Spectroscopic Evidence for Self-Organization of N

Article pubs.acs.org/JPCB

Spectroscopic Evidence for Self-Organization of N‑Iodoacetylamphotericin B in Crystalline and Amorphous Phases Mariusz Gagoś,*,†,‡ Daniel Kamiński,§ Marta Arczewska,† Bartosz Krajnik,∥ and Sebastian Maćkowski∥ †

Department of Biophysics, University of Life Sciences in Lublin, Akademicka 13, 20-950 Lublin, Poland Department of Cell Biology, Institute of Biology and Biochemistry, Maria Curie-Skłodowska University, 20-033 Lublin, Poland § Department of Chemistry, University of Life Sciences in Lublin, Akademicka 15, 20-950 Lublin, Poland ∥ Optics of Hybrid Nanostructures Group, Institute of Physics, Nicolaus Copernicus University, Grudziadzka 5, 87-100 Toruń, Poland ‡

ABSTRACT: In this paper, we propose a new way of thinking about molecular self-organization of the antibiotic amphotericin B (AmB) by examination of its Niodoacetyl derivative (AmB-I). This choice was dictated by the simplicity of AmB-I crystallization as compared to pure AmB. The studies focus on spectroscopic investigations of the monocrystal and the amorphous state of AmB-I. The results of vibrational, FTIR, and Raman spectroscopy show differences between the crystalline and amorphous forms, in particular for bands attributed to CO (1700−1730 cm−1) and CCH groups, as well as CCC (ca. 1010 cm−1) stretching vibrations. The process of crystallization is identified by strong differences in the intensities and locations of these characteristic bands. For the AmB-I crystals, the carbonyl band is shifted toward lower frequencies as a result of intensified hydrogen bonding in the crystalline form. Detailed analysis indicates that bands in the region characteristic for the CCC bending distortion in the chromophore are particularly intense for AmB-I in the crystalline form as compared to the intensity of this band in the amorphous state. These findings are corroborated by the results of fluorescence spectroscopy. We observe a much faster decay of the emission for the AmB-I monocrystal as compared to the DMSO solution of AmB-I. Interestingly, the fluorescence decay in the amorphous form requires three decay times for simulating the observed behavior; two of these decay constants are sufficient for estimating the decay measured for the AmB-I crystals. The proof of the molecular organization of AmB-I molecules is obtained from polarization-resolved fluorescence spectroscopy on a single AmB-I crystal. Strong anisotropy of the emission intensity correlates with the axes of the crystal, providing insight into actual alignment of the molecules in the AmB-I crystals. These findings related to molecular organization in AmB-I crystals are crucial for understanding toxicity mechanisms of the clinically used drug, amphotericin B.

■

INTRODUCTION Amphotericin B (AmB), isolated half a century ago1 from Streptomyces nodosus, remains the gold standard in the treatment of serious systemic mycoses despite its nephrotoxicity and adverse effects.2,3 During the last decades, an alarming rise in life-threatening fungal infections due to the emergence of opportunistic pathogens forced the development of new drugs and new formulations of old antifungal agents.4 Invasive fungal infections are an increasing complication for immunecompromised individuals such as AIDS, cancer, and transplant patients.3,5 Although the stereochemical configuration of the Niodoacetyl derivative of AmB (see Scheme 1) was elucidated by Ganis et al.6 in 1971, we were able to crystallize single AmB-I crystals of much better structural quality, as evidenced by X-ray diffraction data.7 While not being clinically used, AmB-I is considered as a biologically active component analogous to AmB.6 The shape of the molecule and spatial positioning of the antibiotic’s major functional groups plays an important role in the aggregation processes.8−11 In particular, it is often assumed that the toxicity of AmB in biological systems is directly © 2012 American Chemical Society

connected with the formation of associated forms of molecular aggregates.12,13 The qualitative model of AmB pores in biological membrane proposed by de Krujiff et al.14 and Andreoli15 is still commonly accepted, despite its simplicity, which allows simulation of the channel properties at the molecular level.10,16 In this work, we demonstrate a thorough spectroscopic characterization of the optical properties and molecular orientation in AmB-I crystals. AmB-I, a heavy atom derivative of AmB, efficiently crystallizes from tetrahydrofuran. In order to comprehensively characterize the spectroscopic properties of these high-quality crystals, we combine FTIR, Raman, and fluorescence spectroscopies. It has been recently reported that with Raman microspectroscopy it is possible to detect single crystals of β-carotene at the subcellular level directly from a plant cell.17 Received: August 8, 2012 Revised: September 27, 2012 Published: October 3, 2012 12706

dx.doi.org/10.1021/jp307873m | J. Phys. Chem. B 2012, 116, 12706−12713

The Journal of Physical Chemistry B

Article

Scheme 1. Chemical Structure of N-Iodoacetylamphotericin B

Figure 1. (A) AmB-I crystals used in experiments. (B) Amorphous phase of AmB-I obtained by DMSO evaporation (N2 stream) from quartz plate. This phase is transparent for light. The cracks are an effect of a solvent evaporation.

■

MATERIALS AND METHODS Synthesis, crystallization, and sample preparation were described in detail previously.7 The amorphous phase was prepared by DMSO evaporation from AmB-DMSO saturated solution on the ATR crystal. The presence of the amorphous phase from DMSO was confirmed by powder X-ray diffraction, where no diffraction signal was observed. The AmB-I crystals are shown in Figure 1. Spectroscopic Measurements. FTIR spectra were recorded with the 670-IR spectrometer (Varian). The attenuated total reflection (ATR) configuration was used with 10 internal reflections of the HATR Ge crystal plate (45° cut). Typically, 25 scans were collected, Fourier-transformed, and averaged for each measurement. Absorption spectra at a resolution of one data point per 1 cm−1 were obtained for the spectral region from 4000 to 600 cm−1. The sample chamber was continuously purged with N2 for 40 min before and during measurements. The Ge crystal plate was cleaned with ultrapure organic solvents (methanol, chloroform, and acetone) from Sigma-Aldrich. All experiments were carried out at room temperature. Raman spectra were recorded with a Raman microscope inVia Reflex from Renishaw (UK), which used a chargecoupled device (CCD) detector with a spectral resolution of 1 cm−1 for detection. The excitation wavelength for all the measurements was a 514.5 nm line of an Ar+ laser, while the excitation power varied between 0.4 and 20 mW. The laser

The results provide a solid basis for reconsideration of the molecular self-organization of AmB-I. We find that the molecular organization of AmB-I is not only assigned to the well-defined phenomenon of aggregation. This is demonstrated by differences between the amorphous and crystalline forms in both the vibrational spectra and transient behavior of fluorescence emission. The analysis of FTIR and Raman spectra yields substantial changes in the regions of CO and CCC stretching modes, thus proving useful for investigating the structural modifications during the conversion from the amorphous to the crystalline phase. These changes in the spectroscopic characteristics which appear exclusively in the crystalline forms of AmB-I may be considered as a fingerprint for formation of periodic structures in biological lipid membranes. By the analogy, we anticipate that the similar structures should be observed for clinically used AmB. The results of fluorescence spectroscopy give further insight into the molecular organization in AmB-I crystals. Polarizationresolved fluorescence measured for single crystals indicates clearly that AmB-I molecules are arranged perpendicularly to the axis of the crystal. We also find substantial differences in the fluorescence transient behavior between the crystalline and amorphous forms that can be employed for detection of crystalline formation in biologically relevant membrane environments. 12707



Article

beam diameter was defocused to 20 μm in order to minimize any possible structural damage of the crystals (monitored by the higher frequency shift of the band centered at 1558 cm−1). The spectra were accumulated for 10 s. A test for possible laser damage was performed for every spectrum. All spectra were recorded at least four times, and all were reproducible for the spectral effects described in this work. The preresonance Raman scattering spectra were corrected by subtracting a background signal originating from the fluorescence of the system. Spectral analysis was performed with Grams/AI software from ThermoGalactic Industries (USA). Fluorescence spectroscopy was carried out using a homebuilt confocal fluorescence microscope based on the Olympus infinity-corrected microscope objective LMPlan 50×, characterized with a numerical aperture of 0.5 and working distance of 6 mm.18 The samples were either placed as solutions in cuvettes or deposited on clean coverslides. White light illumination coupled with a CCD camera was used to identify locations of the AmB crystals on the coverslide surface. For excitation of steady-state fluorescence, we used 355 or 405 nm lasers. The 405 nm laser was also used in a pulsed mode, which enabled us to measure fluorescence transients. We used a laser excitation power of about 40 μW. The fluorescence was detected in a backscattering geometry with the emission of AmB extracted by appropriate filter sets (610/40, Chroma Technology). The fluorescence spectra were measured with a CCD detector (Andor iDus DV 420A-BV) coupled with an Amici prism. The spectral resolution of the system is about 2 nm. Polarization of the emission was obtained by fixing the polarization of the laser beam either along or perpendicular to the axis of the AmB crystal using a Babine−Soleil compensator and linear polarizer. The same set was applied for analyzing the polarization of the emission. The emission intensity was measured with an avalanche photodiode (PerkinElmer SPCMAQRH-14). Fluorescence lifetimes were determined using a time-correlated single photon counting module (SPC 150 Becker & Hickl) equipped with a fast avalanche photodiode (idQuantique id100-50) triggered by a laser pulse with a 50 MHz repetition rate. The time resolution of the TCSPC setup is about 30 ps.

Figure 2A presents the ATR-FTIR spectra of AmB-I crystals (shown in Figure 1A) deposited adhesively on the Ge ATR.

Figure 2. ATR-FTIR spectra of (A) AmB-I in crystals and its amorphous phase dissolved in DMSO and then evaporated by N2. (B) AmB-I crystals after CH3Cl evaporation and dissolved in CH3Cl ethanol in ratio 5:1. Measurements were carried out on the HATR Ge crystal plate.

■

RESULTS AND DISCUSSION Vibrational Spectroscopy. FTIR and Raman spectroscopic studies allow determination of differences between the monocrystal and the amorphous state of AmB-I. Due to its high sensitivity and precision, vibrational spectroscopy has been a method of choice for the present research, and has been frequently used in studies focusing on molecular organization in biological systems.19−22 ATR-FTIR spectra of pure AmB dissolved in tetrachloromethane, aqueous solutions at different pH values, and lipid membranes were described previously.11,21 It has been clearly shown that the environment has a tremendous influence upon the molecular organization of AmB, leading to pronounced changes of the shape of vibrational spectra. In polar solvents, the hydrogen bonds between molecules are dominant, while in nonpolar solvents the most possible are the van der Waals interactions. On the basis of the solvent properties, different kinds of interactions between AmB molecules in dimers were discussed.23 It is worth noting that FTIR spectra of AmB-I monocrystals have never been published, but they exhibit the well-known spectral features in the polar environments.

The dashed line represents the spectrum measured for the AmB-I crystals dissolved in DMSO and purged by a N2 stream. By analyzing the FTIR spectra, we find that the changes are present over the whole spectrum. Detailed analysis and discussion of all bands was given in our previous works for commercial AmB.11,21,24 Here we focus, in particular, on the bands within the region 1000−1100 cm−1. The characteristic frequencies of the FTIR spectra are compiled in Table 1. The band centered at 1010 cm−1, assigned to the CC vibrations of the CCH group coupled with the CCC bending distortion in the chromophore of AmB,20 is important because of its appearance both in FTIR and Raman spectra of AmB and other polyenes.20,21,24−31 This band has been previously observed in different environments (KBr and NaBr), and its intensity increased at high temperature, low pH values, and during compression of monomolecular layers and lipid multibilayers with AmB.21 The presence of this band might be assigned to the rigid conformation of AmB molecules that the deformation of the CCH vibrations causing the geometrical changes in the chromophore.20 Interestingly, the intensity of the 1010 cm−1 band strongly increases in 12708



Article

Table 1. The Position of FTIR Vibrations in the AmB-I Crystals and Their Amorphous Form Dissolved in DMSO and Chloroform (as Solid Chlsolid and Liquid Films Chlliquid)a wavenumber (cm−1)aa FTIR

Raman

crystal

DMSO

Chlsolid

Chlliquid

*crystal

assignment*

1010s

1010w 1042w 1074w 1111 1166w 1185w

1010s 1039s 1065s 1110s 1130w 1175 1267

1010w 1046w 1074w 1104w 1128s 1161s

1008w

δ(CCH) for chromophore, γ(CH) in trans-polyene, γ(CCC)* γ(NH2), νs(COC) pyranose ring νas(CO)

1156s 1199 1293

ν(CC)* νas(COC) for β-glycosidic linkage, (COCO) ν(COC) for ester + δ(OH) δ(CH2)

1558s

δ(CH3) + δ(OH), νs(COO−), δ(CH) δas(CH2, CH3) δs(NH3+) + νas(COO−), νs(CC)*

1065s 1109 1166s 1183s 1231 1265s

1265w

1315 1377 1448 1551

1315 1377 1459 1553

1644

1654

1710

1725

3400

1297 1338 1315 1375 1455 1554 1566 1652

1710 2800−3000 3385 3386

1381 1459 1550 1569 1651

1599 1633

1710 3386

νas(CC)* νas(CO) in ester, ν(CO) for saturated carboxylic group νs+as(CH2, CH3) + ν(CH) in polyene ν(OH), ν(NH)

The asterisks indicate Raman modes in the AmB-I crystal (Ex 514.5 nm). Temperature: 23 °C. ν, stretching mode; δ, bending in plane; γ, bending out of plane; s, symmetric vibrations; as, asymmetric vibrations; w, weak shoulder; s, strong shoulder.

a

experiments carried out on the AmB-I solid phase after chloroform evaporation (Figure 2B, solid line). The clear appearance of this band for the AmB-I crystal and its virtual absence in the amorphous phase (see Figure 1B) has to be related to the specific molecular interactions due to the geometrical changes in AmB-I molecules. Other evident spectral changes during AmB-I crystallization are visible in the carbonyl band at 1710−1730 cm−1.32 This band shifts from a relatively broad band centered at 1725 cm−1 to a much stronger and sharper band at 1710 cm−1 (see Figure 2A). In structural terms, these spectral differences for the crystalline phase can be ascribed to the oxygen atoms of the carbonyl group being located closer to the hydrogen atoms of the other molecules.33 This gives the possibility of forming H-bonds, thus reducing the carbonyl bond order reflected in the shift of the CO toward lower frequencies in comparison with the amorphous phase.34 Figure 3 shows the preresonance Raman spectra of AmB-I in the crystalline and amorphous forms. The differences in the spectra of the AmB-I crystals (see the inset of the image from Raman microscopy, Figure 3) and the amorphous form (see Figure 1B) can be seen in the region of the CC vibrations (∼1558 cm−1) in the polyene chain. They are mainly related to broadening of the main band and the slight shift toward lower frequencies (by ∼2 cm−1), which is connected with a slight delocalization of the polyene π electron system. A similar effect was also observed for CeO2, where the amorphous phase yielded a much more broadened band compared to the crystal form.35 The preresonance Raman spectra of AmB-I crystals are similar to our previous results for commercial formulation of AmB in powder.24 In this case, in analogy to the FTIR spectra, we observed changes in the intensity and the spectral shift of the CCH vibrations.24 As can be seen, the intensity of the band centered at about 1008 cm−1 strongly decreases for the amorphous phase; see Figure 3. Bunow et al.20 have observed

Figure 3. Raman spectra of AmB-I in crystalline and amorphous forms. In the right upper corner are presented the images of AmB-I crystals (see Figure 1A) recorded by Raman microscopy. All spectra were normalized with respect to the maximum at 1558 cm−1. Spectra were obtained using the 514.5 excitation wavelength. Samples were deposited at the platinum plate.

the band in the solid phase of commercial AmB, but like in our experiment, its intensity was very weak in the DMSO solution. However, these modifications were not considered as being related to the effect of structural changes from the amorphous to the crystalline phase. Similar observations were reported by Ridente et al.29 and Iqbal et al.,27 where Raman spectra of other polyenes were investigated. This study clearly shows that the CCH vibrations in FTIR and Raman spectra occur not only for well-ordered AmBI crystals but also, which we consider more important, in the case of commercial AmB in powder.24 We conclude that the presence of the iodoacetyl group in AmB-I facilitates the crystallization processes.7 After solvent evaporation, the AmB-I molecules crystallize and the intensity 12709



Article

of the absorbance centered at 1010 cm−1 increases, in contrast with the amorphous phase. Additionally, the monomolecular study revealed that the band in question increased upon compression to the condensed state of commercial AmB.21 In regard to the previous discussion, the vertically oriented AmB molecules (at a surface pressure of 25 mN/m) can interact in a similar way as in the AmB-I crystals. The specific area of 36 Å2 obtained from the monomolecular study is close to the theoretical cross section of AmB calculated on the basis of crystallographic parameters (approximately 6 Å × 7 Å × 24 Å). This is an indication of very efficient packing of the drug in the monomolecular layers.9,36 On the other hand, it is known from the crystal structure of AmB-I that the distance between the chromophores of neighboring molecules is close to 4.0 Å.7 Such a small distance gives the possibility of excitonic interactions between the chromophores, which can result in an increase of the intensity of this band. This is a well-known spectroscopic effect in electronic absorption spectroscopy.37−39 The results of FTIR spectroscopy for AmB-I have shown that the process of crystallization can be monitored by the appearance of the 1010 cm−1 band occurring simultaneously with the shift of the carbonyl band toward lower frequencies. In addition, the presence of the band characteristic for the C CC bending distortion in the chromophore of AmB-I can be considered as a fingerprint of crystal formation and can play an important role in determining the crystal phase in lipid membrane systems. By the analogy between AmB-I and AmB, presented results provide a way to understand the molecular organization of clinically used AmB. Fluorescence Spectroscopy. Figure 4 shows electronic absorption spectra of AmB-I deposited on a quartz plate as a

In Figure 5, we show the fluorescence spectra measured for AmB-I dissolved in DMSO and in the amorphous form. The

Figure 5. Room temperature fluorescence spectra of AmB-I in DMSO (dashed blue line), AmB-I crystal form (solid line), and AmB-I amorphous form (dotted red line). The excitation wavelength was 405 nm.

results are compared with the data obtained for the crystalline AmB-I. The spectra are normalized to the maximum intensity in order to facilitate easier comparison. As can be seen, there are minute differences between the fluorescence bands; they are mainly ascribed to different relative intensities of the subbands forming the spectrum. The results indicate that simple fluorescence spectroscopy is not sufficient for determining the crystal formation of AmB-I, in particular when samples involve biological membranes, which would impose additional spectral features and complicate the analysis. Much more pronounced differences in the fluorescence behavior between various forms of AmB-I are measured for transients in the nanosecond time domain. In Figure 6, we

Figure 4. Electronic absorption spectra of AmB-I film deposited on the quartz plate (see Figure 1B). AmB-I was dissolved in DMSO (dashed line) and the same sample as a dry film (N2 stream, solid line).

Figure 6. Experimental fluorescence decays measured for AmB-I in DMSO (A), AmB-I amorphous form (B), and AmB-I crystal form (C). Circles and lines correspond to data and multiexponential fits, respectively. Residues for all three samples are also displayed. The excitation wavelength was 405 nm with laser pulse repetition of 50 MHz.

liquid and dry film (the same sample as the one presented in Figure 1B). As can be seen, we observe a characteristic spectrum for the monomeric form of AmB-I (the electronic absorption spectrum of commercial AmB is almost identical10) in DMSO solution. After solvent evaporation, the shape of the spectrum is changed and is typical for the well-known aggregated but not crystalline state of AmB.40 For the purpose of the fluorescence study, the laser excitation was 405 nm, which gives a possibility to study the monomeric form of AmBI.

present the result obtained for AmB-I crystals and compare it with transients measured for AmB-I in the DMSO solution and in the amorphous form. In the case of AmB-I molecules dissolved in DMSO, the decay is the slowest and can be described with excellent accuracy using two exponential decay constants (see Table 2). The fluorescence decay measured for AmB-I dissolved in DMSO can be described with two decay times of 2.1 and 0.43 ns, with the faster component being 12710



Article

Table 2. The Fluorescence Decay Times Obtained from Fitting the Transients Measured for Various Forms of AmB-I Together with Respective R2 Values form of AmB-I crystal form amorphous form in DMSO solution

τ1 (ns)

τ2 (ns)

τ3 (ns)

R2

0.94 ± 0.026 0.94

0.21 ± 0.03 0.21

0.71 ± 0.12

0.99843 0.99587

2.12 ± 0.28

0.44 ± 0.05

0.99877

dominant over the slow decay (see Figure 6A), which is similar to commercial AmB samples.40,41 The transient behavior of the fluorescence emission of the AmB-I crystals can also be described with two exponential curves, but the characteristic times are considerably shorter. In this case, the transient behavior can be described using two exponential functions with characteristic times of 0.9 and 0.2 ns (the same sample as the one presented in Figure 1A). Similarly, the faster component dominates the decay (see Figure 6C). The significant shortening of the fluorescence decay observed for the crystalline form of AmB-I as compared to the monomeric form in DMSO can be interpreted to be a result of electronic coupling between individual molecules comprising the crystal. In contrast, the decay curve measured for the amorphous form of AmB-I cannot be reliably expressed as a combination of two single exponential decays. The fit quality is rather poor. Therefore, in order to analyze the transient behavior observed for this structure, we introduced a sum of three exponential decays with two of them being identical with the ones determined for the AmB-I crystal. In this way, we assume that in the amorphous AmB-I sample there are regions where the molecules are strongly coupled in a similar manner as for the crystalline structure. In order to describe the decay curve measured for the amorphous AmB-I, it was necessary to add a decay time, which is in between the two times characteristic for the crystals, and equal to 0.7 ns. We note that for all the fits the quality of estimating the experimentally measured decay curves is remarkably good, with the R2 factor exceeding 0.995 for all the samples. At the same time, although the transient behavior of the AmB-I crystals is somewhat different from the amorphous form and it is plausible these differences may be useful for discriminating between both forms of molecular organization in membranes and biomimetic layers, the procedure would require significant analytical effort. The formation of molecular crystals can be convincingly demonstrated in a polarization-resolved experiment. In Figure 7, we show examples of fluorescence intensity measured for a single AmB-I crystal for both characteristic directions with the excitation varying between 0 and 180°. It is important to ensure that indeed only one crystal is probed at a time, as simultaneous excitation of two or more crystals would result in most cases in mixed and complex behavior of the fluorescence polarization. This has been accomplished by combining a fluorescence measurement with white light illumination. In this way, we are able to place a laser beam exactly on a single, well-isolated crystal. Additionally, what is equally important, we can determine the orientation of the crystal and adjust the polarization of the laser for it to be either parallel or perpendicular to the crystal axis. In Figure 7, we present the values of fluorescence intensity measured for polarization direction along the AmB-I crystal (black squares) and across the long axis of the crystal (red circles) for the varied direction of the excitation polarization.

Figure 7. Fluorescence intensities as a function of the exciting light polarization with the polarizers on the detector parallel (black) and perpendicular (red) to the long axis of the crystal; see details of the AmB-I arrangement in a crystal in Figure 8.

The data obtained for two different crystals are shown, and the crystals themselves are included in Figure 7. Altogether, over 10 crystals were measured and the results are qualitatively similar. Namely, when the excitation is polarized along the AmB-I crystal (the longest axis), the emission intensity reaches a maximum (0°); afterward, as the angle of excitation polarization changes, the intensity decreases until it reaches a minimum for the direction perpendicular to the long axis of the AmB-I crystal. Further, as the angle of excitation polarization starts to approach the long axis of the crystal, the fluorescence intensity recovers its value. The behavior obtained for the emission detected in the direction perpendicular to the axis of the crystal is essentially identical but shifted by 90°. In other words, the minima measured for the excitation polarization along the crystal coincide with the maxima measured for the orthogonal polarization of the excitation. This result shows that the polarization of the excitation is in a large portion preserved in the crystal during the fluorescence lifetime. Furthermore, it indicates that the natural axes of the crystal arrangement are identical with the axes of the crystal itself. An analogous experiment carried out on a solution reveals no polarization selectivity, as the molecules in the monomeric form are randomly oriented and diffused in the solvent; thus, the polarization of the emission is randomized over all directions. It is postulated that AmB-I molecules in the crystal are arranged in layers with a plane perpendicular to the long axis of the crystal seen in Figure 8. Consequently, the dipole moments of the AmB-I molecules are also oriented perpendicularly to the long axis of the crystal. Thus, the significant result emerging from the polarization-resolved experiments on a single crystal indicates a strong interaction between the layers or lines of AmB-I in the direction parallel to the long axis of the crystal. Formation of such a strongly coupled molecular system is a direct consequence of crystallization of AmB-I. On the other hand, for the excitation polarization perpendicular to the crystal axis, we excite dipole moments of the molecules themselves, hence the preferred polarization of the emission. We conclude that the polarization-resolved experiment not only proves the crystal formation of AmB-I molecules and can be effectively used to demonstrate crystallinity in biological membranes as well as fluorescence lifetime, but it also sheds light upon electronic coupling between the molecules in such an assembly.

■

CONCLUSIONS Vibrational spectroscopic studies reveal that the intermolecular interaction patterns varied between the crystalline and 12711



Article

The financial support from the WELCOME program “Hybrid nanostructures as a stepping-stone towards efficient artificial photosynthesis” awarded by the Foundation for Polish Science is gratefully acknowledged.

■

(1) Dutcher, J. D.; Gold, W.; Pagano, J. F.; Vandepatte, J. Amphotericin B, its production and its salts; 2. US patent 2,908,611, 1959. (2) Gold, W.; Stout, H. A.; Pagano, J. F.; Donovick, R. Antibiot. Annu. 1956, 579−585. (3) Hartsel, S.; Bolard, J. Trends Pharmacol. Sci. 1996, 17, 445−449. (4) Ruge, E.; Korting, H. C.; Borelli, C. Int. J. Antimicrob. Agents 2005, 26, 427−441. (5) Zotchev, S. B. Curr. Med. Chem. 2003, 10, 211−223. (6) Ganis, P.; Avitabile, G.; Mechlinski, W.; Schaffner, C. P. J. Am. Chem. Soc. 1971, 93, 4560−4564. (7) Jarzembska, K. N.; Kaminski, D.; Hoser, A. A.; Malinska, M.; Senczyna, B.; Wozniak, K.; Gagos, M. Cryst. Growth Des. 2012, 12, 2336−2345. (8) Baginski, M.; Resat, H.; Borowski, E. Biochim. Biophys. Acta 2002, 1567, 63−78. (9) Bonilla-Marin, M.; Moreno-Bello, M.; Ortega-Blake, I. Biochim. Biophys. Acta 1991, 1061, 65−77. (10) Gagoś, M.; Herec, M.; Arczewska, M.; Czernel, G.; Dalla Serra, M.; Gruszecki, W. I. Biophys. Chem. 2008, 136, 44−49. (11) Gagoś, M.; Arczewska, M. J. Phys. Chem. B 2011, 115, 3185− 3192. (12) Bolard, J. Biochim. Biophys. Acta 1986, 864, 257−304. (13) Bolard, J.; Legrand, P.; Heitz, F.; Cybulska, B. Biochemistry 1991, 30, 5707−5715. (14) De Kruijff, B.; Gerritsen, W. J.; Oerlemans, A.; Demel, R. A.; van Deenen, L. L. Biochim. Biophys. Acta 1974, 339, 30−43. (15) Andreoli, T. E. Ann. N.Y. Acad. Sci. 1974, 235, 448−468. (16) Gagoś, M.; Gabrielska, J.; Dalla Serra, M.; Gruszecki, W. I. Mol. Membr. Biol. 2005, 22, 433−442. (17) Baranska, M.; Baranska, R.; Grzebelus, E.; Roman, M. Vib. Spectrosc. 2011, 56, 166−169. (18) Krajnik, B.; Schulte, T.; Piatkowski, D.; Czechowski, N.; Hofmann, E.; Mackowski, S. Cent. Eur. J. Phys. 2011, 9, 293−299. (19) Weakliem, C. L.; Fujii, G.; Chang, J.-E.; Ben-Shaul, A.; Gelbart, W. M. J. Phys. Chem. 1995, 99, 7694−7697. (20) Bunow, M. R.; Levin, I. W. Biochim. Biophys. Acta 1977, 464, 202−216. (21) Gagoś, M.; Arczewska, M. Biochim. Biophys. Acta 2010, 1798, 2124−2130. (22) Kaushal, A. M.; Chakraborti, A. K.; Bansal, A. K. Mol. Pharm. 2008, 5, 937−945. (23) Caillet, J.; Berges, J.; Langlet, J. Biochim. Biophys. Acta 1995, 1240, 179−195. (24) Gagoś, M.; Arczewska, M.; Gruszecki, W. I. J. Phys. Chem. B 2011, 115, 5032−5036. (25) Espuelas, M. S.; Legrand, P.; Irache, J. M.; Gamazo, C.; Orecchioni, A. M.; Devissaguet, J.-P.; Ygartua, P. Int. J. Pharm. 1997, 158, 19−27. (26) Colline, A.; Bolard, J.; Chinsky, L.; Fang, J. R.; Rinehart, K. L., Jr. J. Antibiot. 1985, 38, 181−185. (27) Iqbal, Z.; Weidekamm, E. Biochim. Biophys. Acta 1979, 555, 426−435. (28) Lewis, E. N.; Kalasinsky, V. F.; Levin, I. W. Anal. Chem. 1988, 60, 2306−2309. (29) Ridente, Y.; Aubard, J.; Bolard, J. Biospectroscopy 1995, 2, 1−8. (30) Rimai, L.; Heyde, M. E.; Gill, D. J. Am. Chem. Soc. 1973, 95, 4493−4501. (31) Schwartzman, G.; Asher, I.; Folen, V.; Brannon, W.; Taylor, J. J. Pharm. Sci. 1978, 67, 398−400. (32) Jarute, G.; Kainz, A.; Schroll, G.; Baena, J. R.; Lendl, B. Anal. Chem. 2004, 76, 6353−6358.

Figure 8. The orientation of AmB-I molecules in the crystal derived from X-ray structural crystallography. AmB-I molecules are arranged in layers with a plane perpendicular to the long axis. The model arrangement of AmB-I molecules based on the structure data created in a program Mercury 2.3.

amorphous phase of AmB-I. Conversion of the amorphous state to the crystalline state leads to changes mainly in the spectral regions attributed to the CO, CCH, and CC vibrations in the polyene chain. The increase in the intensity of the band at 1010 cm−1 observed in the FTIR spectra of AmB-I in the crystallized form was accompanied with the shift of the carbonyl band toward lower wavenumbers. Raman spectroscopy showed that the band intensity centered at around 1010 cm−1 was highly sensitive to the crystalline/amorphous state changes as well. According to the above hypothesis, the appearance of this band in AmB-I and AmB as well as in the lipid membranes can suggest, by analogy, that these structures could be also present in other biological systems causing side effects after medical treatment. The decrease in this band together with the broadening of the main maximum centered around 1558 cm−1 indicates not crystalline but aggregated forms of AmB-I in the amorphous phase. The spectral shifts of the above-mentioned bands are related to weak H-bond interactions. The differences in intermolecular interactions could explain the differences in fluorescence lifetimes of AmB-I. The results of fluorescence spectroscopy and polarizationresolved fluorescence corroborate and strengthen the conclusions of vibrational spectroscopy. Crystal formation leads to strong coupling between molecules arranged in neighboring layers that are perpendicular to the long axis of the crystal, and furthermore, it strongly affects the fluorescence lifetime.

■

REFERENCES

AUTHOR INFORMATION

Corresponding Author

*Fax: +(48 81) 4456684. Phone: +(48 81) 4456899. E-mail: [email protected]. Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS This research was financed by the Ministry of Education and Science of Poland from the budget funds for science in the years 2008−2011 within the research project N N401 015035. 12712



Article

(33) Xu, J.; Guo, B. H.; Yang, R.; Wu, Q.; Chen, G. Q.; Zhang, Z. M. Polymer 2002, 43, 6893−6899. (34) Kansiz, M.; Dominguez-Vidal, A.; McNaughton, D.; Lendl, B. Anal. Bioanal. Chem. 2007, 388, 1207−1213. (35) Amer, M. S. Raman Spectroscopy for Soft Matter Applications; John Wiley & Sons, Inc.: Hoboken, NJ, 2009. (36) Gruszecki, W. I.; Gagoś, M.; Kernen, P. FEBS Lett. 2002, 524, 92−96. (37) Kasha, M. Radiat. Res. 1963, 20, 55−71. (38) Kasha, M.; Rawls, H. R.; Ashraf El-Bayoumi, M. Pure Appl. Chem. 1965, 11, 371−392. (39) Gagoś, M.; Koper, R.; Gruszecki, W. I. Biochim. Biophys. Acta 2001, 1511, 90−98. (40) Gruszecki, W. I.; Luchowski, R.; Gagoś, M.; Arczewska, M.; Sarkar, P.; Hereć, M.; Mysliwa-Kurdziel, B.; Strzalka, K.; Gryczynski, I.; Gryczynski, Z. Biophys. Chem. 2009, 143, 95−101. (41) Selvam, S.; Mishra, A. K. J. Photochem. Photobiol., B 2008, 93, 66−70.

12713


Spectroscopic Evidence for Self-Organization of N

Recommend Documents