Evidence of Increased Hydrophobicity and Dynamics inside the Tail

Jul 9, 2019 - •S Supporting Information. ABSTRACT: New ... INTRODUCTION. Membrane proteins contain large surfaces of hydrophobic .... HOMO and LUMO ...
0 downloads 0 Views 6MB Size
Article Cite This: Langmuir XXXX, XXX, XXX−XXX

pubs.acs.org/Langmuir

Evidence of Increased Hydrophobicity and Dynamics inside the Tail Region of Glycolipid Self-Assemblies Using 2‑n‑Alkyl-Pyrene Derivatives to Probe Different Locations N. Idayu Zahid,†,‡ Lei Ji,§ M. Faisal Khyasudeen,†,‡ Alexandra Friedrich,§ Rauzah Hashim,‡ Todd B. Marder,§ and Osama K. Abou-Zied*,† †

Department of Chemistry, Faculty of Science, Sultan Qaboos University, P.O. Box 36, Postal Code 123, Muscat, Sultanate of Oman Centre for Fundamental and Frontier Sciences in Nanostructure Self-Assembly, Department of Chemistry, Faculty of Science, University of Malaya, 50603 Kuala Lumpur, Malaysia § Institut für Anorganische Chemie and Institute for Sustainable Chemistry & Catalysis with Boron, Julius-Maximilians-Universität Würzburg, Am Hubland, 97074 Würzburg, Germany

Downloaded via NOTTINGHAM TRENT UNIV on July 17, 2019 at 06:33:21 (UTC). See https://pubs.acs.org/sharingguidelines for options on how to legitimately share published articles.



S Supporting Information *

ABSTRACT: New designer biofluorophores are being increasingly used in the investigation of complex cellular processes. In this study, we utilized new derivatives of pyrene (Py), i.e., 2-n-alkyl-pyrenes (Py-C4 and Py-C8), in order to probe different regions inside the hydrophobic tail of n-dodecyl β-D-maltoside (βMal-C12) in two different phases (cubic ↔ lamellar). Although the sensitivity to the local environment is reduced compared to that of Py, attaching C4 and C8 at the 2position of Py can provide a possible means to probe the local hydrophobicity in different parts of the tail region. The absence of excimer fluorescence and the ratio of the vibronic fluorescence peak intensities (I1/I3) in a lipid environment indicate the existence of Py as monomers in the hydrophobic region, similar to hydrophobic solvation, yet close to the headgroup region. When Py is replaced by Py-C4 and Py-C8, there is a small increase in hydrophobicity (reduction in I1/I3) as the Py moiety is pulled deeper inside the tail region of both cubic and lamellar phases. The larger space of the tail region in the lamellar phase is reflected as more local hydrophobicity measured by the probes which can penetrate deep inside, whereas the curved structure of the cubic phase limits the available space for the probes. Three fluorescence lifetime components were measured in lipid, indicating the heterogeneous nature of the hydrophobic region. In the lamellar phase, a large reduction in the average lifetime value, led by the long decay component, was measured for Py-C4 (reduction by 25%) and Py-C8 (45%) compared to that of the parent Py. This observation suggests the presence of a mechanism of interaction more collisional than static between the Py moiety and the tail region of the bilayer unit due to the ample space provided by the lamellar phase as the probe is buried deeper inside the hydrophobic region. A much smaller effect was observed in the cubic phase and was correlated with the tight environment around the probes, which stems from the increased curvature of the cubic phase. The current results provide a deeper understanding of the hydrophobic region during phase transition of lipid self-assembly which is important for better control during the process of membrane-protein crystallization.



INTRODUCTION Membrane proteins contain large surfaces of hydrophobic residues embedded in the nonpolar bilayer region of the lipid membrane, and only minor parts of these proteins are solvated by water molecules.1,2 Even though working with these special proteins is challenging, they are significant in controlling many fundamental biochemical processes apart from being useful as pharmaceutical targets.3 Crystal structures of membrane proteins at high resolution are highly desired but demanding, as they require a comprehensive understanding of the structure−function relationship at the molecular level.1,4 These proteins show extremely low solubility in aqueous solution. The crystallization experiment usually involves extraction of the membrane protein from the membrane © XXXX American Chemical Society

environment by solubilization with detergent, thus forming the detergent−protein crystal complex and not simply the protein crystal alone. Among the different detergents, the alkyl maltopyranoside is widely used, accounting for 50% of all structures in the protein database, followed by the alkyl glucopyranosides (23%), amine oxides (7%), and polyethylene glycols (7%).1 Although the choice of detergent depends on many different parameters, the n-dodecyl-β-D-maltopyranoside (βMal-C12) is considered to be the most successful alkyl Received: June 11, 2019 Revised: June 26, 2019

A

DOI: 10.1021/acs.langmuir.9b01767 Langmuir XXXX, XXX, XXX−XXX

Article

Langmuir

assemblies.24−26 These studies established Py fluorescence as a useful tool to characterize the lipophilic region, where the results show that Py molecules disperse among the tails of the hydrophobic part as monomers and the vibronic peak ratio of I1/I3 implies that Py is close to the sugar headgroups. In the present work, we extend the study to probe different locations in the hydrophobic tail region of lipid self-assemblies. Such probing is attainable by attaching an alkyl chain of different lengths to the Py parent molecule without altering the photophysical properties, specifically its fluorescence structure and lifetime. This is challenging because most studies on Py substitution involve derivatives substituted at the 1-, 3-, 6-, and 8-carbon atoms which are electrophilic sites of maximum contribution from the HOMO, hence greatly influencing the spectroscopic properties of the molecule. In contrast, substitution at the 2- and 2,7-positions often preserves the photophysical properties of the parent Py molecule. However, substituting Py at these positions has proven to be synthetically challenging owing to the presence of the nodal plane in the HOMO and LUMO passing through the 2- and 7positions.27,28 Here, we take advantage of the methodology achieved by some of us in successfully preparing 2- and 2,7substituents of Py.27,28 We report in the current study the fluorescence behavior of Py and 2-alkyl substituents of Py, 2-(n-butyl)pyrene (Py-C4), and 2-(n-octyl)pyrene (Py-C8) (shown in Figure 1b), in solution and in probing the phase transition from cubic to lamellar in the βMal-C12 lipid self-assembly. We select these phases because they conceptually represent the membrane system. Using these molecules, we successfully probed different regions inside the hydrophobic part of the lipid. We correlate the fluorescence change of the Py moiety to the change in the local hydrophobicity and dynamics as the probe penetrates deeper inside the tail region. The results from this work should be useful in understanding the hydrophobic properties of different phases of the same lipid, which is important in membrane protein crystallization, and other interactions related to lipid−membrane systems.

maltopyranoside detergent, especially in transporters and respiratory complexes.1,2 βMal-C12, which has a maltose headgroup linked to a lauryl hydrophobic tail (see Figure 1a), has also been employed in

Figure 1. Chemical structures of (a) n-dodecyl β-D-maltoside (βMalC12) and (b) pyrene (generic) with its derivatives. The subtituent R represents H, n-C4H9, or n-C8H17 which denote pyrene (Py), 2-(nbutyl)pyrene (Py-C4), and 2-(n-octyl)pyrene (Py-C8), respectively.

applications such as in the purification and stabilization of RNA polymerase5 and detection of protein−lipid interactions.6,7 βMal-C12 is produced synthetically and also known as both thermotropic and lyotropic liquid crystals.8−10 The phase study of βMal-C12 has mostly focused on lyotropic systems, i.e., lamellar, cubic, and hexagonal phases because these are closely related to biological membranes.11,12 The lyotropic selfassembly properties are suitable as model systems for understanding specific membrane function in biosystems and for studying the phase transition which is often associated with the rupture and repair of membranes.13 Most structural analyses of membranes have utilized biophysical techniques with an emphasis on spectroscopic methods. Fluorescence spectroscopy has been one of the most powerful spectroscopic tools to account for the flexibility, heterogeneous organization, and dynamics of biological and model membranes. It has an appropriate time scale, minimal perturbation, noninvasive nature, and high sensitivity and specificity.14 Pyrene (Py, shown in Figure 1b) is one of the most frequently used lipid-linked fluorophores due to its strong hydrophobic nature.15 Py monomer is polarity-sensitive and has a notably long lifetime (e.g., 382 ns in deoxygenated cyclohexane).16 Furthermore, one Py excited molecule can interact with one ground state Py molecule to form an excited dimer (excimer) which is convenient for addressing intra- and intermolecular interactions upon lipid binding.17 The Py monomer emission is split into resolved vibronic fine structures within the range 373 to 394 nm, whereas its excimer gives a structureless emission band centered at ∼480 nm.18 These interesting spectral characteristics of Py make it often employed as a probe in the study of micellar structures19−21 and membrane-like systems of phospholipid dispersions.22,23 We have previously used Py to probe the local environment in the hydrophobic region of a series of glycolipid self-



EXPERIMENTAL SECTION

Materials. βMal-C12 (98%) and Py were obtained from SigmaAldrich. The synthesis and characterization of Py-C4 and Py-C8 are described in the Supporting Information. All chemicals and solvents were used without further purification. Sample Preparation. All samples were prepared according to the same procedure used in our previous work.24−26,29,30 Briefly, the samples were prepared by dissolving 4.2/5.6/6.4 mL of 2.6/1.9/1.7 μM methanolic Py/Py-C4/Py-C8, respectively, with 120 mg βMal-C12. Methanol was then removed. Due to their hydrophobicity, the probes are expected to reside in the tail region of the βMal-C12. Then, an appropriate amount of buffer (25 mM sodium phosphate buffer, pH 7.2) was added to each of the probe-βMal-C12 mixture in a 4-mmdiameter quartz tube of 4 cm length to form the desired system containing 80% probe-βMal-C12 and 20% buffer (w/w). The quartz ampule was flame-sealed, and the probe-βMal-C12-buffer mixture was then homogenized by heating, combined with up and down centrifuging. A viscous yet homogeneous sample was formed. The concentration of Py and its derivatives Py-C4 and Py-C8 in the βMalC12/water system was 100 μM for the steady-state and time-resolved experiments. The value was based on an estimated density of ∼1.2 g/ mL for the mixture of each probe-lipid-buffer. According to the βMalC12-water binary phase diagram,31 at this water concentration a cubic phase is formed at 50 °C which transforms into a lamellar phase when heated to 75 °C. The phases were confirmed to be formed by illuminating the sample and examining it through a cross-polarizing B

DOI: 10.1021/acs.langmuir.9b01767 Langmuir XXXX, XXX, XXX−XXX

Article

Langmuir filter. The lamellar phase sample showed a birefringent characteristic under the cross-polarizing filter which was not observed in the case of the cubic phase sample. The samples included the lipid mixed with the different probes, in addition to one sample with no probe which was used as a standard in the measurements. For the probes in solution, a stock solution of each probe in methanol (5 mM) was first prepared, then diluted with the buffer to reach the desired concentration of 50 μM. The final methanol:H2O (v/v) mixture in the solution was 10:90. A ratio above 20:80 was shown to have similar properties of pure water.32,33 The samples appeared as translucent (neither cloudy nor completely clear), but measurable. For the probes in cyclohexane, a direct preparation using cyclohexane as a solvent was used to reach a final concentration of 50 μM. Steady-State and Time-Resolved Fluorescence Studies. Steady-state fluorescence spectra were performed on a Shimadzu RF-5301 PC spectrofluorophotometer. Fluorescence emission spectra of pyrene-labeled samples were recorded between 345 and 650 nm by setting the excitation wavelength at 340 nm. Both excitation and emission slit widths were set at 1.5 nm. Lifetime measurements for Py and its derivatives in solution were performed using a TimeMaster fluorescence lifetime spectrometer obtained from Photon Technology International. Excitation was at 340 nm using a light emitting diode. The system response time was estimated from the scattered light to be approximately 1.5 ns. The transients were measured using stroboscopic detection.34 Timeresolved fluorescence of Py and its derivatives in different lipids was measured using time-correlated single photon counting (TCSPC) in order to accurately determine fast dynamics. The TCSPC setup, described elsewhere,35 is part of an ultrafast spectrometer (Halcyone, Ultrafast Systems, LLC) that is used to measure femtosecond fluorescence upconversion. Briefly, the excitation pulse was obtained using a regenerative amplified Ti:sapphire laser (Libra, Coherent). The Libra generates compressed laser pulses (70 fs fwhm) with output of 4.26 W at a repetition rate of 5 kHz and centered at 800 nm. 90% of the output pulse was used to pump a Coherent OPerA Solo (Light Conversion Ltd.) optical parametric amplifier to generate spectrally tunable light spanning the range 240−2600 nm. For the current measurements, the OperA was adjusted at 340 nm and was used as the excitation beam, after passing through a depolarizer (DPU-25, Thorlabs). A PMT was used as the detector and the IRF was ∼250 ps, with a time window up to 200 μs. Fluorescence was attenuated and directed to the detector and a monochromator was used to adjust the detection wavelength. Decays were recorded to ∼10,000 counts in the peak channel. The decay transients were fitted to multiexponential functions convoluted with the instrument response function (IRF). The temperature of the samples was controlled within ±0.1 °C at 50.0 or 75.0 °C. All samples were equilibrated overnight at the desired temperature to ensure that the required phase was obtained before the measurements. Each experiment was repeated three times and the average values were presented. The change of temperature was found to have no effect on the probes’ spectroscopy in solution.

Figure 2. Fluorescence spectra of Py and its derivatives in (a) cyclohexane and (b) pH 7.2 buffer. For ease of comparison, the spectra are normalized at I5 in cyclohexane and at the excimer peak in buffer.

environmentally sensitive and has been used to elucidate the local environment of pyrene.18,38,39 An increase in the I1/I3 ratio is indicative of increased apparent polarity. From our previous study on pyrene, its estimated I1/I3 ratio was 0.58 in cyclohexane.24 When an alkyl chain is attached at the 2position of pyrene, there is an apparent increase in the I1/I3 ratio as shown in Table 1 for Py-C4 and Py-C8. We also Table 1. Fluorescence Intensity Ratio of the Vibronic Peaks (I1/I3) of Py and Its Derivatives in Cyclohexane and pH 7.2 Buffer



RESULTS AND DISCUSSION Fluorescence of Py and Its Derivatives in Solution. In order to probe the hydrophobic region of lipids using the pyrene molecules, it is important first to understand the fluorescence signal of Py, Py-C4, and Py-C8 in nonpolar and polar (aqueous) solvents as a point of reference. Figure 2 shows the fluorescence spectra of pyrene and its derivatives measured in cyclohexane and buffer. In cyclohexane, the sharp and structured band (∼360−450 nm) is due to the monomer fluorescence.36,37 No other peaks were observed for pyrene and its derivatives in cyclohexane. The structured fluorescence from the monomer species has two characteristic peaks with maxima at ∼375 and ∼385 nm (marked in Figure 2a as I1 and I3, respectively). The ratio of the two peak intensities (I1/I3) is

I1/I3

a

Probe

Cyclohexane

pH 7.2 buffer

Py Py-C4 Py-C8

0.61 0.77 0.76

1.08a 0.99a -

Value is not accurate. See text for details.

observed a slight red shift in the I1 and I3 peaks for Py-C4 and Py-C8 (Figure 2a). As these peaks are sensitive to the local polarity, the shift can be correlated with the presence of the alkyl chains. This is supported by the absence of any red shift for the peak at ∼394 nm (I5) which is not polarity dependent.40 C

DOI: 10.1021/acs.langmuir.9b01767 Langmuir XXXX, XXX, XXX−XXX

Article

Langmuir In buffer, Py tends to form excimers even at low concentrations which is manifested in the unstructured broad band centered at ∼465 nm as shown in Figure 2b. The hydrophobic nature of Py is expected to cause the molecules to cluster close to each other in order to avoid the highly disliked polar nature of the solvent. For the two Py derivatives, the excimer fluorescence is red-shifted by ∼20 nm, compared to that of Py. This observation implies greater stability of the excimers in the presence of the alkyl chains. The high affinity to form dimers is evidenced in the much lower fluorescence intensity from the monomers in Py-C4, compared to that in Py, and the absence of monomer fluorescence in Py-C8. The very high concentration of excimers causes the excimer fluorescence peak to distort the intensity of the monomer vibronic peaks as shown in Figure 2b. In fact, the I3 peak is absent in Py. This explains the very small I1/I3 values reported here (1.08 for Py and 0.99 for Py-C4 in Table 1) which are underestimated to a large extent when compared with the reported values for Py of 1.8−1.9.41,42 In the current work, we use the spectral shape of Py and its derivatives in buffer only as a guide in order to determine if the probe is in contact with water in the lipid selfassembly. In order to correlate the spectral changes when the new probes are used to examine the hydrophobic region of lipids, it is necessary to characterize their respective I1/I3 values in different solvents. In this work, we measured this ratio for PyC4 and Py-C8 in nonprotic solvents of varying polarity. Figure 3 displays the change in the I1/I3 ratio as a function of the

Figure 4. Fluorescence decay transients of Py and its derivatives (color coded) after excitation at 340 nm in (a) cyclohexane and (b) pH 7.2 buffer, where the signal was measured using a 380 nm (a) and a 495 nm (b) long-pass filters. Instrument response function (IRF) is shown as a dashed line. Black solid lines represent the best fits. The insets show the corresponding decay transients measured for a longer time range.

lifetime values are summarized in Table 2. Fluorescence decay was measured in the monomer and excimer spectral regions by Table 2. Fluorescence Lifetime Measurements (ns) for Py and Its Derivatives in Solventsa Figure 3. Fluorescence ratio I1/I3 as a function of the empirical parameter ET(30) for Py, Py-C4, and Py-C8. Details about the solvents used and the values of I1/I3 are tabulated in Table S3, SI.

empirical parameter of solvent polarity ET(30).43 The corresponding values for Py are included for comparison. The latter results are in substantial agreement with the results of Dong and Winnik.41,42 The spectra of the current results are included in SI, Figures S8−S10 and Table S3. As can be seen in Figure 3, the response of the I1/I3 ratio for Py-C4 and Py-C8 to the polarity of the solvents is much less sensitive than that of Py, as reported for other 1-substituted Pys.44 On the other hand, the similarity between the two probes in sensitivity may be useful when used in lipids since each one is expected to penetrate with different depth inside the tail region of the selfassembly, thus inducing different response. Time-resolved fluorescence of Py and its derivatives in cyclohexane and buffer was measured under air-saturated conditions. The transients are shown in Figure 4 and the

Solvent

Probe

λdetection(nm)

Cyclohexane

Py Py-C4 Py-C8 Py Py-C4 Py-C8

≥380 ≥380 ≥380 ≥495 ≥495 ≥495

pH 7.2 buffer

τriseb

τdecayc

8.64 (−0.35) 1.12 (−0.10) 2.65 (−0.05)

16.9 17.4 18.6 20.6 (1) 25.5 (1) 24.0 (1)

Emission was detected using a 380 nm and a 495 nm long-pass filters as indicated. λex = 340 nm. Relative contributions are listed in parentheses. b±0.30 ns. c±0.5 ns. a

using two different optical filters as indicated in the caption of Figure 4. In air-saturated cyclohexane, Py shows only one lifetime component of 16.9 ns. This lifetime is assigned to the monomers of Py, which was confirmed by repeating the measurements using a band-pass filter centered at 387 nm with 11 nm bandwidth. The latter produced similar results. The absence of excimers at the current concentration of 50 μM was D

DOI: 10.1021/acs.langmuir.9b01767 Langmuir XXXX, XXX, XXX−XXX

Article

Langmuir confirmed by the absence of any dynamics in the excimer spectral region when using a long-pass filter of 495 nm. Previous work indicates that the fluorescence lifetime of Py is very sensitive to the presence of oxygen in the sample.16 The lifetime in deoxygenated cyclohexane was 382 ns, whereas in air-saturated cyclohexane the lifetime drops to 20 ns. As shown in Table 2, there is a slight increase in the fluorescence lifetime as the chain length increases (17.4 ns for Py-C4 and 18.6 ns for Py-C8). The fluorescence lifetime increase in the presence of the alkyl chain indicates less dynamical quenching by oxygen as a consequence of the reduced mobility of the Py molecule. The fluorescence transient for Py in buffer shows a buildup time of 8.64 ns when probing the excimer spectral region. The rise time refers to the time required for the excimer formation in the excited state. A decay time of 20.6 ns was measured after the excimer formation as shown in Table 2. Both lifetime components (rise and decay) resemble to a large extent those measured for Py crystals,45 indicating that the status of Py in our case is more solid crystalline than dissolved in buffer. A shorter rise time in Py-C4 and Py-C8 (1.12 and 2.65 ns, respectively) reflects the high affinity to form excimers in the presence of the alkyl chains. This is followed by longer decay lifetimes, compared to Py, indicating more stable excimers. The ratio of the negative pre-exponential factor over that of the positive pre-exponential factor for Py in buffer (−0.35/1.0, Table 2) indicates that ground state dimers are present because of Py molecules packed close to each other.45,46 If dimers were formed via diffusion only, this ratio should yield −1.0. The ratio is less negative for Py-C4 and Py-C8. The decrease in the ratio going from pyrene (−0.35) to Py-C4 (−0.10) to Py-C8 (−0.05) shows an inverse correlation with the relative excimer to monomer fluorescence intensity, namely, the IE/IM ratio, where this ratio is largest for Py-C8 and smallest for Py in Figure 2b. A large IE/IM ratio, indicative of efficient excimer formation, should reflect a stronger aggregation of Py derivatives in water, as observed from the pre-exponential ratios. Probing the Hydrophobic Region of Lipids. Due to its hydrophobicity, the Py molecule is expected to favor the tail region of the lipid self-assembly and can be utilized as a local probe to investigate the hydrophobic region of the lipid βMalC12. The local environment around each probe in lipid can be revealed by measuring the I1/I3 intensity ratio of the monomer fluorescence, along with the excimer fluorescence peak. The fluorescence spectra of Py and its derivatives incorporated inside cubic (normal, type I) and lamellar phases of βMal-C12 are shown in Figure 5 for λex = 340 nm. The values of I1/I3 for the two phases of the βMal-C12 lipid are summarized in Table 3. It is clear from Figure 5 that only monomer fluorescence is observed in lipid. The results imply that the probe molecules are distributed among the tails of the lipid assembly and prefer to be isolated from each other, similar to hydrophobic solvation. In the cubic phase, the I1/I3 ratio for Py is ca. 1.13. This value points to a slight polar environment around Py. Molecular dynamics simulations by Hoff and co-workers47 show that Py prefers a position inside the lipid membrane near the headgroups. Therefore, the results indicate that the Py molecules reside close to the polar region, yet buried within the tail region of the lipid. The I1/I3 ratio is close to that measured for Py in ethyl acetate (Table S3, SI), indicative of apolar local environment in the tail region. On the other hand, when Py is replaced by Py-C4 or Py-C8, the I1/I3 ratio was 1.02 for the former and 0.98 for the latter. Given the lower

Figure 5. Fluorescence spectra of Py, Py-C4, and Py-C8 in cubic and lamellar phases of βMal-C12. The corresponding fluorescence spectra in cyclohexane are included for comparison. For ease of comparison, the spectra are normalized at I5. λex = 340 nm.

Table 3. Fluorescence Intensity Ratio of the Vibronic Peaks of Py and Its Derivatives (I1/I3) in βMal-C12a I1/I3 Phase Probe

Cubic

Lamellar

Py Py-C4 Py-C8

1.13 1.02 0.98

0.99 0.86 0.84

λex = 340 nm The values are the averages of at least two measurements.

a

sensitivity of this ratio for the two Py derivatives toward their local environment, the values may not be useful in estimating the local polarity, but can possibly be used to indicate the relative location of the Py moiety in lipid as the chain length increases. The smaller value for Py-C8 may point to deeper penetration of the Py moiety inside the cubic tail region, away from the maltose headgroup, resulting in an increase in the local hydrophobicity. These results are consistent with the E

DOI: 10.1021/acs.langmuir.9b01767 Langmuir XXXX, XXX, XXX−XXX

Article

Langmuir

exposure to the polar side from all directions if the local space is small. In contrast, the lamellar phase self-assembly allows such exposure from one direction only. The unique structure of the cubic phase allows interpenetrating channels of water and lipid that provide compatibility with water-soluble, lipidsoluble, and amphiphilic active ingredients. Fluorescence lifetime measurements were conducted for Py and its derivatives in the cubic and lamellar phases of βMalC12. Figure 6 displays the decay transients, and the lifetime

crystal structure of Py-C8 and its crystal packing (SI, Figures S5−S7) in which the molecules are arranged head-to-tail, minimizing π−π interaction and favoring C−H···π intermolecular interaction. We previously reported a polarity gradient in the headgroup region using tryptophan (Trp) and two of its ester derivatives (Trp-C4 and Trp-C8) as probes, and we observed that Trp and Trp-C4 reside slightly away from the maltoside sugar units, and the local polarity is similar to that of simple alcohols (methanol and ethanol). For Trp-C8, the long chain length pulls the Trp moiety closer to the headgroups and the local polarity approaches that of 1,4-dioxane.24 By integrating these observations with the current hydrophobicity gradient in the tail region, it is clear that the (I1/I3) value decreases gradually from the polar domain to the hydrophobic domain. This smooth transition in behavior is important for the stability, integrity, and functionality of the lipid as abrupt alteration can affect not only lipid−lipid packing but also the nature of the boundaries in lipid domains.48 The values of I1/I3 for the three probes in the lamellar phase indicate an increased hydrophobic environment as compared to the cubic phase (Table 3). This observation can be correlated to the structural difference between the two selfassemblies. In the lamellar phase, the tail region of the bilayer unit provides a large space for the probes to penetrate freely inside the hydrophobic region. In contrast, the curved structure of the cubic phase limits the available space for the probes. The temperature-induced phase transition (cubic to lamellar) is completely reversible as the I1/I3 values after cooling the samples to 23 °C match those measured for the cubic phase before heating. This is expected because the cubic phase is a curved lamellar phase. The results show that the internal structure of the glycolipid−water system is a thermodynamic equilibrium structure that depends only on the current temperature, regardless of whether it was reached upon heating or cooling.49 We compare the current results with our previous results of Py in the glucoside lipid βGlc-C8,25 and we start first with the cubic phase in each lipid. In βGlc-C8, the I1/I3 ratio was 0.93, which is somewhat less polar than in βMal-C12. The slightly more polar environment when Py was probed in βMal-C12 can be correlated to the longer carbon chain length of the tail part. A longer C12-tail in βMal-C12 is expected to have a more random and wobbling motion than C8 in βGlc-C8 which leads to less interaction with the Py molecule (less solvation effect) in the former. The estimated length of the Py molecule (carbon-to-carbon distance) is ca. 7.0 Å. The difference in polarity must then arise from a more compact interaction between Py and a C8-tail (ca. 9.0 Å in length) compared to the interaction with a C12-tail (ca. 14.0 Å).25 We also reported more local hydrophobic effect for Py in the inverse cubic phase (type II) of the βGlc-C10C6 lipid (I1/I3 = 0.78).26 The increase in hydrophobicity (compared to βMal-C12 and βGlc-C8) can be attributed to the increased hydrophobic environment provided by the double alkyl chains of βGlc-C10C6, leading to further increase in the Py local hydrophobicity inside a glucose based lipid.26 For phase transformation from cubic to lamellar, the current results for Py in βMal-C12 show more hydrophobicity in the latter phase. We also reported a slight reduction in the I1/I3 ratio for Py in βGlc-C8 (0.93 for the cubic phase and 0.87 for the lamellar phase).25 In the cubic phase, its interface is curved toward the bulk water and hence gives Py the possibility of

Figure 6. Fluorescence decay transients of Py and its derivatives (color coded) after excitation at 340 nm in (a) cubic phase and (b) lamellar phase. IRF is shown as a dashed line. Black solid lines represent the best fits. The insets show the corresponding decay transients measured for a longer time range. Detection was at 380 nm.

values are summarized in Table 4. We did not detect any rise time when probing in the excimer spectral region (≥495 nm), which confirms the absence of dimers in the lipid assembly. Three decay components were measured in both lipidic phases. The three lifetime components in each phase show a short component (0.30−0.52 ns), an intermediate component (3.6−6.2 ns), and a long component (31.0−57.5 ns). The presence of three decay components reflects the heterogeneity in the local environment of the Py moiety inside the hydrophobic region. This heterogeneity is important for the biological function of the lipid assembly, giving the lipid a degree of flexibility that is necessary to accommodate various molecular sizes. We also observed heterogeneity in the tail region of other lipids using Py as a probe.24−26 More insight into the origin of this heterogeneity may be obtained by F

DOI: 10.1021/acs.langmuir.9b01767 Langmuir XXXX, XXX, XXX−XXX

Article

Langmuir Table 4. Fluorescence Lifetime (ns) Data of Py and Its Derivatives in Lipidsa Cubic

Lamellar

Probe

τ1b

τ2c

τ3d

⟨τ⟩

τ1b

τ2c

τ3d

⟨τ⟩

Py Py-C4 Py-C8

0.52 (0.19) 0.40 (0.28) 0.30 (0.30)

4.1 (0.33) 3.6 (0.30) 4.1 (0.23)

54.8 (0.48) 57.5 (0.42) 48.6 (0.47)

27.7 25.3 23.9

0.34 (0.06) 0.50 (0.27) 0.45 (0.20)

3.8 (0.28) 4.7 (0.22) 6.2 (0.18)

53.0 (0.66) 39.5 (0.51) 31.0 (0.62)

36.1 21.3 20.4

Relative contributions are listed in parentheses. λex = 340 nm and λdetection = 380 nm. b±0.08 ns. c±1.0 ns. d±0.5 ns.

a

conducting molecular dynamics simulations on the Py/lipid complexes. The lifetime values in Table 4 indicate one major difference between the dynamics of Py and its derivatives in the two phases. In the cubic phase, the presence of C4 and C8 chains shows only a small effect on the three lifetime components as indicated by the value of the average lifetime ⟨τ⟩ (dropped by ∼9% for Py-C4 and ∼14% for Py-C8 from that of Py). On the other hand, ⟨τ⟩ in the lamellar phase is reduced by ∼41% and 43% for Py-C4 and Py-C8, respectively, from that of Py. It is also clear that this reduction in lifetime in the lamellar phase is driven mainly by the long lifetime component which is reduced from 53.0 ns for Py to 39.5 ns (25%) and 31.0 ns (45%) for Py-C4 and Py-C8, respectively. The long decay window in the insets in Figure 6 reflects this effect. This large drop must correlate to the dynamics of the probe within the hydrophobic region of the lamellar phase. The hydrophobic effect on the spectroscopy of Py is more pronounced in the lifetime values than for the steady-state I1/I3 values (see Table 3). This observation suggests the presence of a mechanism of interaction more collisional than static between the Py moiety and the tail region of the lamellar phase, as the probe is buried deeper inside the hydrophobic region. The less pronounced effect in the cubic phase indicates a tighter environment around the Py probes resulting from the increased curvature of the cubic phase. In contrast, the tail region of the bilayer unit in the lamellar phase provides ample space for the probes, thus allowing more freedom for the enclosed molecules to move as the probes are buried deeper inside the tail region. This dynamic mechanism of interaction may be important for several biological functions such as the process of membrane fusion that involves lipid bilayers of different phases, including the lamellar phase.50 A schematic diagram showing the effect of the different geometries of the cubic and lamellar phases on the probes is shown in Figure 7. The new probes’ low sensitivity to solvent polarity may be improved by modifying the alkyl chain at the 2-position. In a related work, Farhangi and Duhamel replaced a methylene group in 1-pyrenebutanol by an oxygen atom (1-pyrenemethoxyethanol) which restored the low sensitivity to that of the parent Py.44 A similar modification to the current probes may also be useful.



Figure 7. Schematic diagram showing the effect of the alkyl chain tail attached to Py on its location in the tail region of the cubic phase (top) and the lamellar phase (bottom).

I3 of Py indicate that Py monomers are dispersed in the hydrophobic region, yet reside close to the maltose headgroups. When a C4 or C8 alkyl chain is attached to Py, there is a slight increase in hydrophobicity (reduction in I1/I3) as the chain length increases, indicating the effect of the nonpolar alkyl chain in pulling the Py moiety deeper inside the tail region. The overall hydrophobicity is observed to be greater for the lamellar phase, which provides more space for the probes to penetrate freely inside the hydrophobic region, whereas the curved structure of the cubic phase limits the available space for the probes. Fluorescence lifetime measurements yielded three lifetime components for each probe in lipid, which shows a degree of heterogeneity in the probe’s local environment. In the lamellar phase, a large reduction in the average lifetime value was measured for Py-C4 (reduction by 25%) and Py-C8 (45%) compared to that of the parent Py. This observation can be explained by the dominant role of collisional mechanism of interaction, compared to static interaction, between the Py moiety and the tail region of the bilayer unit due to the ample space provided by the lamellar phase as the probe is buried deeper inside the hydrophobic region. In the cubic phase, this effect is smaller due to the tight environment around the probe which stems from the increased curvature of the cubic phase.

CONCLUSIONS

The fluorescence behavior of Py, Py-C4, and Py-C8 in the hydrophobic region of βMal-C12 self-assembly was characterized in the cubic and lamellar phases in order to gain a deeper understanding of the nature of the tail region during thermotropic phase transition. Although the sensitivity to local environment is reduced, attaching an alkyl chain (C4 or C8) at the 2-position of Py provides a means to probe the local hydrophobicity in different parts of the tail region. The absence of excimer fluorescence in lipid and the vibronic peak ratio I1/ G

DOI: 10.1021/acs.langmuir.9b01767 Langmuir XXXX, XXX, XXX−XXX

Article

Langmuir

(9) Koeltzow, D. E.; Urefer, A. D. Preparation and Properties of Pure Alkyl Glucosides, Maltosides and Maltotriosides. J. Am. Oil Chem. Soc. 1984, 61, 1651−1655. (10) Vill, V.; Böcker, T.; Thiem, J.; Fischer, F. Studies on LiquidCrystalline Glycosides. Liq. Cryst. 1989, 6, 349−356. (11) Seddon, J. M.; Templer, R. H. Polymorphism of Lipid-Water Systems. In Structure and Dynamics of Membranes, Lipowsky, R.; Sackmann, E., Eds.; Science B. V.: Amsterdam, 1995; Vol. 1, pp 97− 160. (12) Sadeghpour, A.; Rappolt, M. Lyotropic Liquid Crystalline Phases for the Formulation of Future Functional Foods. J. Nutr. Health Food Eng. 2016, 5, 00157. (13) Weiskopf, D.; Schmitt, E. K.; Klühr, M. H.; Dertinger, S. K.; Steinem, C. Micro-BLMs on Highly Ordered Porous Silicon Substrates: Rupture Process and Lateral Mobility. Langmuir 2007, 23, 9134−9139. (14) Chattopadhyay, A. Exploring Membrane Organization and Dynamics by the Wavelength-Selective Fluorescence Approach. Chem. Phys. Lipids 2003, 122, 3−17. (15) Somerharju, P. Pyrene-Labeled Lipids as Tools in Membrane Biophysics and Cell Biology. Chem. Phys. Lipids 2002, 116, 57−74. (16) Karpovich, D. S.; Blanchard, G. J. Relating the PolarityDependent Fluorescence Response of Pyrene to Vibronic Coupling. Achieving A Fundamental Understanding of the Py Polarity Scale. J. Phys. Chem. 1995, 99, 3951−3958. (17) Sahoo, D.; Narayanaswami, V.; Kay, C. M.; Ryan, R. O. Pyrene Excimer Fluorescence: A Spatially Sensitive Probe to Monitor Lipidinduced Helical Rearrangement of Apolipophorin III. Biochemistry 2000, 39, 6594−6601. (18) Haque, M. E.; Ray, S.; Chakrabarti, A. Polarity Estimate of the Hydrophobic Binding Sites in Erythroid Spectrin: A Study by Pyrene Fluorescence. J. Fluoresc. 2000, 10, 1−6. (19) Yekta, A.; Duhamel, J.; Brochard, P.; Adiwidjaja, H.; Winnik, M. A. A Fluorescent Probe Study of Micelle-Like Cluster Formation in Aqueous Solutions of Hydrophobically Modified Poly(ethylene oxide). Macromolecules 1993, 26, 1829−1836. (20) Chaudhuri, A.; Haldar, S.; Chattopadhyay, A. Organization and Dynamics in Micellar Structural Transition Monitored by Pyrene Fluorescence. Biochem. Biophys. Res. Commun. 2009, 390, 728−732. (21) Raimbault, J.; Casier, R.; Little, H.; Duhamel, J. Hydrophobic and Elastic Forces Experienced by a Series of Pyrene End-Labeled Poly(ethylene oxide)s Interacting with Sodium Dodecyl Sulfate Micelles. Macromolecules 2018, 51, 5933−5943. (22) Möller, M. N.; Li, Q.; Chinnaraj, M.; Cheung, H. C.; Lancaster, J. R.; Denicola, A. Solubility and Diffusion of Oxygen in Phospholipid Membranes. Biochim. Biophys. Acta, Biomembr. 2016, 1858, 2923− 2930. (23) Cheng, S.; Thomas, J.; Kulpa, C. Dynamics of Pyrene Fluorescence in Escherichia coli Membrane Vesicles. Biochemistry 1974, 13, 1135−1139. (24) Zahid, N. I.; Abou-Zied, O. K.; Hashim, R.; Heidelberg, T. Characterization of the Head Group and the Hydrophobic Regions of a Glycolipid Lyotropic Hexagonal Phase Using Fluorescent Probes. J. Phys. Chem. C 2011, 115, 19805−19810. (25) Zahid, N. I.; Abou-Zied, O. K.; Hashim, R.; Heidelberg, T. Fluorescence Probing of the Temperature-Induced Phase Transition in a Glycolipid Self-Assembly: Hexagonal↔ Micellar and Cubic↔ Lamellar. Langmuir 2012, 28, 4989−4995. (26) Zahid, N. I.; Abou-Zied, O. K.; Hashim, R. Evidence of Basic Medium in the Polar Nanochannels of the Inverse Bicontinuous Cubic Phase of a Guerbet Glycolipid: A Steady-State and TimeResolved Fluorescence Study. J. Phys. Chem. C 2013, 117, 26636− 26643. (27) Crawford, A. G.; Dwyer, A. D.; Liu, Z.; Steffen, A.; Beeby, A.; Pålsson, L.-O.; Tozer, D. J.; Marder, T. B. Experimental and Theoretical Studies of the Photophysical Properties of 2- and 2,7Functionalized Pyrene Derivatives. J. Am. Chem. Soc. 2011, 133, 13349−13362.

The current results provide a deeper understanding of the local environment of the hydrophobic region during phase transformation of lipid self-assembly which is important for membrane-protein crystallization.



ASSOCIATED CONTENT

S Supporting Information *

The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acs.langmuir.9b01767.



Materials and Methods, including analysis and characterization (PDF)

AUTHOR INFORMATION

Corresponding Author

*E-mail: [email protected]. Tel: (+968) 2414-1468. Fax: (+968) 2414-1469. ORCID

N. Idayu Zahid: 0000-0002-4912-0850 Todd B. Marder: 0000-0002-9990-0169 Osama K. Abou-Zied: 0000-0003-0497-8412 Notes

The authors declare no competing financial interest.



ACKNOWLEDGMENTS This work was supported by Sultan Qaboos University through His Majesty’s Trust Fund for Strategic Research (SR/SCI/ CHEM/18/01); University of Malaya, Ministry of Higher Education High Impact Research Grant (UM.C/625/1/HIR/ MOHE/05), Ministry of Education FRGS (FP069-2018A), and the Julius-Maximilians-Ü niverstät Würzburg. We thank Kevin Erdle for assistance with the synthesis of Py-C4 and PyC8.



REFERENCES

(1) Parker, J. L.; Newstead, S. Membrane Protein Crystallisation: Current Trends and Future Perspectives. In The Next Generation in Membrane Protein Structure Determination, Moraes, I., Ed.; Springer International Publishing: Switzerland, 2016; pp 61−72. (2) Breibeck, J.; Rompel, A. Successful Amphiphiles as the Key to Crystallization of Membrane Proteins: Bridging Theory and Practice. Biochim. Biophys. Acta, Gen. Subj. 2019, 1863, 437−455. (3) Seddon, A. M.; Curnow, P.; Booth, P. J. Membrane Proteins, Lipids and Detergents: Not Just a Soap Opera. Biochim. Biophys. Acta, Biomembr. 2004, 1666, 105−117. (4) Caffrey, M.; Cherezov, V. Crystallizing Membrane Proteins Using Lipidic Mesophases. Nat. Protoc. 2009, 4, 706−731. (5) Bujarski, J. J.; Hardy, S. F.; Miller, W. A.; Hall, T. C. Use of Dodecyl-β-D-Maltoside in the Purification and Stabilization of RNA Polymerase from Brome Mosaic Virus-Infected Barley. Virology 1982, 119, 465−473. (6) Sasaki, T.; Demura, M.; Kato, N.; Mukai, Y. Sensitive Detection of Protein-Lipid Interaction Change on Bacteriorhodopsin Using Dodecyl-β-D-Maltoside. Biochemistry 2011, 50, 2283−2290. (7) Lambert, O.; Levy, D.; Ranck, J. L.; Leblanc, G.; Rigaud, J. L. A New “Gel-Like” Phase in Dodecyl Maltoside-Lipid Mixtures: Implications in Solubilization and Reconstitution Studies. Biophys. J. 1998, 74, 918−930. (8) von Minden, H. M.; Brandenburg, K.; Seydel, U.; Koch, M. H. J.; Garamus, V.; Willumeit, R.; Vill, V. Thermotropic and Lyotropic Properties of Long Chain Alkyl Glycopyranosides. Part II: Disaccharides Head Groups. Chem. Phys. Lipids 2000, 106, 157−179. H

DOI: 10.1021/acs.langmuir.9b01767 Langmuir XXXX, XXX, XXX−XXX

Article

Langmuir (28) Crawford, A. G.; Liu, Z.; Mkhalid, I. A.; Thibault, M. H.; Schwarz, N.; Alcaraz, G.; Steffen, A.; Collings, J. C.; Batsanov, A. S.; Howard, J. A.; Marder, T. B. Synthesis of 2-and 2, 7-Functionalized Pyrene Derivatives: An Application of Selective C-H Borylation. Chem. - Eur. J. 2012, 18, 5022−5035. (29) Abou-Zied, O. K.; Zahid, N. I.; Khyasudeen, M. F.; Giera, D. S.; Thimm, J. C.; Hashim, R. Detecting Local Heterogeneity and Ionization Ability in the Head Group Region of Different Lipidic Phases Using Modified Fluorescent Probes. Sci. Rep. 2015, 5, 8699. (30) Zahid, N. I.; Abou-Zied, O. K.; Saari, N. A. N.; Hashim, R. Comparative Study of the Inverse versus Normal Bicontinuous Cubic Phases of the β-D-Glucopyranoside Water-Driven Self-Assemblies Using Fluorescent Probes. RSC Adv. 2016, 6, 227−235. (31) Auvray, X.; Petipas, C.; Dupuy, C.; Louvet, S.; Anthore, R.; Rico-Lattes, I.; Lattes, A. Small-Angle X-Ray Diffraction Study of the Thermotropic and Lyotropic Phases of Five Alkyl Cyclic and Acyclic Disaccharides: Influence of the Linkage between the Hydrophilic and Hydrophobic Moieties. Eur. Phys. J. E: Soft Matter Biol. Phys. 2001, 4, 489−504. (32) Abou-Zied, O. K. Examining [2,2′-bipyridyl]-3,3′-diol as a Possible DNA Model Base Pair. J. Photochem. Photobiol., A 2006, 182, 192−201. (33) Abou-Zied, O. K. The Role of Water in Solvating the Hydrogen-Bonding Center of 2-(2′-hydroxyphenyl)benzoxazole. Chem. Phys. 2007, 337, 1−10. (34) James, D. R.; Siemiarczuk, A.; Ware, W. R. Stroboscopic Optical Boxcar Technique for the Determination of Fluorescence Lifetimes. Rev. Sci. Instrum. 1992, 63, 1710−1716. (35) Zahid, N. I.; Mahmood, M. S.; Subramanian, B.; Mohd Said, S.; Abou-Zied, O. K. New Insight into the Origin of the Red/NearInfrared Intense Fluorescence of a Crystalline 2-Hydroxychalcone Derivative: A Comprehensive Picture from the Excited-State Femtosecond Dynamics. J. Phys. Chem. Lett. 2017, 8, 5603−5608. (36) Lakowicz, J. Principles of Fluorescence Spectroscopy, 3rd ed.; Springer Verlag: New York, 2006; Vol. 1. p 954. (37) Kinnunen, P. K. J.; Koiv, A.; Mustonen, P. Pyrene-Labeled Lipids as Fluorescent Probes in Studies on Biomembranes and Membrane Models. In Fluorescence Spectroscopy: New Methods and Applications, Wolfbeis, O. S., Ed.; Springer-Verlag: New York, 1993; pp 159−171. (38) Saxena, R.; Shrivastava, S.; Chattopadhyay, A. Exploring the Organization and Dynamics of Hippocampal Membranes Utilizing Pyrene Fluorescence. J. Phys. Chem. B 2008, 112, 12134−12138. (39) Ioffe, V.; Gorbenko, G. P. Lysozyme Effect on Structural State of Model Membranes as Revealed by Pyrene Excimerization Studies. Biophys. Chem. 2005, 114, 199−204. (40) Bains, G. K.; Kim, S. H.; Sorin, E. J.; Narayanaswami, V. The Extent of Pyrene Excimer Fluorescence Emission is a Reflector of Distance and Flexibility: Analysis of the Segment Linking the LDL Receptor-Binding and Tetramerization Domains of Apolipoprotein E3. Biochemistry 2012, 51, 6207−6219. (41) Dong, D. C.; Winnik, M. A. The Py Scale of Solvent Polarities: Solvent Effects on the Vibronic Fine Structure of Pyrene Fluorescence and Empirical Correlations with ET and Y Values. Photochem. Photobiol. 1982, 35, 17−21. (42) Dong, D. C.; Winnik, M. A. The Py Scale of Solvent Polarities. Can. J. Chem. 1984, 62, 2560−2565. (43) Reichardt, C. Solvents and Solvent Effects in Organic Chemistry, 3rd ed.; Wiley-VCH: Weinheim, 2003. (44) Farhangi, S.; Duhamel, J. Pyrenyl Derivative with a Four-Atom Linker That Can Probe the Local Polarity of Pyrene-Labeled Macromolecules. J. Phys. Chem. B 2016, 120, 834−842. (45) Keyes-Baig, C.; Duhamel, J.; Fung, S.-Y.; Bezaire, J.; Chen, P. Self-Assembling Peptide as a Potential Carrier of Hydrophobic Compounds. J. Am. Chem. Soc. 2004, 126, 7522−7532. (46) Duhamel, J. Global Analysis of Fluorescence Decays to Probe the Internal Dynamics of Fluorescently Labeled Macromolecules. Langmuir 2014, 30, 2307−2324.

(47) Hoff, B.; Strandberg, E.; Ulrich, A. S.; Tieleman, D. P.; Posten, C. 2H-NMR Study and Molecular Dynamics Simulation of the Location, Alignment, and Mobility of Pyrene in POPC Bilayers. Biophys. J. 2005, 88, 1818−1827. (48) García-Sáez, A. J.; Schwille, P. Stability of Lipid Domains. FEBS Lett. 2010, 584, 1653−1658. (49) de Campo, L.; Yaghmur, A.; Sagalowicz, L.; Leser, M. E.; Watzke, H.; Glatter, O. Reversible Phase Transitions in Emulsified Nanostructured Lipid Systems. Langmuir 2004, 20, 5254−5261. (50) Chernomordik, L. V.; Kozlov, M. M. Protein-Lipid Interplay in Fusion and Fission of Biological Membranes. Annu. Rev. Biochem. 2003, 72, 175−207.

I

DOI: 10.1021/acs.langmuir.9b01767 Langmuir XXXX, XXX, XXX−XXX