Colloids and Surfaces B: Biointerfaces 141 (2016) 595–601
Contents lists available at ScienceDirect
Colloids and Surfaces B: Biointerfaces journal homepage: www.elsevier.com/locate/colsurfb
Chondroitin sulfate interacts mainly with headgroups in phospholipid monolayers Lucinéia F. Ceridório a,∗ , Luciano Caseli a , Osvaldo N. Oliveira Jr. b a b
Institute of Environmental, Chemical and Pharmaceutical Sciences, Federal University of São Paulo, UNIFESP, Diadema, SP, Brazil Institute of Physics of São Carlos, University of São Paulo, USP, São Carlos, SP, Brazil
a r t i c l e
i n f o
Article history: Received 8 November 2015 Received in revised form 21 January 2016 Accepted 11 February 2016 Available online 15 February 2016 Keywords: Chondroitin sulfate Cell membrane model Langmuir monolayers PM-IRRAS spectroscopy
a b s t r a c t Sulfated glycosaminoglycans are precursors of the extracellular matrix used to treat diseases related to blood clotting and degenerative joint diseases. These medical applications have been well established, but the mode of action at the molecular level, which depends on the interaction with cell membranes, is not known in detail. In this study, we investigated the interaction between chondroitin sulfate (CS) and phospholipid monolayers that mimic cell membranes. From surface pressure isotherms and polarizationmodulated infrared reflection absorption spectroscopy (PM-IRRAS), CS was found to interact mainly with the polar groups of dipalmitoyl phosphatidylcholine (DPPC) and dipalmitoyl phosphatidylglycerol (DPPG), with negligible penetration into the hydrophobic tails and only small changes in monolayer elasticity for the packing corresponding to a real cell membrane. The changes in surface pressure and surface potential isotherms depended on CS concentration and on the time allowed for its adsorption onto the monolayer, which points to a dynamic adsorption-desorption process. The charge of the phospholipid was also relevant, since CS induced order into DPPC monolayers while the opposite occurred for DPPG, according to the PM-IRRAS spectra. In summary, interaction with polar groups is responsible for the CS effects on model cell membranes. © 2016 Elsevier B.V. All rights reserved.
1. Introduction Glycosaminoglycans (GAGs) are long unbranched anionic polysaccharides found in the extracellular matrix at the cell surface combined with proteins and forming a structure known as proteoglycan (PG) [1,2]. Its chemical diversity arises from the relative quantity of sulfate groups and their position, in addition to the molecular weight. They comprise hyaluronanic acid, chondroitin sulfate (CS), heparan sulfate, keratan sulfate, and dermatan sulfate. These molecules differ in the kinds of the charged groups, the charge density and positions. Chondroitin sulfate, in particular, contains alternated sulfated units of N-acetyl-galactosamine (o-sulphated on C-6 or/and C-4) and glucuronic acid (O-sulphated on C-3 or C-2). Among the GAGS, CS is particularly important because this kind of molecule participates in various biological functions in the central nervous system, wound repair, cell division, differentiation, adhesion, migration and response to growth factors [3–5]. With these several functions, CS is of interest for pharmaceutical sciences and used as therapeutic agents in
∗ Corresponding author. E-mail addresses: [email protected], [email protected] (L.F. Ceridório). http://dx.doi.org/10.1016/j.colsurfb.2016.02.030 0927-7765/© 2016 Elsevier B.V. All rights reserved.
atherosclerosis/thrombosis [6–9], osteoarthritis and osteoarthrosis [10–13] and infections [14–16]. These potential physiological and pathological roles are likely dependent on the interaction with cell membranes, whose framework is made of a phospholipid bilayer that also contains embedded proteins and polysaccharides [17]. These phospholipids normally possess a double alkyl chain and typical headgroups are choline, ethanolamine, serine, glycerol, glucose and inositol [18]. Additionally, many researchers have been studying CS due to its role in cartilage and bone tissues and its interactions with growth factors and other proteins. Therefore, the characterization of the CS behavior with cell membrane molecules at molecular level is necessary. Thus, in this work, we investigated the interaction of CS with Langmuir monolayers of phospholipids with the same chain length and saturation level, but with different polar heads: dipalmitoyl phosphatidylcholine (DPPC), a major lipid found in outer leaflet of the membrane [18], and dipalmitoyl phosphatidylglycerol (DPPG), a negatively charged phospholipid. This methodology was chosen because Langmuir monolayers mimic half a membrane, with ease control over lateral packing and composition [19,20]. Furthermore, even though extensive studies exist for CS in the medical literature, little is known about molecular-level interactions in such uses [21–23].
596
L.F. Ceridório et al. / Colloids and Surfaces B: Biointerfaces 141 (2016) 595–601
2. Materials and methods Dipalmitoyl phosphatidylcholine (DPPC) and dipalmitoyl phosphatidylglycerol (DPPG) sodium salt were purchased from Avanti Polar Lipids. Chondroitin sulfate sodium salt (CS) from shark cartilage (CAS number 9007-28-7) acquired from Sigma Aldrich is a mixture of derivatives of chondroitin, mainly Chondroitin 4-Sulfate and Chondroitin 6-Sulfate. For CS from shark cartilage ca. 70% of 6-O sulfatation of N-acetylGalactosamine and 2-O sulfatation of glucuronic acid were found [24]. CS was dissolved in water to obtain a concentration of 10 mg/mL and then used as stock solution. DPPC was dissolved in chloroform while DPPG was dissolved in a mixture of chloroform/methanol (3:1 in volume) at a concentration of 0.5 mg mL−1 . Chloroform and methanol of analytical grade were purchased from Merck. Ultrapure water from a Milli-Q Plus system with resistivity of 18.2 M cm and pH 6.2 was used as subphase for the Langmuir monolayers for all CS solutions. It is important to emphasize that pure water was employed as subphase instead of phosphate buffer because in previous experiments we observed that the salts present in the buffer affected the interfacial properties of CS-lipid monolayer, especially for the surface potential measurements, leading us to inconclusive results. Because of that, we avoided the use of buffer in the aqueous subphase and pure water was employed for all cases. The structures of the phospholipids and CS are shown in Fig. 1. Surface activity of CS aqueous solutions was monitored with a tensiometer. Adsorption kinetics of CS in the presence of DPPC and DPPG was carried out for four solutions of CS; two of which were relatively highly concentrated (1.8 and 0.6 mg/mL), while the other two were diluted (1.8 × 10−3 and 0.6 × 10−3 mg/mL). The measurements were carried out in a Kibron tensiometer for the most concentrated solutions and in a KSV mini-though for the most diluted solutions. Since the results indicated no surface tension for CS aqueous solutions below 2.0 mg/mL, the concentrations of 6 × 10−3 mg/mL and 18 × 10−3 mg/mL were chosen for measurements in Langmuir monolayers. The Langmuir monolayers were obtained with a computer controlled KSV mini-though (KSV-Instruments, Finland) housed in a class 10,000 clean room. Phospholipid solutions were carefully spread on the aqueous surface with a Hamilton microsyringe and left for 15 min for solvent evaporation before the measurements. Then the monolayer was symmetrically compressed using movable barriers at the rate of 3.65 A2 /mol min. The surface pressure and surface potential were measured during compression using a Wilhelmy plate made of filter paper connected to a balance and a vibrating plate with a Kelvin probe, respectively. For mixed monolayers, the subphase was prepared with CS aqueous solutions in 250 mL of ultrapure water at 6 × 10−3 and 18 × 10−3 mg/mL. In the presence of CS, -A isotherms were measured after 1 h or 4 h of phospholipid spreading to ensure the adsorption n of CS on the monolayer. The mean molecular area was obtained without considering the CS molecules adsorbed at the monolayer. Measurements were repeated at least three times, showing reproducibility, with standard deviation of the points of the isotherms lower than 2%. In order to verify reversibility of the mixed monolayers, hysteresis experiments were conducted by compressing and subsequently decompressing the monolayer ten times. CS/phospholipid mixed monolayers were cycled to a maximum pressure of 40 mN/m and expanded at high areas per molecule, = 0 mN/m. The limit of compression should be below collapse to avoid irreversible changes with formation of lipid aggregates and multilayers. The morphology of DPPC and DPPG monolayers on subphases containing CS was investigated by Brewster Angle Microscopy (BAM) using an ULTRABAM instruments (Acccurion GMbH Gottingen, Germany) equipped with a 50 mW laser emitting p-polarized
light at 658 nm, a 10× magnification objective polarizer, an analyzer and a CCD camera. The spatial resolution was 2 m. The molecular-level interactions in Langmuir monolayers were investigated with polarization modulation infrared reflectionabsorption spectroscopy (PM-IRRAS) with a KSV instrument. The incidence angle to the normal was 80◦ , where intensity is maximum and noise level is lowest. The setup allows one to measure p and s polarization reflectivities (Rp and Rs respectively) where p and s correspond to the fraction of radiation polarized parallel and perpendicular to the plane of incidence, respectively. From the relation (Rp − Rs)/(Rp + Rs), a spectrum is obtained, with surface specific absorption bands. For a PM-IRRAS spectrum, a negative band indicates a transition moment oriented preferentially perpendicular to the surface, whereas a positive reflection absorption band indicates a transition moment oriented preferentially along the surface plane. The temperature in all experiments was kept at 20 ± 1 ◦ C. 3. Results and discussion The surface tension measurements of CS aqueous solutions showed no relevant decrease in water surface tension for concentrations below 2.0 mg/mL, which reveal a negligible surface activity for CS. CS was then inserted below phospholipid monolayers at 30 mN/m in several points of injection, and the surface pressure was monitored along time as shown in Fig. 2. To minimize problems with lateral diffusion this experiments were carried out in miniaturized wells with aqueous subphase of few milliliters. For neat phospholipids the surface pressure fall due to aggregation considering that this surface pressure is attained out-of-equilibrium. Particularly, the decrease of surface pressure for DPPG from 30 mN/m to 20 mN/m is related to the fact that this lipid is in a condition above its equilibrium surface pressure, which is a few mN/m. Ordered monolayers are, consequently, generally metastable with highs surface pressures attained because molecular rearrangements is not fast enough to follow the conditions established to attain initially the high values of surface pressure. For mixed monolayers, it is observed a considerable decrease in surface pressure for DPPC within 3 h, thus pointing to either removal of lipids from the interface by CS or condensation of the DPPC monolayer. In contrast, the surface pressure increased by 2 mN/m for the DPPG/CS monolayer for the same interval. This opposite effect suggests that the interaction with CS depends on the charge of the phospholipid head group. Investigation on other initial surface pressures was carried out and we observed or constant values of surface pressure either a low decrease of the surface pressure upon injection of CS. Furthermore, as the surface pressure of 30 mN/m is those related to natural membranes, we focused the discussion on this value. The incorporation of CS into DPPC and DPPG monolayers affected the surface pressure-area isotherms, depending on both the CS concentration and the adsorption times, as shown in Fig. 3. The adsorption time corresponds to the time elapsed between the spreading of the lipid on the CS solutions and the beginning of the compression. For DPPC, CS had almost no effect at large areas per molecule and the liquid-expanded to liquid-condensed phase transition was little affected. At high pressures, there was little effect with two exceptions: (i) large condensation for 1 h waiting time, especially for the highest concentration (18 × 10−3 mg/mL); (ii) higher collapse pressure for 4 h with 6 × 10−3 mg/mL. In contrast, for DPPG an expansion on the monolayer was observed at large areas (low surface pressures) in all cases. No significant effect occurred at high pressures, with the exception of a condensation for 1 h and 18 × 10−3 mg/mL. It is important to emphasize that for small concentrations of CS, at an adsorption controlled by diffusion, CS will take more time than for the higher concentrations. As
L.F. Ceridório et al. / Colloids and Surfaces B: Biointerfaces 141 (2016) 595–601
DPPG
O
O O
O
597
OH
P
O
OH
-
O
O
O
DPPC
O
O O
O
CH3
O
P
+
N
-
O
O
CH3
CH3
O
CS O
-
H2C OR
O
4
RO
O H OR
2
O H3C
6
O
O
H n
NH
CS-6S +
Na
CS-4S CS-2,6S
R2
R4
R6
H
H
SO3-
H SO3
-
SO3-
H
H
H
O
Fig. 1. Chemical structures of dipalmitoyl phosphatidyl choline (DPPC), dipalmitoyl phosphatidyl glycerol (DPPG), and chondroitin sulfate (CS).
Fig. 2. Kinetics of adsorption for CS from the subphase onto DPPC and DPPG monolayers at 30 mN m−1 .
no significant expansion was observed for the monolayers at high surface pressures, it can be said that CS does not penetrate into the hydrophobic tails of either DPPC or DPPG monolayers at these conditions, especially at a pressure corresponding to the packing of a cell membrane (30 mN/m) [25]. This does not necessarily mean that CS was completely removed from the interface since it can remain at a subsurface below the phospholipid polar heads. The results for CS are in contrast to previous reports of polysaccharide moieties penetrating into the alkyl tails of DPPG [26–28]. Furthermore, it is expected that coulombic interactions should occur between a negatively charged CS and the negatively charged head groups of DPPG. It is important to emphasize, that although adsorption is generally irreversible for polymers like CS, for the compression isotherms, the adsorption primarily occurs at low values of surface pressure, and the air-water interface is compressed sequentially.
Therefore, the dynamic process involving the interaction of CS with the lipids may be directly related to the new lipid surface density provided for CS, which may molecularly rearrange, changing its conformation in order to better adapt the anchor points for the interaction with the lipid at the interface. The surface elasticity of biomembranes has important biological implications, since it is crucial for membrane rupture or leakage [29]. From the slope of the -A isotherms, the compression modulus CS −1 can be calculated using Cs−1 = −A.(d/dA), where A is the molecular area at a given surface pressure. As expected from the small effects observed in the surface pressure isotherms, CS almost did not affect the compressibility curves for DPPC (see Fig. S1 in Supplementary material). For DPPG, Fig. S1 also shows a small effect induced by CS, though there was a large decrease in Cs−1 at very high pressures (∼45 mN/m). Overall, the compressibility results are consistent with CS interacting only with head groups,
598
L.F. Ceridório et al. / Colloids and Surfaces B: Biointerfaces 141 (2016) 595–601
Fig. 3. Surface pressure-area isotherms for phospholipid monolayers on aqueous subphases with different CS concentrations and incorporation times. (A) DPPC and (B) DPPG. The error bar represents the average of at least 3 measurements, with a standard deviation lower than 2%.
therefore causing only small changes in Cs−1, with a larger effect for the negatively charged DPPG. Similar conclusions can be drawn from the changes in monolayer morphology induced by CS on the BAM images in Fig. S2 in Supplementary material. The incorporation of the negatively charged CS could affect the monolayer surface potential significantly. According to the Demchak-Fort model [32,33] the monolayer surface potential depends on the normal component of the dipole moments of the film-forming molecules, on the orientation of the subphase water molecules close to the interface, and on the contribution from the electrical double-layer when the monolayer is totally or partially ionized [30,31]. Fig. 4 shows that the effects are again dependent
on the waiting time and CS concentration. For DPPC, the surface potential increased in all cases, generally with slightly larger critical area (i.e., area at which the surface potential increases steeply upon compression). The largest increase occurred for 4 h with 6 × 10−3 mg/mL (the same monolayer that displayed very high collapse pressure). Since CS is negatively charged, if it contributed to the double-layer potential by adsorbing on the monolayer (and making it negatively charged), the surface potential should decrease. Therefore, CS probably affects the head groups, altering their orientation. With the large dipole moment of the zwitterion, a small change in orientation should be sufficient to increase the surface potential. No clear trend could be established in the results
L.F. Ceridório et al. / Colloids and Surfaces B: Biointerfaces 141 (2016) 595–601
599
Fig. 4. Surface potential-area isotherms for phospholipid monolayers on aqueous subphases with different CS concentrations and incorporation times. (A) DPPC and (B) DPPG. The error bar represents the average of at least 3 measurements, with a standard deviation lower than 2%.
for DPPG. For a waiting time of 1 h, CS caused a small increase in potential at large areas, then large decreases at small areas (high surface pressures). For 4 h, there was a large increase at large areas and a small decrease at small areas. Therefore, the conclusion that can be drawn is that for DPPC CS interacts with the head groups. In order to access information on how chemical groups from the phospholipids and CS are affected by adsorption, PMIRRAS measurements were performed [32,33]. The spectra of the neat phospholipid and mixed monolayers were collected in the wavenumber range from 800 to 4000 cm−1 , at a surface pressure of 30 mN/m, and were subtracted from the spectrum of the free-film interface. Fig. 5 shows bands at 2850 cm−1 and 2916 cm−1 , assigned to the antisymmetric ass (CH2 ) and symmetric s (CH2 ) stretching of methylene groups, respectively. The ratio between the intensi-
ties of these bands can be taken as a parameter of order [34,35]. It increases from 1.74 to 2.81 with CS incorporation in DPPC, while for DPPG it decreases from 1.60 to 1.24. Therefore, CS induces order in the DPPC monolayer, but causes disorder in the DPPG monolayer. Although for both a condensation of the monolayer occurs at 30 mN/m, as observed in the surface pressure-area isotherms, the surface potential for DPPC increased with CS, and this must be related to the increase of lipid order. The lack of significant changes in the position of the bands associated with the tails indicates that CS does not penetrate into either DPPC or DPPG, consistent with the surface pressure isotherms. Therefore, the effects on the packing of the monolayers should be due to interactions with headgroups, as confirmed below.
600
L.F. Ceridório et al. / Colloids and Surfaces B: Biointerfaces 141 (2016) 595–601
Fig. 5. PM-IRRAS spectra for phospholipid monolayers on aqueous subphases containing 1.8 × 10−3 mg/mL of chondroitin sulfate (1 h) at a surface pressure of 30 mN m−1 in the CH asymmetric and symmetric stretching region.
Fig. 6. PM-IRRAS spectra for phospholipid monolayers on aqueous subphases containing 1.8 × 10−3 mg mL−1 of chondroitin sulfate (1 h) at a surface pressure of 30 mN m−1 in the PO2 asymmetric and symmetric stretching region.
The spectra in the region between 1000 and 1800 cm−1 provide information on hydrogen bonds, hydration and conformation of the head group. Fig. 6 features two bands at 1088 cm−1 and 1232 cm−1 assigned to vibrational modes of the phosphate group. For the DPPC monolayer there is an additional band at 1175 cm−1 due to C N stretching. With CS adsorption, the band assigned to asymmetric PO2 stretching (1232 cm−1 ) overlaps the S O stretch band, being relatively more intense. Also, the band at 1170 cm−1 assigned to C
O C stretching for cyclic ethers of CS in the spectrum of DPPG/CS monolayer coincides with the band assigned to C C N amine from DPPC in the spectrum for the DPPC/CS monolayer. Other changes in band intensity and shape are observed, which can be related to interaction of glucoside moieties of CS with the lipid monolayers [26]. These findings confirm the interaction of CS with the head groups. For DPPG, this also happens for the antisymmetric PO2 stretching mode, whose width varies upon incorporation of CS. A band assigned to HO CR2 vibrations from CS appears at 1190 cm−1 , again confirming CS incorporation. Phosphate bands in the region 1000–1100 cm−1 also seem to disappear with CS incorporation. The interaction between CS and headgroups was also manifested in the spectral region between 1600 and 1800 cm−1 (results not shown). For CS-containing monolayers, the band assigned to C O stretching of amide appears as negative band, which point to the vibration transition dipoles for C O groups of amide are oriented preferentially perpendicular to the monolayer. The band at 1740 cm−1 due to ester C O stretching is shifted by 5 and 9 cm−1 for DPPC and DPPG, respectively, owing to hydrogen bonds between CS and phospholipids, while a shoulder appears at 1740 cm−1 for the CS-containing monolayers. We have not analyzed the region between 1650 and 1690 cm−1 because it is considered as interference from water [36,37]. Shifts to lower wavenumbers to the C O stretching in esters are evidence of the interaction of CS with the phospholipid monolayer. In summary, CS does not penetrate in the hydrophobic region of either DPPC or DPPG monolayer, but rather interacts with their head groups, this interaction being modulated by the charges in the polar head. For DPPC, CS induces order which may explain the decrease in pressure in the adsorption kinetics of Fig. 2. In contrast, in DPPG monolayers the incorporation of CS induces disorder, again consistent with the adsorption kinetics data. 4. Conclusions The polysaccharide CS, soluble and with poor surface activity in aqueous solutions, adsorbs at the air-water interface in the presence of DPPC and DPPG monolayers. CS thus has affinity for
L.F. Ceridório et al. / Colloids and Surfaces B: Biointerfaces 141 (2016) 595–601
both membrane phospholipids, but with different means of action. Since DPPC and DPPG have the same alkyl chain, differences in the adsorption profile indicate that the interaction is affected by the polar head. Changes in the PM-IRRAS spectra showed that CS forms a subsurface below the DPPC and DPPG monolayers, with the stronger electrostatic interaction with DPPG causing disorder in the monolayer. We hope that these findings may help understand in detail the interaction of GAGs with membranes at the molecular level, which is of biological interest since many biological processes depend on lipid-GAG interactions. Acknowledgments This work was supported by CAPES, FAPESP (Grant 2013/142627), CNPq and nBioNet network (Brazil). Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/j.colsurfb.2016.02. 030. References [1] H.A. Santos, V. García-Morales, R.-J. Roozeman, J.A. Manzanares, K. Kontturi, Interfacial interaction between dextran sulfate and lipid monolayers: an electrochemical study, Langmuir 21 (2005) 5475–5484, http://dx.doi.org/10. 1021/la046825u. [2] R.M. Lauder, T.N. Huckerby, I.A. Nieduszynski, I.H. Sadler, Characterisation of oligosaccharides from the chondroitin/dermatan sulphates: H and 13 C NMR studies of oligosaccharides generated by nitrous acid depolymerisation, Carbohydr. Res. 346 (2011) 2222–2227, http://dx.doi.org/10.1016/j.carres. 2011.06.033. [3] A. Varki, R. Cummings, J. Esko, H. Freeze, G. Hart, J. Marth, Proteoglycans and Sulfated Glycosaminoglycans—Essentials of Glycobiolog, 2nd edition, Cold Springer Harbor Laboratoty Press, NY-EUA, 1999. [4] R.J. Linhardt, T. Toida, Role of glycosaminoglycans in cellular communication, Acc. Chem. Res. 37 (2004) 431–438, http://dx.doi.org/10.1021/ar030138x. [5] N.S. Gandhi, R.L. Mancera, The structure of glycosaminoglycans and their interactions with proteins, Chem. Biol. Drug Des. 72 (2008) 455–482, http:// dx.doi.org/10.1111/j.1747-0285.2008.00741.x. [6] M. Krumbiegel, K. Arnold, Microelectrophoresis studies of the binding of glycosaminoglycans to phosphatidylcholine liposomes, Chem. Phys. Lipids 54 (1990) 1–7, http://dx.doi.org/10.1016/0009-3084(90)90053-T. [7] K. Nakazawa, K. Murata, K. Izuka, Y. Oshima, The short-term effects of chondroitin sulfates A and C on coronary atherosclerotic subjects: with reference to its anti-thrombogenic activities, Jpn. Heart J. 10 (1969) 289–296. [8] M. McGee, W.D. Wagner, Chondroitin sulfate anticoagulant activity is linked to water transfer: relevance to proteoglycan structure in atherosclerosis, Arterioscler. Thromb. Vasc. Biol. 23 (2003) 1921–1927, http://dx.doi.org/10. 1161/01.ATV.0000090673.96120.67. [9] S. Chen, G. Li, N. Wu, X. Guo, N. Liao, X. Ye, et al., Sulfation pattern of the fucose branch is important for the anticoagulant and antithrombotic activities of fucosylated chondroitin sulfates, Biochim. Biophys. Acta 1830 (2013) 3054–3066, http://dx.doi.org/10.1016/j.bbagen.2013.01.001. [10] D.O. Clegg, D.J. Reda, C.L. Harris, M.A. Klein, J.R.O. Dell, M.M. Hooper, et al., Glucosamine and chondroitin sulfate for knee osteoarthritis, N. Engl. J. Med. 354 (2006) 795–808. [11] N. Volpi, Analytical aspects of pharmaceutical grade chondroitin sulfates, J. Pharm. Sci. 96 (2007) 3168–3180, http://dx.doi.org/10.1002/jps. [12] T.E. McAlindon, M.P. Lavalley, J.P. Gulin, D.T. Felson, Glucosamine and chondroitin for treatment of osteoarthritis, J. Am. Med. Assoc. 283 (2000) 1469–1475. [13] N. Volpi, Anti-inflammatory activity of chondroitin sulphate: new functions from an old natural macromolecule, Inflammopharmacology 19 (2011) 299–306, http://dx.doi.org/10.1007/s10787-011-0098-0. [14] N. Huang, M.-Y. Wu, C.-B. Zheng, L. Zhu, J.-H. Zhao, Y.-T. Zheng, The depolymerized fucosylated chondroitin sulfate from sea cucumber potently inhibits HIV replication via interfering with virus entry, Carbohydr. Res. 380 (2013) 64–69, http://dx.doi.org/10.1016/j.carres.2013.07.010.
601
[15] W. Lian, M. Wu, N. Huang, N. Gao, C. Xiao, Z. Li, et al., Anti-HIV-1 activity and structure-activity-relationship study of a fucosylated glycosaminoglycan from an echinoderm by targeting the conserved CD4 induced epitope, Biochim. Biophys. Acta 1830 (2013) 4681–4691, http://dx.doi.org/10.1016/j. bbagen.2013.06.003. [16] M. Marchetti, E. Trybala, F. Superti, M. Johansson, T. Bergström, Inhibition of herpes simplex virus infection by lactoferrin is dependent on interference with the virus binding to glycosaminoglycans, Virology 318 (2004) 405–413, http://dx.doi.org/10.1016/j.virol.2003.09.029. [17] M. Edidin, Lipids on the frontier: a century of cell-membrane bilayers, Nat. Rev. Mol. Cell Biol. 4 (2003) 414–418, http://dx.doi.org/10.1038/nrm1102. [18] P.L. Yeagle (Ed.), The Structure of Biological Membranes, CRC Press, New York, EUA, 2012. [19] G. Brezesinski, H. Mohwald, Langmuir monolayers to study interactions at model membrane surfaces, Adv. Colloid Interface Sci. 102 (2003) 563–584, http://dx.doi.org/10.1016/S0001-8686(02)00071-4. [20] H. Brockman, Lipid monolayers: why use half a membrane to characterize protein-membrane interactions? Curr. Opin. Struct. Biol. 9 (1999) 438–443. [21] S. Xu, M. Yonese, Two dimensional and three dimensional interactions between bovine serum albumin and chondroitin sulfate, Polym. J. 39 (2007) 298–303, http://dx.doi.org/10.1295/polymj.PJ2006123. [22] N. Volpi, Influence of charge density sulfate group position and molecular mass on adsorption of chondroitin sulfate onto coral, Biomaterials 23 (2002) 3015–3022. [23] J. Ouyang, S. Deng, J. Zhong, Crystallization of calcium oxalate monohydrate at dipalmitoylphosphatidylcholine monolayers in the presence of chondroitin sulfate A, J. Cryst. Growth 270 (2004) 646–654, http://dx.doi.org/10.1016/j. jcrysgro.2004.06.058. [24] R.M. Lauder, T.N. Huckerby, I.A. Nieduszynski, I.H. Sadler, Characterisation of oligosaccharides from the chondroitin/dermatan sulphates: H and 13 C NMR studies of oligosaccharides generated by nitrous acid depolymerisation, Carbohydr. Res. 346 (2011) 2222–2227, http://dx.doi.org/10.1016/j.carres. 2011.06.033. [25] A. Blume, A comparative study of the phase transitions of phospholipid bilayers and monolayers, Biochim. Biophys. Acta: Biomembr. 557 (1979) 32–44, http://dx.doi.org/10.1016/0005-2736(79)90087-7. [26] A.K. de Brito, L. Caseli, C.S.F. Nordi, Interaction of algal polysaccharide with lipid Langmuir monolayers, Mater. Sci. Eng. C 31 (2011) 1857–1860, http://dx. doi.org/10.1016/j.msec.2011.09.002. [27] M. Pinheiro, M. Lu, S. Reis, J.L.F.C. Lima, T. Mart, L. Camacho, et al., Molecular interaction of rifabutin on model lung surfactant monolayers, J. Phys. Chem. B (2012), http://dx.doi.org/10.1021/jp303725j. [28] A.A. Torrano, Â.S. Pereira, O.N. Oliveira Jr., A. Barros-timmons, Probing the interaction of oppositely charged gold nanoparticles with DPPG and DPPC Langmuir monolayers as cell membrane models, Colloids Surf. B: Biointerfaces 108 (2013) 120–126, http://dx.doi.org/10.1016/j.colsurfb.2013.02.014. [29] L. Caseli, C.P. Pascholati, F. Teixeira, S. Nosov, C. Vebert, H.E. Müeller, et al., Interaction of oligonucleotide-based amphiphilic block copolymers with cell membrane models, J. Colloid Interface Sci. 347 (2010) 56–61, http://dx.doi. org/10.1016/j.jcis.2010.03.006. [30] O.N. Oliveira Jr., C. Bonardi, The surface potential of Langmuir monolayers revisited, Langmuir 7463 (1997) 5920–5924. [31] P. Dynarowicz-łatka, A. Dhanabalan, O.N. Oliveira Jr., Modern physicochemical research on Langmuir monolayers, Adv. Colloid Interface Sci. 91 (2001) 221–293, http://dx.doi.org/10.1016/S0001-8686(99)00034-2. [32] L. Caseli, D. Sousa-Martins, M. Maia, A.A.S. Lima-Filho, E.B. Rodrigues, R. Belfort, An intraocular dye solution based on lutein and zeaxanthin in a surrogate internal limiting membrane model: a Langmuir monolayer study, Colloids Surf. B: Biointerfaces 107 (2013) 124–129, http://dx.doi.org/10.1016/ j.colsurfb.2013.01.076. [33] T.E. Goto, L. Caseli, Understanding the collapse mechanism in Langmuir monolayers through polarization modulation-infrared reflection absorption spectroscopy, Langmuir (2013), http://dx.doi.org/10.1021/la402044c. [34] I.W. Levin, T.E. Thompson, Y. Barenholz, C. Huang, Two types of hydrocarbon chain interdigitation in Sphingomyelin bilayers, Biochemistry (1985) 6282–6286, http://dx.doi.org/10.1021/bi00343a036. [35] R.F. Lopez, T.M. Nobre, C.D.M. Accardo, P.C. Pernambuco Filho, H.B. Nader, C.C. Lopes, et al., Effect of carrageenans of different chemical structures in biointerfaces: a Langmuir film study, Colloids Surf. B: Biointerfaces 111 (2013) 530–535, http://dx.doi.org/10.1016/j.colsurfb.2013.06.039. [36] D. Blaudez, T. Buffeteau, B. Desbat, J. Marie, Infrared and Raman spectroscopies of monolayers at the air-water interface, Curr. Opin. Colloid Interface Sci. 4 (1999) 265–272, http://dx.doi.org/10.1016/S13590294(99)90006-7. [37] H. Wang, S. Coss, A. Mudalige, R.L. Polt, J.E. Pemberton, A PM-IRRAS investigation of monorhamnolipid orientation at the air-water interface, Langmuir (2013), http://dx.doi.org/10.1021/la3051725.
Contents lists available at ScienceDirect
Colloids and Surfaces B: Biointerfaces journal homepage: www.elsevier.com/locate/colsurfb
Chondroitin sulfate interacts mainly with headgroups in phospholipid monolayers Lucinéia F. Ceridório a,∗ , Luciano Caseli a , Osvaldo N. Oliveira Jr. b a b
Institute of Environmental, Chemical and Pharmaceutical Sciences, Federal University of São Paulo, UNIFESP, Diadema, SP, Brazil Institute of Physics of São Carlos, University of São Paulo, USP, São Carlos, SP, Brazil
a r t i c l e
i n f o
Article history: Received 8 November 2015 Received in revised form 21 January 2016 Accepted 11 February 2016 Available online 15 February 2016 Keywords: Chondroitin sulfate Cell membrane model Langmuir monolayers PM-IRRAS spectroscopy
a b s t r a c t Sulfated glycosaminoglycans are precursors of the extracellular matrix used to treat diseases related to blood clotting and degenerative joint diseases. These medical applications have been well established, but the mode of action at the molecular level, which depends on the interaction with cell membranes, is not known in detail. In this study, we investigated the interaction between chondroitin sulfate (CS) and phospholipid monolayers that mimic cell membranes. From surface pressure isotherms and polarizationmodulated infrared reflection absorption spectroscopy (PM-IRRAS), CS was found to interact mainly with the polar groups of dipalmitoyl phosphatidylcholine (DPPC) and dipalmitoyl phosphatidylglycerol (DPPG), with negligible penetration into the hydrophobic tails and only small changes in monolayer elasticity for the packing corresponding to a real cell membrane. The changes in surface pressure and surface potential isotherms depended on CS concentration and on the time allowed for its adsorption onto the monolayer, which points to a dynamic adsorption-desorption process. The charge of the phospholipid was also relevant, since CS induced order into DPPC monolayers while the opposite occurred for DPPG, according to the PM-IRRAS spectra. In summary, interaction with polar groups is responsible for the CS effects on model cell membranes. © 2016 Elsevier B.V. All rights reserved.
1. Introduction Glycosaminoglycans (GAGs) are long unbranched anionic polysaccharides found in the extracellular matrix at the cell surface combined with proteins and forming a structure known as proteoglycan (PG) [1,2]. Its chemical diversity arises from the relative quantity of sulfate groups and their position, in addition to the molecular weight. They comprise hyaluronanic acid, chondroitin sulfate (CS), heparan sulfate, keratan sulfate, and dermatan sulfate. These molecules differ in the kinds of the charged groups, the charge density and positions. Chondroitin sulfate, in particular, contains alternated sulfated units of N-acetyl-galactosamine (o-sulphated on C-6 or/and C-4) and glucuronic acid (O-sulphated on C-3 or C-2). Among the GAGS, CS is particularly important because this kind of molecule participates in various biological functions in the central nervous system, wound repair, cell division, differentiation, adhesion, migration and response to growth factors [3–5]. With these several functions, CS is of interest for pharmaceutical sciences and used as therapeutic agents in
∗ Corresponding author. E-mail addresses: [email protected], [email protected] (L.F. Ceridório). http://dx.doi.org/10.1016/j.colsurfb.2016.02.030 0927-7765/© 2016 Elsevier B.V. All rights reserved.
atherosclerosis/thrombosis [6–9], osteoarthritis and osteoarthrosis [10–13] and infections [14–16]. These potential physiological and pathological roles are likely dependent on the interaction with cell membranes, whose framework is made of a phospholipid bilayer that also contains embedded proteins and polysaccharides [17]. These phospholipids normally possess a double alkyl chain and typical headgroups are choline, ethanolamine, serine, glycerol, glucose and inositol [18]. Additionally, many researchers have been studying CS due to its role in cartilage and bone tissues and its interactions with growth factors and other proteins. Therefore, the characterization of the CS behavior with cell membrane molecules at molecular level is necessary. Thus, in this work, we investigated the interaction of CS with Langmuir monolayers of phospholipids with the same chain length and saturation level, but with different polar heads: dipalmitoyl phosphatidylcholine (DPPC), a major lipid found in outer leaflet of the membrane [18], and dipalmitoyl phosphatidylglycerol (DPPG), a negatively charged phospholipid. This methodology was chosen because Langmuir monolayers mimic half a membrane, with ease control over lateral packing and composition [19,20]. Furthermore, even though extensive studies exist for CS in the medical literature, little is known about molecular-level interactions in such uses [21–23].
596
L.F. Ceridório et al. / Colloids and Surfaces B: Biointerfaces 141 (2016) 595–601
2. Materials and methods Dipalmitoyl phosphatidylcholine (DPPC) and dipalmitoyl phosphatidylglycerol (DPPG) sodium salt were purchased from Avanti Polar Lipids. Chondroitin sulfate sodium salt (CS) from shark cartilage (CAS number 9007-28-7) acquired from Sigma Aldrich is a mixture of derivatives of chondroitin, mainly Chondroitin 4-Sulfate and Chondroitin 6-Sulfate. For CS from shark cartilage ca. 70% of 6-O sulfatation of N-acetylGalactosamine and 2-O sulfatation of glucuronic acid were found [24]. CS was dissolved in water to obtain a concentration of 10 mg/mL and then used as stock solution. DPPC was dissolved in chloroform while DPPG was dissolved in a mixture of chloroform/methanol (3:1 in volume) at a concentration of 0.5 mg mL−1 . Chloroform and methanol of analytical grade were purchased from Merck. Ultrapure water from a Milli-Q Plus system with resistivity of 18.2 M cm and pH 6.2 was used as subphase for the Langmuir monolayers for all CS solutions. It is important to emphasize that pure water was employed as subphase instead of phosphate buffer because in previous experiments we observed that the salts present in the buffer affected the interfacial properties of CS-lipid monolayer, especially for the surface potential measurements, leading us to inconclusive results. Because of that, we avoided the use of buffer in the aqueous subphase and pure water was employed for all cases. The structures of the phospholipids and CS are shown in Fig. 1. Surface activity of CS aqueous solutions was monitored with a tensiometer. Adsorption kinetics of CS in the presence of DPPC and DPPG was carried out for four solutions of CS; two of which were relatively highly concentrated (1.8 and 0.6 mg/mL), while the other two were diluted (1.8 × 10−3 and 0.6 × 10−3 mg/mL). The measurements were carried out in a Kibron tensiometer for the most concentrated solutions and in a KSV mini-though for the most diluted solutions. Since the results indicated no surface tension for CS aqueous solutions below 2.0 mg/mL, the concentrations of 6 × 10−3 mg/mL and 18 × 10−3 mg/mL were chosen for measurements in Langmuir monolayers. The Langmuir monolayers were obtained with a computer controlled KSV mini-though (KSV-Instruments, Finland) housed in a class 10,000 clean room. Phospholipid solutions were carefully spread on the aqueous surface with a Hamilton microsyringe and left for 15 min for solvent evaporation before the measurements. Then the monolayer was symmetrically compressed using movable barriers at the rate of 3.65 A2 /mol min. The surface pressure and surface potential were measured during compression using a Wilhelmy plate made of filter paper connected to a balance and a vibrating plate with a Kelvin probe, respectively. For mixed monolayers, the subphase was prepared with CS aqueous solutions in 250 mL of ultrapure water at 6 × 10−3 and 18 × 10−3 mg/mL. In the presence of CS, -A isotherms were measured after 1 h or 4 h of phospholipid spreading to ensure the adsorption n of CS on the monolayer. The mean molecular area was obtained without considering the CS molecules adsorbed at the monolayer. Measurements were repeated at least three times, showing reproducibility, with standard deviation of the points of the isotherms lower than 2%. In order to verify reversibility of the mixed monolayers, hysteresis experiments were conducted by compressing and subsequently decompressing the monolayer ten times. CS/phospholipid mixed monolayers were cycled to a maximum pressure of 40 mN/m and expanded at high areas per molecule, = 0 mN/m. The limit of compression should be below collapse to avoid irreversible changes with formation of lipid aggregates and multilayers. The morphology of DPPC and DPPG monolayers on subphases containing CS was investigated by Brewster Angle Microscopy (BAM) using an ULTRABAM instruments (Acccurion GMbH Gottingen, Germany) equipped with a 50 mW laser emitting p-polarized
light at 658 nm, a 10× magnification objective polarizer, an analyzer and a CCD camera. The spatial resolution was 2 m. The molecular-level interactions in Langmuir monolayers were investigated with polarization modulation infrared reflectionabsorption spectroscopy (PM-IRRAS) with a KSV instrument. The incidence angle to the normal was 80◦ , where intensity is maximum and noise level is lowest. The setup allows one to measure p and s polarization reflectivities (Rp and Rs respectively) where p and s correspond to the fraction of radiation polarized parallel and perpendicular to the plane of incidence, respectively. From the relation (Rp − Rs)/(Rp + Rs), a spectrum is obtained, with surface specific absorption bands. For a PM-IRRAS spectrum, a negative band indicates a transition moment oriented preferentially perpendicular to the surface, whereas a positive reflection absorption band indicates a transition moment oriented preferentially along the surface plane. The temperature in all experiments was kept at 20 ± 1 ◦ C. 3. Results and discussion The surface tension measurements of CS aqueous solutions showed no relevant decrease in water surface tension for concentrations below 2.0 mg/mL, which reveal a negligible surface activity for CS. CS was then inserted below phospholipid monolayers at 30 mN/m in several points of injection, and the surface pressure was monitored along time as shown in Fig. 2. To minimize problems with lateral diffusion this experiments were carried out in miniaturized wells with aqueous subphase of few milliliters. For neat phospholipids the surface pressure fall due to aggregation considering that this surface pressure is attained out-of-equilibrium. Particularly, the decrease of surface pressure for DPPG from 30 mN/m to 20 mN/m is related to the fact that this lipid is in a condition above its equilibrium surface pressure, which is a few mN/m. Ordered monolayers are, consequently, generally metastable with highs surface pressures attained because molecular rearrangements is not fast enough to follow the conditions established to attain initially the high values of surface pressure. For mixed monolayers, it is observed a considerable decrease in surface pressure for DPPC within 3 h, thus pointing to either removal of lipids from the interface by CS or condensation of the DPPC monolayer. In contrast, the surface pressure increased by 2 mN/m for the DPPG/CS monolayer for the same interval. This opposite effect suggests that the interaction with CS depends on the charge of the phospholipid head group. Investigation on other initial surface pressures was carried out and we observed or constant values of surface pressure either a low decrease of the surface pressure upon injection of CS. Furthermore, as the surface pressure of 30 mN/m is those related to natural membranes, we focused the discussion on this value. The incorporation of CS into DPPC and DPPG monolayers affected the surface pressure-area isotherms, depending on both the CS concentration and the adsorption times, as shown in Fig. 3. The adsorption time corresponds to the time elapsed between the spreading of the lipid on the CS solutions and the beginning of the compression. For DPPC, CS had almost no effect at large areas per molecule and the liquid-expanded to liquid-condensed phase transition was little affected. At high pressures, there was little effect with two exceptions: (i) large condensation for 1 h waiting time, especially for the highest concentration (18 × 10−3 mg/mL); (ii) higher collapse pressure for 4 h with 6 × 10−3 mg/mL. In contrast, for DPPG an expansion on the monolayer was observed at large areas (low surface pressures) in all cases. No significant effect occurred at high pressures, with the exception of a condensation for 1 h and 18 × 10−3 mg/mL. It is important to emphasize that for small concentrations of CS, at an adsorption controlled by diffusion, CS will take more time than for the higher concentrations. As
L.F. Ceridório et al. / Colloids and Surfaces B: Biointerfaces 141 (2016) 595–601
DPPG
O
O O
O
597
OH
P
O
OH
-
O
O
O
DPPC
O
O O
O
CH3
O
P
+
N
-
O
O
CH3
CH3
O
CS O
-
H2C OR
O
4
RO
O H OR
2
O H3C
6
O
O
H n
NH
CS-6S +
Na
CS-4S CS-2,6S
R2
R4
R6
H
H
SO3-
H SO3
-
SO3-
H
H
H
O
Fig. 1. Chemical structures of dipalmitoyl phosphatidyl choline (DPPC), dipalmitoyl phosphatidyl glycerol (DPPG), and chondroitin sulfate (CS).
Fig. 2. Kinetics of adsorption for CS from the subphase onto DPPC and DPPG monolayers at 30 mN m−1 .
no significant expansion was observed for the monolayers at high surface pressures, it can be said that CS does not penetrate into the hydrophobic tails of either DPPC or DPPG monolayers at these conditions, especially at a pressure corresponding to the packing of a cell membrane (30 mN/m) [25]. This does not necessarily mean that CS was completely removed from the interface since it can remain at a subsurface below the phospholipid polar heads. The results for CS are in contrast to previous reports of polysaccharide moieties penetrating into the alkyl tails of DPPG [26–28]. Furthermore, it is expected that coulombic interactions should occur between a negatively charged CS and the negatively charged head groups of DPPG. It is important to emphasize, that although adsorption is generally irreversible for polymers like CS, for the compression isotherms, the adsorption primarily occurs at low values of surface pressure, and the air-water interface is compressed sequentially.
Therefore, the dynamic process involving the interaction of CS with the lipids may be directly related to the new lipid surface density provided for CS, which may molecularly rearrange, changing its conformation in order to better adapt the anchor points for the interaction with the lipid at the interface. The surface elasticity of biomembranes has important biological implications, since it is crucial for membrane rupture or leakage [29]. From the slope of the -A isotherms, the compression modulus CS −1 can be calculated using Cs−1 = −A.(d/dA), where A is the molecular area at a given surface pressure. As expected from the small effects observed in the surface pressure isotherms, CS almost did not affect the compressibility curves for DPPC (see Fig. S1 in Supplementary material). For DPPG, Fig. S1 also shows a small effect induced by CS, though there was a large decrease in Cs−1 at very high pressures (∼45 mN/m). Overall, the compressibility results are consistent with CS interacting only with head groups,
598
L.F. Ceridório et al. / Colloids and Surfaces B: Biointerfaces 141 (2016) 595–601
Fig. 3. Surface pressure-area isotherms for phospholipid monolayers on aqueous subphases with different CS concentrations and incorporation times. (A) DPPC and (B) DPPG. The error bar represents the average of at least 3 measurements, with a standard deviation lower than 2%.
therefore causing only small changes in Cs−1, with a larger effect for the negatively charged DPPG. Similar conclusions can be drawn from the changes in monolayer morphology induced by CS on the BAM images in Fig. S2 in Supplementary material. The incorporation of the negatively charged CS could affect the monolayer surface potential significantly. According to the Demchak-Fort model [32,33] the monolayer surface potential depends on the normal component of the dipole moments of the film-forming molecules, on the orientation of the subphase water molecules close to the interface, and on the contribution from the electrical double-layer when the monolayer is totally or partially ionized [30,31]. Fig. 4 shows that the effects are again dependent
on the waiting time and CS concentration. For DPPC, the surface potential increased in all cases, generally with slightly larger critical area (i.e., area at which the surface potential increases steeply upon compression). The largest increase occurred for 4 h with 6 × 10−3 mg/mL (the same monolayer that displayed very high collapse pressure). Since CS is negatively charged, if it contributed to the double-layer potential by adsorbing on the monolayer (and making it negatively charged), the surface potential should decrease. Therefore, CS probably affects the head groups, altering their orientation. With the large dipole moment of the zwitterion, a small change in orientation should be sufficient to increase the surface potential. No clear trend could be established in the results
L.F. Ceridório et al. / Colloids and Surfaces B: Biointerfaces 141 (2016) 595–601
599
Fig. 4. Surface potential-area isotherms for phospholipid monolayers on aqueous subphases with different CS concentrations and incorporation times. (A) DPPC and (B) DPPG. The error bar represents the average of at least 3 measurements, with a standard deviation lower than 2%.
for DPPG. For a waiting time of 1 h, CS caused a small increase in potential at large areas, then large decreases at small areas (high surface pressures). For 4 h, there was a large increase at large areas and a small decrease at small areas. Therefore, the conclusion that can be drawn is that for DPPC CS interacts with the head groups. In order to access information on how chemical groups from the phospholipids and CS are affected by adsorption, PMIRRAS measurements were performed [32,33]. The spectra of the neat phospholipid and mixed monolayers were collected in the wavenumber range from 800 to 4000 cm−1 , at a surface pressure of 30 mN/m, and were subtracted from the spectrum of the free-film interface. Fig. 5 shows bands at 2850 cm−1 and 2916 cm−1 , assigned to the antisymmetric ass (CH2 ) and symmetric s (CH2 ) stretching of methylene groups, respectively. The ratio between the intensi-
ties of these bands can be taken as a parameter of order [34,35]. It increases from 1.74 to 2.81 with CS incorporation in DPPC, while for DPPG it decreases from 1.60 to 1.24. Therefore, CS induces order in the DPPC monolayer, but causes disorder in the DPPG monolayer. Although for both a condensation of the monolayer occurs at 30 mN/m, as observed in the surface pressure-area isotherms, the surface potential for DPPC increased with CS, and this must be related to the increase of lipid order. The lack of significant changes in the position of the bands associated with the tails indicates that CS does not penetrate into either DPPC or DPPG, consistent with the surface pressure isotherms. Therefore, the effects on the packing of the monolayers should be due to interactions with headgroups, as confirmed below.
600
L.F. Ceridório et al. / Colloids and Surfaces B: Biointerfaces 141 (2016) 595–601
Fig. 5. PM-IRRAS spectra for phospholipid monolayers on aqueous subphases containing 1.8 × 10−3 mg/mL of chondroitin sulfate (1 h) at a surface pressure of 30 mN m−1 in the CH asymmetric and symmetric stretching region.
Fig. 6. PM-IRRAS spectra for phospholipid monolayers on aqueous subphases containing 1.8 × 10−3 mg mL−1 of chondroitin sulfate (1 h) at a surface pressure of 30 mN m−1 in the PO2 asymmetric and symmetric stretching region.
The spectra in the region between 1000 and 1800 cm−1 provide information on hydrogen bonds, hydration and conformation of the head group. Fig. 6 features two bands at 1088 cm−1 and 1232 cm−1 assigned to vibrational modes of the phosphate group. For the DPPC monolayer there is an additional band at 1175 cm−1 due to C N stretching. With CS adsorption, the band assigned to asymmetric PO2 stretching (1232 cm−1 ) overlaps the S O stretch band, being relatively more intense. Also, the band at 1170 cm−1 assigned to C
O C stretching for cyclic ethers of CS in the spectrum of DPPG/CS monolayer coincides with the band assigned to C C N amine from DPPC in the spectrum for the DPPC/CS monolayer. Other changes in band intensity and shape are observed, which can be related to interaction of glucoside moieties of CS with the lipid monolayers [26]. These findings confirm the interaction of CS with the head groups. For DPPG, this also happens for the antisymmetric PO2 stretching mode, whose width varies upon incorporation of CS. A band assigned to HO CR2 vibrations from CS appears at 1190 cm−1 , again confirming CS incorporation. Phosphate bands in the region 1000–1100 cm−1 also seem to disappear with CS incorporation. The interaction between CS and headgroups was also manifested in the spectral region between 1600 and 1800 cm−1 (results not shown). For CS-containing monolayers, the band assigned to C O stretching of amide appears as negative band, which point to the vibration transition dipoles for C O groups of amide are oriented preferentially perpendicular to the monolayer. The band at 1740 cm−1 due to ester C O stretching is shifted by 5 and 9 cm−1 for DPPC and DPPG, respectively, owing to hydrogen bonds between CS and phospholipids, while a shoulder appears at 1740 cm−1 for the CS-containing monolayers. We have not analyzed the region between 1650 and 1690 cm−1 because it is considered as interference from water [36,37]. Shifts to lower wavenumbers to the C O stretching in esters are evidence of the interaction of CS with the phospholipid monolayer. In summary, CS does not penetrate in the hydrophobic region of either DPPC or DPPG monolayer, but rather interacts with their head groups, this interaction being modulated by the charges in the polar head. For DPPC, CS induces order which may explain the decrease in pressure in the adsorption kinetics of Fig. 2. In contrast, in DPPG monolayers the incorporation of CS induces disorder, again consistent with the adsorption kinetics data. 4. Conclusions The polysaccharide CS, soluble and with poor surface activity in aqueous solutions, adsorbs at the air-water interface in the presence of DPPC and DPPG monolayers. CS thus has affinity for
L.F. Ceridório et al. / Colloids and Surfaces B: Biointerfaces 141 (2016) 595–601
both membrane phospholipids, but with different means of action. Since DPPC and DPPG have the same alkyl chain, differences in the adsorption profile indicate that the interaction is affected by the polar head. Changes in the PM-IRRAS spectra showed that CS forms a subsurface below the DPPC and DPPG monolayers, with the stronger electrostatic interaction with DPPG causing disorder in the monolayer. We hope that these findings may help understand in detail the interaction of GAGs with membranes at the molecular level, which is of biological interest since many biological processes depend on lipid-GAG interactions. Acknowledgments This work was supported by CAPES, FAPESP (Grant 2013/142627), CNPq and nBioNet network (Brazil). Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/j.colsurfb.2016.02. 030. References [1] H.A. Santos, V. García-Morales, R.-J. Roozeman, J.A. Manzanares, K. Kontturi, Interfacial interaction between dextran sulfate and lipid monolayers: an electrochemical study, Langmuir 21 (2005) 5475–5484, http://dx.doi.org/10. 1021/la046825u. [2] R.M. Lauder, T.N. Huckerby, I.A. Nieduszynski, I.H. Sadler, Characterisation of oligosaccharides from the chondroitin/dermatan sulphates: H and 13 C NMR studies of oligosaccharides generated by nitrous acid depolymerisation, Carbohydr. Res. 346 (2011) 2222–2227, http://dx.doi.org/10.1016/j.carres. 2011.06.033. [3] A. Varki, R. Cummings, J. Esko, H. Freeze, G. Hart, J. Marth, Proteoglycans and Sulfated Glycosaminoglycans—Essentials of Glycobiolog, 2nd edition, Cold Springer Harbor Laboratoty Press, NY-EUA, 1999. [4] R.J. Linhardt, T. Toida, Role of glycosaminoglycans in cellular communication, Acc. Chem. Res. 37 (2004) 431–438, http://dx.doi.org/10.1021/ar030138x. [5] N.S. Gandhi, R.L. Mancera, The structure of glycosaminoglycans and their interactions with proteins, Chem. Biol. Drug Des. 72 (2008) 455–482, http:// dx.doi.org/10.1111/j.1747-0285.2008.00741.x. [6] M. Krumbiegel, K. Arnold, Microelectrophoresis studies of the binding of glycosaminoglycans to phosphatidylcholine liposomes, Chem. Phys. Lipids 54 (1990) 1–7, http://dx.doi.org/10.1016/0009-3084(90)90053-T. [7] K. Nakazawa, K. Murata, K. Izuka, Y. Oshima, The short-term effects of chondroitin sulfates A and C on coronary atherosclerotic subjects: with reference to its anti-thrombogenic activities, Jpn. Heart J. 10 (1969) 289–296. [8] M. McGee, W.D. Wagner, Chondroitin sulfate anticoagulant activity is linked to water transfer: relevance to proteoglycan structure in atherosclerosis, Arterioscler. Thromb. Vasc. Biol. 23 (2003) 1921–1927, http://dx.doi.org/10. 1161/01.ATV.0000090673.96120.67. [9] S. Chen, G. Li, N. Wu, X. Guo, N. Liao, X. Ye, et al., Sulfation pattern of the fucose branch is important for the anticoagulant and antithrombotic activities of fucosylated chondroitin sulfates, Biochim. Biophys. Acta 1830 (2013) 3054–3066, http://dx.doi.org/10.1016/j.bbagen.2013.01.001. [10] D.O. Clegg, D.J. Reda, C.L. Harris, M.A. Klein, J.R.O. Dell, M.M. Hooper, et al., Glucosamine and chondroitin sulfate for knee osteoarthritis, N. Engl. J. Med. 354 (2006) 795–808. [11] N. Volpi, Analytical aspects of pharmaceutical grade chondroitin sulfates, J. Pharm. Sci. 96 (2007) 3168–3180, http://dx.doi.org/10.1002/jps. [12] T.E. McAlindon, M.P. Lavalley, J.P. Gulin, D.T. Felson, Glucosamine and chondroitin for treatment of osteoarthritis, J. Am. Med. Assoc. 283 (2000) 1469–1475. [13] N. Volpi, Anti-inflammatory activity of chondroitin sulphate: new functions from an old natural macromolecule, Inflammopharmacology 19 (2011) 299–306, http://dx.doi.org/10.1007/s10787-011-0098-0. [14] N. Huang, M.-Y. Wu, C.-B. Zheng, L. Zhu, J.-H. Zhao, Y.-T. Zheng, The depolymerized fucosylated chondroitin sulfate from sea cucumber potently inhibits HIV replication via interfering with virus entry, Carbohydr. Res. 380 (2013) 64–69, http://dx.doi.org/10.1016/j.carres.2013.07.010.
601
[15] W. Lian, M. Wu, N. Huang, N. Gao, C. Xiao, Z. Li, et al., Anti-HIV-1 activity and structure-activity-relationship study of a fucosylated glycosaminoglycan from an echinoderm by targeting the conserved CD4 induced epitope, Biochim. Biophys. Acta 1830 (2013) 4681–4691, http://dx.doi.org/10.1016/j. bbagen.2013.06.003. [16] M. Marchetti, E. Trybala, F. Superti, M. Johansson, T. Bergström, Inhibition of herpes simplex virus infection by lactoferrin is dependent on interference with the virus binding to glycosaminoglycans, Virology 318 (2004) 405–413, http://dx.doi.org/10.1016/j.virol.2003.09.029. [17] M. Edidin, Lipids on the frontier: a century of cell-membrane bilayers, Nat. Rev. Mol. Cell Biol. 4 (2003) 414–418, http://dx.doi.org/10.1038/nrm1102. [18] P.L. Yeagle (Ed.), The Structure of Biological Membranes, CRC Press, New York, EUA, 2012. [19] G. Brezesinski, H. Mohwald, Langmuir monolayers to study interactions at model membrane surfaces, Adv. Colloid Interface Sci. 102 (2003) 563–584, http://dx.doi.org/10.1016/S0001-8686(02)00071-4. [20] H. Brockman, Lipid monolayers: why use half a membrane to characterize protein-membrane interactions? Curr. Opin. Struct. Biol. 9 (1999) 438–443. [21] S. Xu, M. Yonese, Two dimensional and three dimensional interactions between bovine serum albumin and chondroitin sulfate, Polym. J. 39 (2007) 298–303, http://dx.doi.org/10.1295/polymj.PJ2006123. [22] N. Volpi, Influence of charge density sulfate group position and molecular mass on adsorption of chondroitin sulfate onto coral, Biomaterials 23 (2002) 3015–3022. [23] J. Ouyang, S. Deng, J. Zhong, Crystallization of calcium oxalate monohydrate at dipalmitoylphosphatidylcholine monolayers in the presence of chondroitin sulfate A, J. Cryst. Growth 270 (2004) 646–654, http://dx.doi.org/10.1016/j. jcrysgro.2004.06.058. [24] R.M. Lauder, T.N. Huckerby, I.A. Nieduszynski, I.H. Sadler, Characterisation of oligosaccharides from the chondroitin/dermatan sulphates: H and 13 C NMR studies of oligosaccharides generated by nitrous acid depolymerisation, Carbohydr. Res. 346 (2011) 2222–2227, http://dx.doi.org/10.1016/j.carres. 2011.06.033. [25] A. Blume, A comparative study of the phase transitions of phospholipid bilayers and monolayers, Biochim. Biophys. Acta: Biomembr. 557 (1979) 32–44, http://dx.doi.org/10.1016/0005-2736(79)90087-7. [26] A.K. de Brito, L. Caseli, C.S.F. Nordi, Interaction of algal polysaccharide with lipid Langmuir monolayers, Mater. Sci. Eng. C 31 (2011) 1857–1860, http://dx. doi.org/10.1016/j.msec.2011.09.002. [27] M. Pinheiro, M. Lu, S. Reis, J.L.F.C. Lima, T. Mart, L. Camacho, et al., Molecular interaction of rifabutin on model lung surfactant monolayers, J. Phys. Chem. B (2012), http://dx.doi.org/10.1021/jp303725j. [28] A.A. Torrano, Â.S. Pereira, O.N. Oliveira Jr., A. Barros-timmons, Probing the interaction of oppositely charged gold nanoparticles with DPPG and DPPC Langmuir monolayers as cell membrane models, Colloids Surf. B: Biointerfaces 108 (2013) 120–126, http://dx.doi.org/10.1016/j.colsurfb.2013.02.014. [29] L. Caseli, C.P. Pascholati, F. Teixeira, S. Nosov, C. Vebert, H.E. Müeller, et al., Interaction of oligonucleotide-based amphiphilic block copolymers with cell membrane models, J. Colloid Interface Sci. 347 (2010) 56–61, http://dx.doi. org/10.1016/j.jcis.2010.03.006. [30] O.N. Oliveira Jr., C. Bonardi, The surface potential of Langmuir monolayers revisited, Langmuir 7463 (1997) 5920–5924. [31] P. Dynarowicz-łatka, A. Dhanabalan, O.N. Oliveira Jr., Modern physicochemical research on Langmuir monolayers, Adv. Colloid Interface Sci. 91 (2001) 221–293, http://dx.doi.org/10.1016/S0001-8686(99)00034-2. [32] L. Caseli, D. Sousa-Martins, M. Maia, A.A.S. Lima-Filho, E.B. Rodrigues, R. Belfort, An intraocular dye solution based on lutein and zeaxanthin in a surrogate internal limiting membrane model: a Langmuir monolayer study, Colloids Surf. B: Biointerfaces 107 (2013) 124–129, http://dx.doi.org/10.1016/ j.colsurfb.2013.01.076. [33] T.E. Goto, L. Caseli, Understanding the collapse mechanism in Langmuir monolayers through polarization modulation-infrared reflection absorption spectroscopy, Langmuir (2013), http://dx.doi.org/10.1021/la402044c. [34] I.W. Levin, T.E. Thompson, Y. Barenholz, C. Huang, Two types of hydrocarbon chain interdigitation in Sphingomyelin bilayers, Biochemistry (1985) 6282–6286, http://dx.doi.org/10.1021/bi00343a036. [35] R.F. Lopez, T.M. Nobre, C.D.M. Accardo, P.C. Pernambuco Filho, H.B. Nader, C.C. Lopes, et al., Effect of carrageenans of different chemical structures in biointerfaces: a Langmuir film study, Colloids Surf. B: Biointerfaces 111 (2013) 530–535, http://dx.doi.org/10.1016/j.colsurfb.2013.06.039. [36] D. Blaudez, T. Buffeteau, B. Desbat, J. Marie, Infrared and Raman spectroscopies of monolayers at the air-water interface, Curr. Opin. Colloid Interface Sci. 4 (1999) 265–272, http://dx.doi.org/10.1016/S13590294(99)90006-7. [37] H. Wang, S. Coss, A. Mudalige, R.L. Polt, J.E. Pemberton, A PM-IRRAS investigation of monorhamnolipid orientation at the air-water interface, Langmuir (2013), http://dx.doi.org/10.1021/la3051725.
