (DCM) has been studied in a dipalmitoyl-phosphatidylcholine (DPPC) vesicle entrapped in a sodium silicate derived sol-gel glass. Solvation dynamics in DPPC in a sol-gel glass is described by two components of 350 ( 50 ps (50%) and 2300 ( 200 ps (50%) with a total dynamic Stokes shift of 1300 cm-1. The fast component (350 ps) is similar to the fast component in a DPPC vesicle in bulk water (320 ( 50 ps). This component may be ascribed to the dynamics of the water molecules inside the water pool of the vesicle. However, the slow component (2300 ( 200 ps) is about 2.5 times slower compared to the slow component of solvation dynamics of DCM in a DPPC vesicle in bulk solvent (900 ( 100 ps). The anisotropy decay of DCM in a DPPC vesicle both in sol-gel glass and in bulk water exhibits a very fast initial decay with a large residual anisotropy, which does not decay in ∼10 ns. The time scale of anisotropy decay is very different from that of solvation dynamics.
1. Introduction In recent years, numerous research groups have demonstrated successful incorporation of a wide range of biomolecules into sol-gel derived porous silica matrixes.1-5,7d,e In a porous silica matrix biological molecules remain in their active form for a very long time. Sol-gel processes carried out in aqueous medium at room temperature are well suited for the entrapment of sensitive molecules in porous matrixes. The sol-gel process usually involves hydrolysis of an appropriate alkoxide (e.g., tetraethyl orthosilicate, TEOS, or tetramethyl orthosilicate, TMOS), which leads to the formation of a sol containing silanol groups (Si-OH). Subsequently they condense to form a gel having siloxane (Si-O-Si) groups. The use of TMOS or TEOS as starting material leads to generation of methanol or ethanol, which in large quantities is detrimental to biological macromolecules.6 The synthesis of a porous silica matrix from sodium silicate has several advantages over hydrolysis of orthosilicates. First, the sodium silicate system does not involve generation of alcohol by hydrolysis. Second, for a sodium silicate derived sol-gel glass, the encapsulation may be carried out at neutral pH, which is very important to preserve the biological activity of a protein.6 In this work, we have studied a lipid vesicle encapsulated in a sodium silicate derived porous silica matrix. In bulk water, solvation dynamics is extremely fast with a major component in the 0.1 ps time scale and a minor component of ∼1 ps.9 However, water molecules confined in many organized and constrained media exhibit a 100-1000 ps component, which is slower by 2-3 orders of magnitude.7,8,10-15 Such a slow component has been reported in many biological * To whom correspondence should be addressed. (A.P.) E-mail: apatra@ cgcri.res.in. (K.B.) E-mail:
[email protected]. Fax: (91)-332473-2805. † Indian Association for the Cultivation of Science. ‡ Central Glass & Ceramic Research Institute.
and complex systems, e.g., sol-gel glasses,7,8 proteins,11 DNA,12 reverse micelles,10,13a,b nanoparticles,13c and lipid vesicles.14 The slow component of relaxation in confined systems is also detected in NMR15a,b and dielectric relaxation15c studies. The experimental studies have inspired phenomenological models16 and large-scale computer simulations.17,18 We have earlier reported that the average solvation time of water molecules inside the pores of a TEOS sol-gel matrix is 150 ps, which is markedly slower than that in bulk water.7a Baumann et al. studied solvation dynamics of ethanol using Nile blue and coumarin 153 (C153) as probes in sol-gel matrixes of different pore sizes. They found that for Nile blue7b the average solvation time of ethanol increases from 18.6 ps in 75 Å pores to 39 ps in 50 Å pores, while for C1537c the average solvation time increases from 28.6 ps in 50 Å pores to 36.9 ps in 25 Å pores. Using the optical Kerr effect Fourkas and coworkers showed that various liquids confined in a sol-gel glass display a major bulk-like component and an additional component that is nearly 4 times slower.8 Bright and co-workers studied solvation dynamics of acrylodan-labeled human serum albumin (HSA) in a sol-gel matrix and detected a nanosecond component.4a In a very recent study, we reported that the solvation time of coumarin 480 in a DMPC-entrapped sol-gel matrix is about 14 times faster than that in a DMPC liposome in bulk water, as liposomes are ruptured inside the TEOS sol-gel.7d Brennan and co-workers5b reported that the liposomes do not exhibit phase transitions when entrapped in TEOS-derived glasses. This indicates that the DMPC liposomes are ruptured during encapsulation. However, DPPC displays phase transitions when entrapped in a sol-gel glass derived from sodium silicate.5b Thus the DPPC liposomes remain structurally intact when entrapped inside the inorganic sol-gel matrix. In this present study, we wish to study the effect of entrapment of DPPC
10.1021/jp0455327 CCC: $30.25 © 2005 American Chemical Society Published on Web 02/08/2005
3320 J. Phys. Chem. B, Vol. 109, No. 8, 2005
Sen et al.
liposomes on the solvation dynamics of the probe 4-(dicyanomethylene)-2-methyl-6(p-(dimethylamino)styryl) 4H-pyran (DCM). 2. Experimental Section 4-(Dicyanomethylene)-2-methyl-6(p-dimethylaminostyryl) 4Hpyran (DCM, laser grade, Exciton), sodium silicate (Aldrich), DOWEX cation-exchange resin (Aldrich), and dipalmitoylphosphatidylcholine (DPPC, Sigma) were used as received. The sodium silicate derived materials were prepared as follows. To a mixture of 1.3 g of sodium silicate and 3.95 g of double distilled water was added 1.74 g of DOWEX cation-exchange resin in small installments spanning over 1 h with constant stirring. It was left undisturbed for another 1 h, which resulted in the formation of a clear transparent sol, the pH being about 4. The unilamellar vesicle was prepared separately by the ethanol injection method. Briefly, 2 mg of DPPC was first dissolved in 50 µL of an ethanol solution of DCM. Then the whole solution was rapidly injected using a micro-liter syringe into 2 mL of 100 mM tris-HCl buffer of pH 7.4 at 45 °C. The solution was kept at 45 °C (i.e., above the gel transition temperature of DPPC, 41 °C) for 1 h. One milliliter of sodium silicate sol and 1 mL of a solution of DPPC-tris buffer containing DCM were mixed in a quartz tube and within 1 h a semitransparent gel resulted. The final concentration of the DPPC is 1.3 mM. The fluorescence experiments were done with a substantially aged semitransparent sol-gel composite after 15 days. The steady-state absorption and emission spectra were recorded in a Shimadzu UV-2401 spectrophotometer and a Perkin-Elmer 44B spectrofluorimeter, respectively. For lifetime measurements, the samples were excited at 373 nm using a picosecond diode laser (IBH Nanoled-07) at a repetition rate 900 kHz. The fluorescence decays were collected at a magic angle polarization using an analyzer and a Hamamatsu MCP photomultiplier (C487802). The time correlated single photon counting (TCSPC) setup consists of an Ortec 9327 CFD and a Tennelec TC 863 TAC. The data are collected with a PCA3 card (Oxford) as a multichannel analyzer. The typical fwhm of the system response is about 80 ps. To study fluorescence anisotropy decay, the analyzer was rotated at regular intervals to get perpendicular (I⊥) and parallel (I|) components. Then the anisotropy function, r(t), was calculated using the formula
r(t) )
I(t) - GI|(t) I(t) + 2GI|(t)
(1)
The G value of our setup is 1.8. 3. Results and Discussion 3.1. Steady-State Results. The emission maximum of DCM exhibits a red shift with increase in solvent polarity, from 530 nm in n-heptane to 620 nm in methanol. In a DPPC liposome DCM exhibits an emission maximum at 615 nm (for λex ) 375 nm). The emission maximum of DCM in a DPPC vesicle is close to that of DCM in methanol (620 nm). DCM is insoluble in bulk water and, hence, in a DPPC vesicle in water, DCM does not reside in bulk water. The emission maximum of DCM in a DPPC vesicle is very different from that in n-heptane (530 nm). Thus in a DPPC vesicle in water DCM does not stay in the hydrocarbon-like membrane region. This suggests that DCM molecules may stay in a polar region near the membrane surface and inside the water pool of the DPPC vesicle.
Figure 1. Steady-state emission spectra of DCM (λex ) 375 nm) in a DPPC-entrapped sodium silicate derived sol-gel matrix.
Figure 2. Fluorescence decays of DCM in a DPPC vesicle in bulk water containing tris buffer at (i) 525 nm, (ii) 585 nm, and (iii) 690 nm.
In the absence of the liposomes, DCM cannot be encapsulated inside the inorganic sol-gel. However, DCM is easily incorporated in the sol-gel glass in the presence of DPPC liposomes. In a DPPC liposome doped sodium silicate sol-gel matrix, the emission maximum of DCM is found to be at 595 nm (Figure 1). This is 20 nm blue shifted compared to that of DCM in a DPPC liposome in bulk water. The blue shift in emission maximum of DCM indicates that the microenvironment inside the DPPC-doped sol-gel matrix is less polar than the liposome alone. Eisenthal and co-workers19 reported that at pH 7 the silica surface of the sol-gel matrix comprises Si-OH and Si-O- moieties. Since DPPC is a cationic lipid, it binds strongly to Si-OH and Si-O-. The thin structured water layer at the silica-DPPC interface is severely constrained. This may be responsible for the reduction in polarity and the marked blue shift of the emission maximum of DCM. 3.2. Time-Resolved Studies: Solvation Dynamics. 3.2.1. SolVation Dynamics of DCM in DPPC Vesicles in Bulk Water. Solvation dynamics of DCM in DPPC liposomes in bulk water is manifested in the wavelength dependence of emission decays of the probe, DCM. At the blue end (525 nm) the fluorescence decay is biexponential with two decay components of 170 ps (78%) and 1.15 ns (22%), while at the red end (690 nm), the decay of the time constant 2.30 ns is preceded by a distinct rise with a time constant of 180 ps. A few representative decay profiles of DCM in a DPPC vesicle in bulk water are shown in Figure 2. Following the procedure given by Fleming and Maroncelli,20 the time-resolved emission spectra (TRES) were constructed using the parameters of best fit to the fluorescence decays and the steady-state emission spectrum. The TRES show a time-dependent Stokes shift (Figure 3). The solvation dynamics
Solvation Dynamics of DCM
J. Phys. Chem. B, Vol. 109, No. 8, 2005 3321
Figure 5. Fluorescence decays of DCM in a DPPC-entrapped sodium silicate derived sol-gel matrix at (i) 525 nm, (ii) 585 nm, and (iii) 690 nm. Figure 3. Time-resolved emission spectra of DCM in DPPC in bulk water at 0 ps (9), 200 ps (O), 600 ps (2), and 4000 ps (3).
Figure 4. Decay of response function, C(t), of DCM in a DPPC vesicle in bulk water (O) and in a DPPC-entrapped sodium silicate derived sol-gel matrix (9). The points denote the actual values of C(t), and the solid line denotes the best fit to a biexponential decay.
TABLE 1: Decay Parameters of C(t) for DCM in Different Systems system
∆νb (cm-1)
a1a
τ1b (ps)
a2a
τ2b (ps)
DPPC vesicle DPPC entrapped in sol-gel matrix
850 1300
0.45 0.50
320 350
0.55 0.50
900 2300
a
(10%. b (10%.
is described by the decay of the solvent correlation function C(t), defined as
C(t) )
ν(t) - ν(∞) ν(0) - ν(∞)
(2)
where ν(0), ν(t), and ν(∞) are the peak frequencies at time 0, t, and ∞, respectively. The decay of C(t) is shown in Figure 4, and the decay parameters are summarized in Table 1. The decay of C(t) for DCM inside DPPC liposomes in bulk water is found to be biexponential, with one component of 320 ps (45%) and another 900 ps (55%). The dynamic Stokes shift of DCM
observed in the case of DPPC vesicles is 850 ( 100 cm-1. It is evident that the time constants of decay of C(t) of DCM in DPPC vesicles are similar to those in DMPC (230 and 1600 ps).14a,21 The two time constants of decay of C(t) may arise from different locations of the probe (DCM) within the DPPC vesicle. It may be recalled that Chattopadhyay and Mukherjee studied location dependence of solvent relaxation in a lipid using red edge excitation shift (REES) and fluorescence anisotropy using two anthroyloxy stearic acid probes (2-AS and 12-AS).22 They observed appreciable REES for a probe located at the lipidwater interface. However, for the probe 12-AS buried in the hydrophobic membrane there is no REES. Thus we ascribe the two time constants (320 and 900 ps) of solvation dynamics in a DPPC vesicle in bulk water to different microenvironments within the DPPC liposomes. The 900 ps component may arise from the highly restricted water molecules in the vicinity of the polar headgroup of DPPC. The fast component (320 ps) may originate from the dynamics of those water molecules that are residing in the water pool of the liposomes. 3.2.2. SolVation Dynamics of DCM in DPPC-Doped SodiumSilicate Sol-Gel Glass. In a DPPC-doped sodium silicate solgel matrix, at the blue end (525 nm) the fluorescence decay of DCM is biexponential with two decay components of 410 ps (63%) and 2.30 ns (37%), while at the red end (690 nm) the decay of time constant of 2.85 ns is preceded by a distinct rise with a time constant of 230 ps. A few decay profiles for this system are shown in Figure 5. The TRES for DCM in DPPC in sol-gel glass are given in Figure 6. For DPPC liposomes entrapped in a sodium silicate sol-gel glass, C(t) is found to be biexponential, with one component of 350 ps (50%) and another 2300 ps (50%) (Figure 4, Table 1). The dynamic Stokes shift is observed to be 1300 ( 100 cm-1. Evidently, the decay parameters of C(t) for DCM in a DPPC vesicle entrapped in sol-gel glass are very different than that in a DPPC vesicle in bulk water. The slow component of solvation dynamics of DCM bound to a DPPC vesicle (900 ps) in bulk water is slowed by a factor of 2.5 when a DPPC vesicle is encapsulated in the sodium silicate derived sol-gel matrix (2300 ps). Evidently, in the DPPC-doped sol-gel matrix, a fraction of the probe (DCM) molecules experiences a microenvironment very different from that inside the DPPC vesicles in bulk water. The very long component of 2300 ps may originate from the dynamics of the water molecules squeezed between the DPPC vesicle and the silica surface. The fast component of the solvation dynamics
3322 J. Phys. Chem. B, Vol. 109, No. 8, 2005
Sen et al.
Figure 6. Time-resolved emission spectra of DCM in a DPPCentrapped sodium silicate derived sol-gel matrix at 0 ps (9), 50 ps (O), 150 ps (2), and 7000 ps (3).
of DCM in a DPPC entrapped sol-gel matrix (350 ps) is very close to the fast component of solvation dynamics (320 ps) of DCM in DPPC vesicles alone and, hence, may be attributed to those in the water pool of the vesicle. There could be many reasons for the slowing down of solvation dynamics in liposome-doped glasses. The water molecules at the silica-lipid interface remain hydrogen bonded to both silica and lipid molecules and are thus severely constrained. This may cause slowing down of the solvation dynamics in the liposome-doped glasses. The other possibility is the hindered exchange between bound and free water molecules.16 If this is assumed, then according to the dynamic exchange model proposed by Nandi and Bagchi,16 the slow component of the solvation dynamics originates from the interconversion of the bound state to the free state of the water molecules. The magnitude of the slow component of solvent relaxation depends on the energy difference (∆G°) between the bound and the free state of water molecules. In the limit of very high ∆G°, the slow component of solvation (τslow) is given by16
τslow ≈ kbf-1
(3)
where, kbf is the rate constant for bound-to-free interconversion,
kbf )
( ) (
)
kBT -(∆G0 + ∆G*) exp h RT
(4)
where ∆G* (∼1.5 kBT, as used by Nandi and Bagchi16) is the activation energy for the conversion of free-to-bound water molecules. From eqs 3 and 4 and using the average solvation times one may calculate the energy difference (∆G°) between the bound and free water molecules. In the case of DPPC vesicles present in bulk water there is only one kind of bound water: bound to lipid. Thus, for DPPC in bulk water ∆G° ) -4.10 ( 0.05 kcal mol-1 (using τslow ) 900 ps). Obviously, straightforward application of the Nandi-Bagchi model is difficult in the case of vesicles in a sol-gel glass. In this system there may be several kinds of bound water molecules, e.g., those bound either or both to the silica surface and lipid. 3.3. Fluorescence Anisotropy Decay of DCM in DPPC Vesicles in Bulk Water and Trapped in Sodium Silicate Derived Sol-Gel Glass. The fluorescence anisotropy decay of
Figure 7. Fluorescence anisotropy decay of DCM in a DPPC vesicle entrapped sodium silicate derived sol-gel matrix.
SCHEME 1: Structure of DCM
DCM in DPPC vesicles in bulk water and entrapped in sodium silicate derived sol-gel matrixes displays a very fast initial decay component, which is faster than the time resolution of our setup (
