Comparison between cucurbiturils and β-cyclodextrin interactions with cholesterol molecules present in Langmuir monolayers used as a biomembrane model

Colloids and Surfaces B: Biointerfaces 111 (2013) 398–406

Contents lists available at SciVerse ScienceDirect

Colloids and Surfaces B: Biointerfaces journal homepage: www.elsevier.com/locate/colsurfb

Comparison between cucurbiturils and ␤-cyclodextrin interactions with cholesterol molecules present in Langmuir monolayers used as a biomembrane model Camila Bussola Tovani a , João Francisco Ventrici de Souza a,1 , Thiago de Souza Cavallini a , Grégoire Jean-Franc¸ois Demets a , Amando Ito b , Marina Berardi Barioni b , Wallance Moreira Pazin b , Maria Elisabete Darbello Zaniquelli a,∗ a b

Departamento de Química, Faculdade de Filosofia, Ciências e Letras de Ribeirão Preto, Universidade de São Paulo, 14040901 Ribeirão Preto, SP, Brazil Departamento de Física, Faculdade de Filosofia, Ciências e Letras de Ribeirão Preto, Universidade de São Paulo, 14040901 Ribeirão Preto, SP, Brazil

a r t i c l e

i n f o

Article history: Received 13 November 2012 Received in revised form 4 May 2013 Accepted 6 May 2013 Available online xxx Keywords: Host–guest complex at interfaces Cholesterol Phospholipids Biomembrane models Cucurbiturils

a b s t r a c t Specific surface techniques can probe the interaction of cholesterol (Chol) with substances that are able to host and/or sequester this biomolecule, provided that the additives are properly assembled at the interface. Reports on inclusion complexes of Chol with ␤-cyclodextrins exist in the literature. Here we compare the interaction of ␤-cyclodextrin and cucurbiturils with Chol present in Langmuir phospholipid (dipalmitoylphosphatidylcholine, DPPC) monolayers, used as a biomembrane model. Cucurbiturils, CB[n], comprise macrocyclic host molecules consisting of n glycoluril units. Classic surface pressure curves, dilatational surface viscoelasticity measurements, and fluorescence emission spectra and images obtained by time-resolved fluorescence of the corresponding Langmuir–Blodgett films have shown that homologues with 5 and 6 glycoluril units, CB[5] and CB[6], do not form inclusion complexes. Higher-order homologues, such as CB[7], are likely to complex with Chol with changes in the minimum molecular areas recorded for DPPC/Chol monolayers, the fluorescence decay lifetimes, and the dilatational surface viscosities of the monolayers generated in the presence of these molecules. Moreover, we proof the removal of cholesterol from the biomimetic interface in the presence of CB[7] by means of fluorescence spectra from the subphase support of monolayers containing fluorescent-labeled Chol. © 2013 Elsevier B.V. All rights reserved.

1. Introduction Elasticity and viscoelasticity influence various processes occurring in cells and biomembrane models, e.g. protein conformation, enzyme activity, receptor functions, and pathological processes evolving malignant cells [1–4], among others. Chol is an essential lipid, that has different functions in the human body–it affects the surface elasticity, permeability, phase transitions and morphologies of cell membranes and biomembrane systems [5–7]. Controling Chol in the human body cholesterol is a complex task. For example, individuals with acute respiratory distress syndrome, caused by accumulation of particulate material accumulation in the lungs, present 5 up to 40% (w/w) increased Chol levels in relation to total phospholipids present in the bronchoalveolar lavage fluid [8]. On the other hand people living in an environment with polluted

∗ Corresponding author. E-mail address: [email protected] (M.E.D. Zaniquelli). 1 Present address: Department of Chemical Engineering and Materials Science, University of California, Davis, CA, United States. 0927-7765/$ – see front matter © 2013 Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.colsurfb.2013.05.006

(particulate-rich) atmosphere may develop coronary disease due to Chol deposition inside the arteries, even if they consume low amounts of fat. These examples demonstrate how difficult is to control Chol and how hazardous to the human body elevated Chol levels can be. Therefore, it is relevant to seek for substances that can potentially sequester Chol which should mitigate its effect on the surface elasticity of biomembrane model systems. Gunasekara et al. [9] have recently demonstrated that a modified ␤-cyclodextrin, CD, a glucose condensed macrocyclic with a hydrophobic cavity, can restore some functions as well as the structure of the surfactant pulmonary system. Since the 1970s, cyclodextrins have found several applications [10], including Chol extraction [11]. Ref. [12] brings a recent review of cyclodextrins. Cucurbiturils, (CB[n]), constitute another type of macrocyclic host molecules originated from the acid-catalyzed condensation of n glycoluril units with formaldehyde [13,14]. Freeman et al. [15] determined the structure of CB[6]. Depending on the conditions employed during the synthesis, it is possible to obtain different ratios of several homologues, from the pentamer, CB[5], to the decamer, CB[10]. CB[6] usually is the easiest to obtain, with the highest yield. This reveals an analogy with ␣−, ␤− and ␥ − CDs,

C.B. Tovani et al. / Colloids and Surfaces B: Biointerfaces 111 (2013) 398–406

399

Fig. 1. Shape and dimensions of additives used in this work [35,36]. Lower right corner show a top illustration of CB[7] hosting the hydrophobic portion of cholesterol molecule. Dimensions of Chol molecule are included for comparison.

which correspond to hexamers, heptamers and octomers of dglucose. The synthesis of cucurbiturils normally produce a mixture of macrocycles with different molecular weights. The distinct solubility of the products allow for their separation. Logically, the size of the cavity of these macrocycles depends on the number of glycoruril units (see Fig. 1); for example, CB[6] may include hydrophobic moieties of different compounds containing between 6 and 7 aliphatic carbon chains, whereas the carbonyl oxygens present in the portals may complex with cationic species or metal cations [16]. Their toxicity of CB[n] is low [17], make them potentially applicable as additives for medical and pharmaceutical purpose; e.g., to encapsulate molecules for drug delivery or to sequester Chol.

In this paper, we investigate the interaction of cucurbit[n]urils with Chol present in biomembranes. For this study we have chosen a simple biomembrane model system consisting of dipalmitoylphosphatidylcholine (DPPC)-Chol Langmuir monolayers, because the kinetics for Chol extraction from ternary lipid mixtures evolving unsaturated lipids is much more complex [18]. We measure the surface dilatational viscoelasticities, and constructed equilibrium surface pressure curves at room temperature, 21 ◦ C, and at the human body temperature, 36.5 ◦ C. We also employed fluorescence time-resolved microscopy using labeled Chol, to follow the presence and distribution of this biomolecule along the interface of the monolayer. Additionally, we conducted

400

C.B. Tovani et al. / Colloids and Surfaces B: Biointerfaces 111 (2013) 398–406

parallel studies of the interaction of ␤-cyclodextrin with Chol, to compare the complexation of Chol with CB[n] and CD in the biomembrane model. 2. Materials and methods 2.1. Materials Dipalmitoylphosphatidycholine (DPPC), cholesterol (Chol), and ␤-cyclodextrin (␤-CD) were purchased from Sigma and used without further purification. Chloroform (J.T. Baker) was HPLC grade. All the aqueous solutions were prepared using dust free Milli-Q® water (surface tension of 72.8 mN m−1 and resistivity of 18.2 M cm). 2.2. Cucurbiturils preparation Cucurbit[n]urils were prepared and purified using the methodology described elsewhere [14]. At the end of the condensation reaction, all the CB[n] (solid 1) is precipitated from the solution using pure acetone. Solid 1 is filtered off and treated with boiling water in order to dissolve almost all the CB[5], CB[7] and CB[8] (a minor product), leaving almost pure CB[6] which is separated by filtration. Briefly, after the separation of CB[6] the remaining solutions, from several synthesis, were mixed in order to have enough material to be extracted from them. This solution was concentrated by heating up to ¼ of volume and it is filtered to remove any quantity of CB[6] or CB[8]. Acetone is added once again to the supernatant containing mainly CB[5], and CB[7] doubling its volume. Adding acetone to the mixture precipitates most of the CB[5] and some CB[7], (CB[5]-richer sample), leaving a liquid that containing mainly CB[7] and traces of CB[5] (CB[7]-richer sample). Three kinds of samples were employed in the present study: CB[6], with 6 glycoluril units, a mixture called CB[5] corresponding to CB[5]-richer sample, which contains mostly CB[5] and CB[7] in smaller amounts, and also another mixture called CB[7], corresponding to the CB[7]richer sample, which contains mostly CB[7] and smaller amounts of CB[5]. ˚ The cavity sizes for CB[5], CB[6] and CB[7] are 4.4, 5.8 and 7.3 A, respectively, and it is known the diameter of the interior is larger than the diameter at the portals [19]. Due to its low solubility and high reaction yield, CB[6] sample is the most easily obtained. The other two CB[5]- and CB[7]-richer samples were confirmed by means of mass spectrometry and NMR of 13 C and 1 H. As we are mostly interested in the formation of inclusion complexes with cholesterol, the experiments with CB[5], with smaller cavity, may disclose other kind of CB[n < 6]-Chol interaction. 2.3. Langmuir monolayers preparation DPPC and DPPC-Chol (80:20 w/w) Langmuir monolayers were used as membrane models. The phospholipid solutions were prepared in chloroform and spread on a 216 cm2 (Insight-Brazil) Langmuir trough. After 15 min of solvent evaporation the monolayers formed at the air/liquid interface were compressed through the moving barrier at a rate of 0.041 A˚ 2 molecule−1 s−1 to record the surface pressure-area curves. The subphase support for the monolayers was pure Milli-QTM water (surface tension 72.6 mN m−1 at 21 ◦ C and 70.2 at 36.5 ◦ C, pH 5.5 after air exposure), or aqueous solutions of CB[5], CB[6], CB[7], or ␤-CD at concentrations of 0.018 mmol L−1 or 4 mmol L−1 , appropriately labeled on the results. The temperature was controlled by means of a circulating water bath operated at 21 or 36.5 ± 0.5 ◦ C. Moreover, a smaller truncatecone-shaped trough with a constant perimeter (surface area of 12.5 cm2 and subphase volume of 10 mL), under mild stirring, was used to investigate whether Chol solubilized into the subphase. In

these experiments, only CB[7] was employed, using the same concentration as the one used to obtain the surface pressure-molecular area, (␲-A), curves. The contact time between the subphase and the monolayer was 16 h. After that, the monolayer was sucked out, and a 2 mL sample of the subphase was withdrawn, to record the emission spectra. 2.4. Emission spectra The fluorescence spectra were recorded on an F-7000 FL spectrofluorometer, at an excitation wavelength of 465 nm; the slits had an optical resolution of 10 nm, and the scan speed was 1200 nm min−1 . These parameters were employed when an aqueous solution of CB[7] was used as the subphase of DPPC/Chol monolayers and also when this same solution was not left in contact with the monolayer. A control emission spectrum was also recorded for water stirred with 1 ␮L of labeled Chol, for 16 h. 2.5. Langmuir–Blodgett films The monolayers were deposited on BK-7 glass Plates 10 × 15 mm previously washed with H2 SO4 /H2 O2 solution and abundantly washed with Milli-Q water. A one-layer Langmuir–Blodgett film was formed by immersing the solid substrates in the subphase solution before spreading the phospholipid; the plates were suspended at a constant withdrawal speed of 0.038 mm s−1 , keeping the surface pressure constant at 30 mN m−1 . The transfer ratio (TR) to hydrophilic BK-7 glass plates was ∼0.87 for the DPPC monolayer formed on water, and in the range of 0.80–0.84 for the mixed DPPC/Chol monolayer, TR was slightly higher for monolayers formed on CB[7] and CD solutions, as compared with other subphases. 2.6. Surface dilatational rheology A pendant drop tensiometer coupled to an Oscillating Drop System (OCA20-ODS, Dataphysics, Germany) was employed to measure the dilatational surface rheological parameters. The surface dilatational elasticity (also called surface elasticity or surface elasticity modulus) is defined as E = d␥/dlnA, where d␥ accounts for the infinitesimal surface tension gradients upon a variation in the relative area (A). The equilibrium value is the Young modulus, identified as the surface compressional modulus (Cs −1 ) [20] obtained from the ␲-A curves. In dynamic conditions, films ranging from perfectly viscous to perfectly elastic exist. Perfectly elastic films do not undergo relaxation process in the timescale of the experiment. Therefore, there are no phase differences (described by the phase angle, ) between the perturbation (deformation) and the response of the system (change in surface tension). On the other hand, phase differences exist when the interface exhibit a viscoelastic behavior. By introducing a complex modulus: E =|E|cos  + i |E|sin␪, one can account for this feature [21]. The imaginary part of this quantity explains the energy dissipation process and relates to the surface dilatational viscosity (␩␴ ): ␩␴ = (|E|sin␪)/␻, where the angular frequency, ω = 2␲f, is established when a sinusoidal disturbance with a frequency f is imposed to the interface. In the experiments described here the drop deformation, (A/A), was 5%, with oscillation frequency 0.15 to 0.45 Hz. 2.7. Time-resolved fluorescence microscopy This technique, also named Fluorescence Lifetime Imaging Microscopy, (FLIM), is based [22] on differences in the fluorescence decay of probes conveniently added to the system. In our study, 25-{n-[(7-nitro-2-1,3-benzoxadiazol-4-il)metil]amino}-27-norcolesterol (Chol-NBD), Avanti Polar Lipids,

C.B. Tovani et al. / Colloids and Surfaces B: Biointerfaces 111 (2013) 398–406

was used as probe at a concentration of 1 mol% in relation to total lipids. The liquid monolayers were transferred to BK-7 glass substrates at a surface pressure of 30 mN m−1 and temperature of 36.5 ◦ C. The images were recorded by a confocal Olympus IX71 fluorescence microscope coupled to a PicoQuant Microtime 200 time resolved fluorescence apparatus, with a pulsed diode laser operated at excitation wavelength of 470 nm; detection was accomplished using a bandpass filter above 470 nm, with an optical slit resolution of 28.0 nm.

3. Results and discussions 3.1. Equilibrium studies: surface pressure – molecular area curves At room temperature, 21.5 ◦ C, DPPC exists in the gel state; at 36.5 ◦ C, it is very close to the transition temperature [23]. We verify the liquid-expanded to liquid-condensed phase transition at 21 ◦ C which agrees with previous results [24]; Chol extinguished this transition (Fig. 2A and C). The presence of CB[5] and CB[6] at a molar concentration of 18 ␮M in the subphase support of the monolayers does not change the equilibrium surface pressure curves at any of the investigated temperatures (Fig. 2A and B). We decided to use low CB[n] concentration to ensure all the CB[n] samples were soluble at the subphases. The DPPC phase transition increased from 4.4 mN m−1 in pure water to 5.8 mN m−1 in the presence of CB[5] and CB[6], and to 6.6 mN m−1 in the presence of CB[7] and ␤-CD. At 36.5 ◦ C, the ␲-A curves recorded for DPPC are practically coincident and quite expanded in all cases. For the mixed monolayers, DPPC/Chol (80/20, w/w), CB[n] and ␤-CD do not affect the ␲-A isotherms at 21 ◦ C. At the human body temperature, (36.5 ◦ C, Fig. 2D), though one can detect two different kinds of behavior as compared to the ␲-A curve obtained for DPPC monolayer formed on pure water: a condensing effect arises when the mixed monolayer is formed on CD or CB[7]. Consequently, the minimum molecular area decreases and the monolayers formed on CB[5] and CB[6] expand. The latter phenomenon often stems from monolayer insertion or repulsion effects. Condensing effects, on the other hand, may result from other phenomena: diminished molecular repulsion (which occurs mainly with electrically charged molecules), complex formation, and molecular realignment or withdrawal of molecules from the interface. The pKa value reported for the phosphate group of phosphatidylcholine is 1.7–3.0 [23], complete protonation occurs at pH 5 [25]. Chol is a neutral lipid; recently it has been shown [26] that egg PC interacts strongly Chol–the intermolecular forces can render a 1:1 complex at the interface. In this paper we used a DPPC/Chol weight proportion (80/20, w/w) that is equivalent to a Chol molar fraction, (XChol ) of 0.32. At this proportion, the complex would occupy about 64% of the surface area as compared with the maximum occupation for XChol = 0.5. Better surface packing or molecular realignment would only take place if additional molecules present in the subphase (␤-CD or CBs) mediated an even stronger interaction among the molecules already present at the air/liquid interface. Otherwise, the macrocycles added to the subphase may form inclusion complexes with lipids, especially with Chol molecules, removing them from the interface. Table 1 summarizes the quantification of these effects. For example, the area of the DPPC/Chol monolayer formed on CB[7] decreases by a value equivalent to 1/5 of the area occupied by a Chol molecule inserted perpendicularly to the interface. Therefore, in the experimental conditions used in this work CB[7] may solubilize one Chol molecule every five DPPC molecules. To better characterize this behavior, we also investigated the dynamic rheological parameters.

401

Table 1 Relative minimum molecular area variations as calculated from the ␲-A curves presented in Fig. 2D, recorded for DPPC/Chol monolayers on different subphases at 36.5 ◦ C. Subphase

Aa (A˚ 2 )

␤-CD CB[5] CB[6] CB[7]

−2.5 +6.5 +5.0 −6.0

Acond = molecular area at ␲ = 30 mN m−1 . a A = (Acond obtained for the isotherm recorded on the subphase containing the indicated additive)–(Acond obtained for the isotherm recorded on pure water). Table 2 Surface dilatational viscoelastic modulus (E) at ␲ 6.4 mN m−1 , f 0.15–0.45 Hz, A/A 5%, for DPPC/Cholesterol 80:20 monolayers at 21 ◦ C and 36.5 ◦ C. E (mN/m) Temperatures Subphases

21 ± 0.5◦ C

Pure water ␤-CD CB[5] CB[6] CB[7]

62.3 42.2 67.8 75.2 45.2

± ± ± ± ±

0.9 1.0 1.3 2.5 1.3

36.5 ± 0.5 ◦ C 67.2 45.4 68.5 72.8 44.3

± ± ± ± ±

1.5 1.3 1.4 1.3 2.0

3.2. Dynamic studies: dilatational surface rheology at low additive concentration The equilibrium ␲-A curves revealed only very small changes in the compressional modulus (Cs −1 ), except for the mixed monolayer at 36.5 ◦ C. For the dynamic experiments, we investigated different surface pressures and deformation frequencies. We decided to maintain the surface pressure in the range of 6–7 mN m−1 , which is higher than the surface pressure of the DPPC monolayer and for which we detected larger variations in the viscoelastic modulus (E) for the different subphases. In all studied systems, E increases with the frequency in the range of 0.15–0.45 Hz and depends on the initial surface pressure (Fig. 3). We conducted experiments in quadruplicate; to compare the influence of the different additives, we accomplished linear regression of the plot of E versus surface pressure, measured at the highest frequency. An interpolation helped us to compare the rheological data for the different additives at the same initial surface pressure. At 21 ◦ C, below the normal human body temperature, E is higher for pure DPPC monolayer and decreases in the presence of additives (␤-CD or CB[n]), with no significant differences among them. The average E value is 42.9 ± 2.0 mN m−1 with an individual deviation in the range of 0.6 and 1.5 mN m−1 . At 36.5 ◦ C the additives virtually do not affect the pure DPPC monolayer, except that the E value recorded for CB[7] is slightly higher; the average E value was 38.8 ± 1.5, at 7.0 mN m−1 , including the value obtained for the pure water subphase. The presence of Chol in DPPC:Chol the 80:20 w/w increases both Cs−1 and E, when the monolayer is formed on a subphase consisting of pure water. The increased rigidity of DPPC monolayer is already known and can be attributed to the role of Chol as a space filler [27] in some PC:Chol ratios. As for additives at a concentration of 18 ␮mol L−1 and at 21 ◦ C, E decreases in the presence of ␤-CD or CB[7] (Table 2), whereas a condensing effect arises in the presence of CB[5] and CB[6]. In all cases, the phase angles are not higher than 3◦ . A similar trend occurs at 36.5 ◦ C. Because Chol rises the value of E for DPPC monolayers, E approaches the E value obtained for the pure DPPC monolayer when ␤-CD or CB[7] is present in the subphase. This result evidences that Chol is removed from the interface, even at low additive concentration (18 ␮mol L−1 ), corroborating the

C.B. Tovani et al. / Colloids and Surfaces B: Biointerfaces 111 (2013) 398–406

(A)

50

Surface Pressure ( mN m -1)

Surface Pressure ( mN m -1)

402

40

30

20

10

0

(B)

50

40

30

20

10

0

30

40

50

60

70

80

90

100

30

40

50

60

70

2

90

Molecular Area (Å )

(C)

50

80 2

Surface Pressure ( mN m -1)

Surface Pressure ( mN m -1)

Molecular Area (Å )

40

30

20

10

0

50

(D)

40

30

20

10

0 30

40

50

60

70

80

90

100

110

2

Molecular Area (Å )

30

40

50

60

70

80

90

100

110

2

Molecular Area (Å )

Fig. 2. Surface pressure-molecular area curves at 21.5 ◦ C (left) and 36.5 ◦ C (right) for: DPPC (A and B) DPPC/Chol (C and D) monolayers, formed on pure water (), ␤-CD (), CB[6] (), CB[5] () and CB[7] subphases (). The molecular area is related to DPPC for all isotherms. DPPC:Col is 80:20 (w/w).

equilibrium surface data. However, further evidence is necessary to confirm whether this result should be really attributed to Chol removal as a consequence of the formation of an inclusion complex between Chol and ␤-CD or CB at the interface.

3.3. Time-resolved fluorescence microscopy Fluorescent probes are often used in mono- and bilayers, to provide information on the molecular environment of the

Fig. 3. Time resolved fluorescence micrographies of LB films formed from DPPC/Chol (80/20 w/w) monolayers prepared on pure water and on aqueous solutions of ␤-CD and CB[n] 0.018 mM. Excitation wavelength was 470 nm.

C.B. Tovani et al. / Colloids and Surfaces B: Biointerfaces 111 (2013) 398–406

A

403

70

11.5 O

DPPC/COL-H2O: 36 C -1

Cs = (70.0 +1.4) mN m

-1

65

-1

Surface Pressure ( mN m )

11.0

E -1 (mN m )

10.5 10.0

60 55

9.5 50

9.0

--

Emax = 69.7+1.3 mN m

-1

45

8.5 8.0

1.0

4.19

4.20

4.21

4.22

4.23

1.5

2.0

2.5

3.0

ω (rad/s)

4.24

ln A

B

11.5 o

DPPC/Chol - 4 mM -CD: 36 C -1 -1 Cs = 71,5 + 1,3 mN m 45

10.5

E -1 (mN m )

Surface Pressure mN m

-1

11.0

10.0 9.5

40

-- Emax = 59.4+0.7 mN m

9.0

-1

35

8.5 8.0

1.0

4.18

4.19

4.20

4.21

4.22

4.23

4.24

1.5

2.0

ω (rad/s)

2.5

3.0

ln A

C

11,5

75 O

DPPC/COL - 4 mM CB[7]: 36 C -1 -1 Cs = (71.9 +0.8) mN m

70

10,5 E -1 (mN m )

-1

Surface Pressure (mN m )

11,0

10,0 9,5

65 60 55

9,0 --

50

8,5

Emax = 73.8+1.9 mN m

-1

45

8,0

1.0

4,12

4,13

4,14

4,15

4,16

4,17

1.5

2.0

2.5

3.0

ω (rad/s)

ln A Fig. 4. Surface compressional modulus, Cs−1 , (left) and dilatational viscoelastic modulus, E, variation with angular frequency (right) at ∼10 ± 1 mN m−1 , A/A 5% for DPPC/Chol monolayers formed on pure water and 4 mmol L−1 aqueous solutions of ␤-CD and CB[7] at 36.5 ◦ C.

C.B. Tovani et al. / Colloids and Surfaces B: Biointerfaces 111 (2013) 398–406

membrane model. In this sense, different emission features may change, such as the intensity, maximum wavelength, and lifetime. The emission spectra depend on the relaxation dynamics, which in turn is frequently related to hydration, especially for liposomes, and reflects the mobility of the lipid layer [28]. For instance, a high lipid order and slow lipid mobility characterize the gel phase. Probes with different polarities may be located in different regions of the monolayer or bilayer. Polar fluorophores, such as NBD used in the present study, are useful to probe headgroup regions [29]. For LB films transferred at 30 mN m−1 , the gel phase exists, so the probe should adopt a vertical orientation [30]. As seen from the ␲-A curves and viscoelastic moduli, Chol should induce an even more ordered phase, represented by the increase in Cs −1 and E as compared with the pure DPPC monolayer. The most familiar fluorescence microscopy technique is based on steady state fluorescence detection and the contrast is provided by differences in the affinity of the molecules for the probe and their distribution within the sampled surface. Time-resolved fluorescence can furnish information about the emission intensity and the lifetime decay of the probe, the latter culminates in the contrast observed in the images. Fig. 3 depicts micrographies obtained for LB films prepared from monolayers of DPPC/Chol 80:20 on different subphases and transferred at 30 mN m−1 . We observed considerable fluorescence enhancement only when ␤-CD was present in the subphase. In this case, the pre-exponential factor increased from ∼0.09 for monolayers formed on pure water to ∼10,680 in the presence of ␤-CD. Enoch and Swaminathan [31] had already reported this effect for inclusion complexes of diphenyl ethers with CD. The image of the DPPC/Chol LB film formed on ␤-CD displays intense spots distributed in a quite “dark background” modeled by a bi-exponential decay curve with emission lifetime decays of 0.56 ns and 4.7 ns. Meanwhile, the image recorded for the mixed monolayer formed on pure water is quite dark, translated by a mono-exponential decay with ␶1 of 0.81 ns. The emission decay detected in the presence of CB[5] and CB[6] are described by biexponential decay curves, but with the pre-exponential factor B2 for ␶2 <10−4 . Therefore, the contribution of the image background, corresponding to ␶1 and expressed by the pre-expontential factor B1 , is higher, except for CB[7]. In this case the contribution of the slower emission decay is higher than the “background”. The fast lifetime decay, ␶1 , does not change significantly among the different CB[n] being 0.515, 0.512, 0.527 ns for CB[6], CB[5] and CB[7], respectively. For ␤-CD ␶1 is slightly higher, 0.555 ns, but it is still lower than the value measured for the LB film formed on pure water. On the other hand, for the slowest decay, ␶2 , values follow the sequence ␤-CD ∼CB[7] < CB[5] ∼ CB[6], being (4.7 ± 0.2), (5.3 ± 0.2), (5.7 ± 0.3), and (6.2 ± 0.4) ns, respectively. Wagner et al. [32] were the first to describe the enhanced fluorescence of dyes in the presence of CB[n]; for 2-anilinonaphthalene-6-sulfonate (2–6ANS probe) in solution. The authors attributed this enhancement to the formation of a host–guest inclusion complex, with the phenyl group of the 2,6-ANS lying inside the CB cavity. They also proposed that the enhancement is a result of loss of rotational mobility of the phenyl ring encapsulated in the CB cavity. Therefore, in the present case the bi-exponential decay may be ascribed to the presence of free and complexed forms of Chol-NBD. It is likely that the probe reacts to the different environment provided by the additive, ␤-CD or CB[n], which causes a delay in the emission decay. Moreover, it is also reported in reference [33] that, in deaerated water, some inclusion complexes involving CB[7] may even have lower quantum yield, but always longer fluorescence lifetime. It was also found that the fluorescence enhancement depend on concentration and was more easily detected at concentrations higher than 1 mmol L−1 . Because the differences previously described for E were not so evident at 18 ␮mol L−1 , one could wonder whether

the rheology of the interface would also be more sensitive to higher CB[n] concentrations. 3.4. Surface rheology at a high additive concentration: comparison between ˇ-CD and CB[7] Considering the limited amounts of available CB[n] for the experiments, their low solubilities and, their striking similarity with ␤-CD in terms of response in the presence of Chol, we decided to perform additional experiments to compare the rheological behavior of DPPC/Chol monolayers formed on pure water and 4 mmol L−1 ␤-CD and on CB[7] aqueous solutions. We compared the equilibrium surface compressional modulus, (Cs −1 ), obtained from the ␲-A curves with the maximum dilatational viscoelastic modulus, E. The results are organized as plots (Fig. 4). Both moduli are practically coincident for the monolayers formed on water and on CB[7]. Comparing the E values with those obtained at lower additive concentration (Table 2), we found that the different behavior observed for CB[7] disappears. Small differences observed for the monolayer formed on pure water could be related to slight differences in surface pressure, which may override the low concentration effect of the additive. Surface viscosity (␩␴ ), however, is not negligible in this case (see Fig. 5). This parameter is related to dissipation processes and, in mixed systems with soluble compounds, it represents an apparent viscosity [34]; in the present circumstances, it is most likely connected to surface reaction. Fig. 5 also shows that the values obtained for the pure DPPC monolayer and DPPC/Chol formed on pure water are leveled off. In the presence of ␤-CD, surface viscosities decrease with the frequency. At about 2 rad s−1 , corresponding to a characteristic time of about 3 s, the apparent viscosity reaches the value obtained in the absence of the additive. Thus, around 3 s or for deformation frequencies higher than 0.3 Hz, the diffusion and/or reaction processes should be slower than the disturbance applied to the surface, and the monolayer behave as elastic. In the presence of CB[7], though this characteristic time is not reached in the conditions of the experiment (maximum frequency of 0.45 Hz). ˚ and portal CB[7] has significantly larger internal cavity (7.3 A) ˚ as compared with than CB[6] (5.8 and 3.9 A, ˚ diameters (5.4 A) respectively) [35]. Their solubilities are 0.018 mM and 20–30 mM, respectively, lower than that of ␤-CD, which also presents a larger cavity although it is still small to include a cholesterol molecule.

15 DPPC on:

pure water DPPC/Chol on:

pure water and on 4 mM aqueous subphases: β-cyclodextrin

10

CB[7]

ησ

404

5

0 1.0

1.5

2.0

2.5

3.0

-1

ω (rad s ) Fig. 5. Surface dilatational viscosity variation with angular frequency at ∼10 ± 1 mN m−1 , A/A 5% for DPPC/Chol monolayers formed on pure water and 4 mmol L−1 aqueous solutions of ␤-CD and CB[7] at 36.5 ◦ C.

C.B. Tovani et al. / Colloids and Surfaces B: Biointerfaces 111 (2013) 398–406

solid substrates indicated formation of the inclusion complex formation on the basis of the bi-exponential emission curve decay and intensification of the fluorescence. Finally, the increased fluidity of the mixed monolayers in the presence of CB[7], but not in the presence of CB[5] or CB[6] in the subphase, may suggest that the properties of the monolayer are comparable to those obtained in the absence of cholesterol, indicating that these molecules may be displaced from the interface at higher compression. Finally, the detection of emission due to labeled cholesterol in the CB[7] aqueous solution used as the subphase support for the mixed monolayer, but not for the solution that was not in contact with the monolayer nor for the pure water subphase constitutes definitive proof of cholesterol solubilization from the biomembrane model system by the cucurbituril CB[7].

Fluorescence Intensity (arbritary units)

50

40

30

20

10

0 480

405

520

560

600

Emission Wavelength (nm) Fig. 6. Fluorescence spectra recorded for: (—) CB[7] aqueous solution and (----) pure water, as the subphases of DPPC/Chol-NBD 80:20 monolayers, after a contact time of 16 h under mild agitation. Both were subtracted from the spectra of the respective subphases before contact with the monolayer.

Indeed, two stacked ␤-CD are necessary to completely include Chol [35]. As attested by mass spectrometry, proton, and 13 C NMR the CB[7] samples used in this study contain predominantly CB[7] and some traces of CB[5] which is not able to include cholesterol. The finding that surface viscosity in the presence of CB[7] does not reach the value obtained for the DPPC/Chol monolayer formed on pure water in the same conditions may indicate a less reversible inclusion process in this case; this is much likely due to the lower solubility of CB[7] as compared with ␤-CD. Moreover, the rate constants for the formation of inclusion complexes depend on association processes, which in turn rely on available concentrations as well as cavity sizes. Even in the case of a simple Na+ -CB[n], the association rate constant can be much higher for CB[7] than for CB[6] and exceed the one expected for a simple diffusion mechanism [19]. Nevertheless, the data on surface viscosities once again show that the differences of DPPC/Chol monolayer formed on CB[7] solution behaves differently from other subphases. 3.5. Emission spectra of CB[7] aqueous solution used as subphase of DPPC/Chol monolayer To corroborate the presence of Chol dissolved in the subphase we used fluorescent-labeled cholesterol, Chol-NBD, to prepare the spreading solution. We recorded the emission spectrum after a contact time of 16 h between the mixed monolayer and the subphase. We also conducted control experiments using water as subphase and recording the spectrum for a simple CB[7] aqueous solution. We present the results in Fig. 6, where we subtracted the fluorescence spectra recorded for the subphases before the contact. Both water and CB[7] aqueous solution before contact only display Raman scattering peaks. In the case using the CB[7]solution as the subphase, we verified an additional peak at. In the sample containing NBD-chol an emission at 530 nm, which is characteristic of Chol-NBD. 4. Conclusion We present evidence for the formation of an inclusion complex between cucurbiturils and cholesterol for the first time. Dilatational surface rheology data have helped to understand the dynamics of the chemical process occuring at the liquid/air interface. Timeresolved fluorescence microscopy of the monolayers transferred to

Acknowledgements This work was developed with the financial support of FAPESP (São Paulo State Research Foundation) research grants: 11/008509 and 09/54044-3. C.B. Tovani and J.F.V. Souza and M. B. Barioni are grateful to FAPESP and W. M. Pazin is grateful to CAPES for their fellowships, received during the development of this research work A.S. Ito is member of INCT-FCx Brazil, and thank CNPq for research grant. The authors thank Dr. Cynthia M.C.P. Manso for revision of the English language. References [1] J.A. Lundbæk, P. Birn, A.J. Hansen, R. Søgaard, C. Nielsen, J. Girshman, M.J. Bruno, S.E. Tape, J. Egebjerg, D.V. Greathouse, G.L. Mattice, R.E. Koeppe II., O.S. Andersen, J. Gen. Physiol. 123 (2004) 599–621. [2] L. Caseli, R.G. Oliveira, D.C. Masui, R.P.M. Furriel, F.A. Leone, B. Maggio, M.E.D. Zaniquelli, Langmuir 21 (9) (2005) 4090–4095. [3] R. Sogaard, T.M. Werge, C. Bertelsen, C. Lundbye, K.L. Madsen, C.H. Nielsen, J.A. Lundbk, Biochemistry 45 (43) (2006) 13118–13129. [4] S. Guanbin, Y. Weiqun, L. Baoan, L. Mian, W. Zezhi, W. Bochu, Colloids Surf. B 24 (2002) 327–332. [5] J.A. Lundbaek, P. Birn, A.J. Hansen, R. Sogaard, C. Nielsen, J. Girshman, M.J. Bruno, S.E. Tape, J. Egebjerg, D.V. Greathouse, G.L. Mattice, R.E. Koeppe II., O.S. Andersen, J. Gen. Physiol. 123 (5) (2004) 599–621. [6] C. Socaciu, R. Jessel, H.A. Diehl, C. Socaciu, R. Jessel, H.A. Diehl, Chem. Phys. Lipids 106 (1) (2000) 79–88. [7] S.M. Nomura, Y. Mizutani, K. Kurita, A. Watanabe, K. Akiyoshi, Biochim. Biophys. Acta 1669 (2) (2005) 164–169. [8] G. Karagiorga, G. Nakos, E. Galiatsou, M.E. Lekka, Intensive Care Med. 32 (2006) 116–123. [9] L.C. Gunasekara, R.M. Pratt, W.M. Schoel, S. Gosche, E.J. Prenner, M.W. Amrein, Biochim. Biophys. Acta 1798 (2010) 986. [10] R. Bergeron, Y. Machida, K. Bloch, J. Biol. Chem. 250 (1975) 1223–1230. [11] D. Castagne, M. Fillet, L. Delattre, B. Evrard, B. Nusgens, G. Piel, J. Inclusion Phenom. Macrocyclic Chem. 63 (2009) 225–231. [12] R. Zidovetzki, I. Levitan, Biochim. Biophys. Acta 1768 (2007) 1311–1324. [13] E. Masson, X. Ling, R. Joseph, L. Kyeremeh-Mensah, X. Lu, R.S.C. Adv. 2 (2012) 1213–1247. [14] A. Day, A.P. Arnold, R.J. Blanch, B. Snushall, J. Org. Chem. 66 (2001) 8094–8100. [15] W.A. Freeman, W.L. Mock, N.Y. Shih, J. Am. Chem. Soc. 103 (1981) 7367–7368. [16] M.V. Rekharsky, Y. Ko Ho, N. Selvapalam, K. Kim, Y. Inoue, Supramol. Chem. 19 (2007) 39–46. [17] V.D. Uzunova, C. Cullinane, K. Brix, W.M. Nau, A.I. Day, Org. Biomol. Chem. 8 (2010) 2037–2042. [18] M.P. Beseniˇcar, A. Bavdek, A. Kladnik, P. Maˇcek, G. Anderluh, Biochim. Biophys. Acta 1778 (2008) 175–184. [19] H. Tang, D. Fuentealba, Y. Ho Ko, N. Selvapalam, K. Kim, C. Bohne, J. Am. Chem. Soc. 133 (2011) 20623–20633. [20] J.T. Davies, E.K. Rideal, Interfacial Phenomena, 2nd Ed., Academic Press, N.Y., U.S.A., 1963. [21] J. Lucassen, M. Van Den Tempel, Chem. Eng. Sci. 27 (1972) 1283–1291. [22] T.W.G. Gadella Jr., T.M. Jovin, R.M. Clegg, Biophys. Chem. 48 (1993) 221–239. [23] S.A. Tatulian, in: G. Cevc (Ed.), Phospholipids Handbook, Marcel Dekker, New York, 1993, Chapter 14. ˜ [24] P. Toimil, G. Prieto, J. Minones Jr., F. Sarmiento, Phys. Chem. Chem. Phys. 12 (2010) 13323–13332. [25] H.H.J. Girault, D.J.J. Schiffrin, Electroanal. Chem. 179 (1984) 277. [26] I. Brzozowska, Z.A. Figaszewski, Colloids Surf. B 23 (2002) 51–58. [27] F.R. Maxfield, I. Tabas, Nature 438 (2005) 612–621. [28] A.P. Demchenko, Y. Me’ly, G. Duportail, A.S. Klymchenko, Biophys. J. 96 (2009) 3461–3470.

406 [29] [30] [31] [32]

C.B. Tovani et al. / Colloids and Surfaces B: Biointerfaces 111 (2013) 398–406

A. Chattopadhyay, S. Mukherjee, Langmuir 15 (1999) 2142–2148. L.A. Bagatolli, E. Gratton, Biophys. J. 78 (2000) 290–305. I.V.M.V. Enoch, M. Swaminathan, J. Lumin. 127 (2007) 713–720. B.D. Wagner, S.J. Fitzpatrick, M.A. Gill, A.I. MacRae, N. Stojanovic, N. Can. J. Chem. 79 (2001) 1101–1104. [33] V. Ramamurthy, Y. Inoue, Supramolecular Photochemistry Controlling Photochemical Process, Wiley, N.J., U.S.A, 2011, pp. 102, Chapter 3.

[34] I.B. Ivanov, K.D. Danov, K.P. Ananthapadmanabhan, A. Lips, Adv. Colloid Interface Sci. 114–115 (2005) 61–92. [35] J. Kim, I.-S. Jung, S.-Y. Kim, E. Lee, J.-K. Kang, S. Sakamoto, K. Yamaguchi, K. Kim, J. Am. Chem. Soc. 122 (2000) 540–541. [36] W. Saenger, J. Jacob, K. Gessler, T. Steiner, D. Hoffmann, H. Sanbe, K. Koizumi, S.M. Smith, T. Takaha, Chem. Rev 98 (1998) 1787–1802.