Effect of Heavy Water on Nonlamellar Structures of ... - CSJ Journals

2 downloads 0 Views 348KB Size Report
Sep 29, 2012 - The effect of replacing water (H2O) with heavy water (D2O) on both lamellar and nonlamellar structures formed by phospholipids and ...
doi:10.1246/cl.2012.1101 Published on the web September 29, 2012

1101

Effect of Heavy Water on Nonlamellar Structures of Phospholipid and Monoolein Molecular Assemblies Hiroshi Takahashi* and Kohtaro Jojiki Biophysics Laboratory, Department of Chemistry and Chemical Biology, Gunma University, 4-2 Aramaki-cho, Maebashi, Gunma 371-8510 (Received May 14, 2012; CL-120516; E-mail: [email protected]) The effect of replacing water (H2O) with heavy water (D2O) on both lamellar and nonlamellar structures formed by phospholipids and monoolein was studied by X-ray diffraction methods. Except for lamellar crystal and inverted hexagonal phases, the lattice constants of all phases investigated were decreased. The changes in the cubic phases of monoolein can be interpreted as a reduction in the interfacial area caused by D2O. The largest change was observed in the Im3m bicontinuous cubic phase of monoolein.

Many biochemical processes are sensitive to replacing water (H2O) with heavy water (D2O).1 High concentrations of D2O are toxic for most mammals. D2O retards the circadian rhythm of some animals.2 Detailed molecular mechanisms of these effects are as yet unknown. Experimental studies on the stability of proteins in D2O have suggested that (1) D2O alters hydrophobic interactions in proteins and (2) D2O affects inter- and/or intramolecular hydrogen bonds.3 Lipids are essential components of biomembranes, where important biochemical process are performed. Hence, study of the interactions between D2O and lipids will provide a clearer understanding of the biological effects of D2O. The effect of D2O on the phase transition behaviors of phospholipid bilayers in lamellar phases has been investigated by calorimetry.4­9 These studies indicated that D2O reduces molecular area occupied by the phospholipids at the bilayer interface. Using X-ray diffraction, Kobayashi and Fukada10 proved that D2O does not affect the thickness of phosphatidylcholine (PC) bilayers but results in a slight decrease in the interlamellar distance as compared with that in H2O. This effect of D2O was observed only in a lamellar liquid-crystalline (L¡) phase of PC; no difference could be detected in a lamellar gel (L¢) phase of the phospholipid. Kobayashi and Fukada10 interpreted this result as an effect of the change of undulation forces11 between the bilayers. Some lipids can form various molecular assemblies including micellar, hexagonal, cubic structures, etc. Some cell organelles incorporate nonlamellar lipid structure, including hexagonal and cubic structures that are believed to play an important role in membrane fusion.12 In this report, we studied the effect of substituting of H2O with D2O on both lamellar and nonlamellar structures of phospholipids and monoolein (MO) using X-ray diffraction methods. MO forms bicontinuous cubic phases that have been used for the crystallization of membrane proteins.13 Here, we used dihexadecyl-PC (DHPC) and dielaidoylphosphatidylethanolamine (DEPE) as phospholipids. In addition to normal L¢ and L¡ phases, DHPC14,15 and DEPE16,17 form an interdigitated gel (L¢I) and inverted hexagonal (HII) phases, respectively.

Chem. Lett. 2012, 41, 1101­1103

DHPC, DEPE, and MO were purchased from Fluka (Buchs, Switzerland), Avanti Polar Lipids (Alabaster, AL), and Sigma Chemical Co. (St Louis, USA), respectively. These lipids had purities >99% and were used without further purification. D2O (99.8 atom % D) was purchased from Aldrich (Milwaukee, USA). Samples were prepared by methods described elsewhere,18­21 except that the water was unbuffered here. The lipids were dispersed in excess of pure H2O or D2O. X-ray diffraction measurements were carried out at the beamlines 9C and 15A of the Photon Factory (Tsukuba, Japan). Details of the methods of measurement have been described elsewhere.20,22 The diffraction data were recorded at 0.33 or 1.33 °C intervals during temperature scans at a rate of 2.0 °C min¹1. Figure 1 represents the temperature dependence of the lattice constants of DHPC (a), DEPE (b), and MO (c) dispersed in H2O and D2O. Because of the broadness of the diffraction pattern of DHPC in a ripple phase that appears between L¢I and L¡ phases, we could not determine the unit cell of the ripple phase precisely.14,15 For this reason we did not plot the data in Figure 1. Although, in the presence of excess water under equilibrium conditions, MO only forms lamellar crystal (Lc), bicontinuous cubic (Pn3m), and HII phases,13,23,24 we have reported using temperature scanning measurements that another bicontinuous cubic phase (Im3m) is formed after melting of the Lc phase.21 All data in Figure 1 were obtained with a scan rate of 2.0 °C min¹1. The Im3m phase of MO is a metastable phase. The situation is similar as in the metastable L¢ phase of dilauroyl-PE (DLPE) bilayers.25 The thermodynamically stable phase of fully hydrated DLPE is an Lc phase in the temperature range where a metastable L¢ phase of DLPE is observed.25 The appearance of Im3m phase in MO­water systems was first reported by Caffrey26 in his kinetic study of phase transitions of MO­water systems. Recently, the formation of Im3m phase has also been reported in a cubosome made in a MO­water system by ultrasonic irradiation.27 The transition temperature from Im3m to Pn3m for MO dispersed in D2O is about 5 °C lower than that in H2O. The transition temperature from cubic to HII phase is 95 °C for the excess H2O condition.24 Thus, D2O reduced the transition temperature by about 10 °C. Because the lattice constants are temperature dependent, we calculated their mean values (Table 1). For comparison, the ratios of the mean values of the lattice constants observed in H2O and D2O are also shown in Table 1. Except for the Lc and HII phases, D2O reduced the lattice constants of all phases as compared to those with H2O. In contrast to the previous study of Kobayashi and Fukada,10 a difference was also detected in the L¢ phase. In all phases investigated here, the biggest difference was observed for Im3m cubic phase of MO. D2O induced about 14% decrease in the lattice constants. The second most conspicuous effect was on the Pn3m cubic phase. Previously,

© 2012 The Chemical Society of Japan

www.csj.jp/journals/chem-lett/

1102 Table 1. Comparison of the observed mean values of lattice constants (a) in H2O and D2O

7.5



7.0

Lattice constant / nm

(a)

DHPC DHPC DEPE DEPE DEPE MO MO MO L¢I L¡ L¢ L¡ HII Lc Im3m Pn3m

6.5

a in H2Oa 4.85 6.63 6.45 5.29 a in D2Oa 4.82 6.59 6.40 5.21 D2O/H2Ob 0.994 0.994 0.992 0.985

6.0 5.5

L βI

4.0

Lattice constant / nm

8.0

(b)

H II

7.5 7.0



6.5 6.0



5.5 5.0 4.5

(c)

Lattice constant / nm

20.0

Im 3m Cubic 16.0

12.0

Pn 3m Cubic 8.0

Lc

H

4.0 10

20

5.00 16.42 9.08 5.00 14.12 8.28 1.00 0.860 0.912

a The units are nanometers (nm). bThe ratios of lattice constants in D2O to those in H2O.

5.0 4.5

7.37 7.37 1.00

30

40

50

60

70

II

80

90

Temperature / °C

Figure 1. Temperature dependence of lattice constants of various phases of (a) DHPC, (b) DEPE, and (c) MO in excess H2O (red open circles ) and excess D2O (blue close circles ). our group reported that the lattice constants of cubic phases of MO or monoelaidin are extremely sensitive to the presence of cosolute.18­21 The lattice constants of the lamellar and HII phases of phospholipids are also affected by the presence of cosolute.15,28 However, the change observed in the lamellar and HII phases is significantly less than those in the cubic phases. The interfacial area between the lipid and the aqueous phase for a lipid of the HII phase is smaller than that for a lipid of the Pn3m cubic phase. Hence, the lowering of the Pn3m-to-HII phase transition temperature by D2O means that D2O reduces the interfacial area of MO. The decrease in the lattice constants of the cubic phase can also be interpreted as a reduction in the interfacial area caused by D2O, based on the geometry of the cubic phases.18­21 This is consistent with the conclusion derived from phase transition behavior of phospholipids,4­9 indicating that this effect of D2O is not sensitive to the type of polar headgroups in the lipids. The decreasing area effect caused by D2O may be related to the change in inter- and/or intramolecular Chem. Lett. 2012, 41, 1101­1103

hydrogen bonds induced by D2O. Probably because the packing of headgroups in the HII phase is already too tight so as to reduce the interfacial area, D2O cannot change the lattice constant of the HII phase. Finally, let us briefly discuss the biological relevance of the effect of D2O on the interfacial area of lipid bilayers. As described above, D2O generally has a deleterious effect on many biological processes. The reduction of interfacial area is equivalent to the increasing of lipid molecular packing density in the bilayers. This would reduce the dynamics or fluidity of the lipid membranes. It is also expected that the reduction in interfacial area increases the lateral pressure inside bilayers. An appropriate lateral pressure is believed to be important for the functioning of membrane proteins.29 Hence, it is conceivable that the harmful biological influence of D2O may be mediated by the changes in the physical properties needed for the membranes to perform their required functions. A direct effect of D2O on the protein structure and function must also be considered in explaining the toxic effects of D2O. In conclusion, D2O affects the lateral interaction at the interface between the headgroups of the lipids and the aqueous phase. From the present and previous18­21 studies, it can be also concluded that the lattice constant of the bicontinuous cubic phase of monoacylglycerols is a sensitive parameter to measure the effect of interfacial molecular interaction of cosolutes or solvents on the lipid­water systems. We thank Prof. M. Nomura and Mr. A. Koyama (Photon Factory, KEK) for their help in measurements of the beamline 9C at the Photon Factory. X-ray measurements were performed under approval of the Photon Factory Program Advisory Committee (Proposal No. 2007G656). We also thank Prof. P. J. Quinn (King’s College London) for the English revision of the manuscript. Paper based on a presentation made at the International Association of Colloid and Interface Scientists, Conference (IACIS2012), Sendai, Japan, May 13­18, 2012. References and Notes 1 a) J. J. Kats, Sci. Am. 1960, 203, 106. b) P. R. Gross, W. Spindel, Science 1960, 131, 37. c) J. F. Thomson, Biological Effects of Deuterium, Pergmon Press, New York, 1963, pp. 1­133. 2 a) C. S. Pittendrigh, P. C. Caldarola, E. S. Cosbey, Proc. Natl. Acad. Sci. U.S.A. 1973, 70, 2037. b) C. P. Richter, Proc. Natl. Acad. Sci. U.S.A. 1977, 74, 1295. 3 Y. Cho, L. B. Sagle, S. Iimura, Y. Zhang, J. Kherb, A. Chilkoti, J. M. Scholtz, P. S. Cremer, J. Am. Chem. Soc.

© 2012 The Chemical Society of Japan

www.csj.jp/journals/chem-lett/

1103 4 5 6 7 8 9 10 11 12

13 14 15

2009, 131, 15188, and references therein. C.-H. Chen, J. Phys. Chem. 1982, 86, 3559. G. Lipka, B. Z. Chowdhry, J. M. Sturtevant, J. Phys. Chem. 1984, 88, 5401. L. D. Ma, R. L. Magin, G. Bacic, F. Dunn, Biochim. Biophys. Acta, Biomembr. 1989, 978, 283. K. Ohki, Biochem. Biophys. Res. Commun. 1991, 174, 102. K. Kinoshita, M. Yamazaki, Biochim. Biophys. Acta, Biomembr. 1996, 1284, 233. H. Matsuki, H. Okuno, F. Sakano, M. Kusube, S. Kaneshina, Biochim. Biophys. Acta, Biomembr. 2005, 1712, 92. Y. Kobayashi, K. Fukada, Chem. Lett. 1998, 1105. J. N. Israelachvili, Intermolecular and Surface Forces, 2nd ed., Academic, New York, 1991, p. 395. S. Hyde, S. Andersson, K. Larsson, Z. Blum, T. Landh, S. Lidin, B. W. Ninham, The Language of Shape: The Role of Curvature in Condensed Matter: Physics, Chemistry and Biology, Elsevier, Amsterdam, 1997. M. Caffrey, Biochem. Soc. Trans. 2011, 39, 725, and references therein. P. Laggner, K. Lohner, G. Degovics, K. Müller, A. Schuster, Chem. Phys. Lipids 1987, 44, 31. H. Takahashi, H. Ohmae, I. Hatta, Biophys. J. 1997, 73, 3030.

Chem. Lett. 2012, 41, 1101­1103

16 R. Koynova, M. Caffrey, Chem. Phys. Lipids 1994, 69, 1. 17 H. Takahashi, K. Sinoda, I. Hatta, Biochim. Biophys. Acta, Gen. Subj. 1996, 1289, 209. 18 H. Takahashi, A. Matsuo, I. Hatta, Mol. Cryst. Liq. Cryst. Sci. Technol., Sect. A 2000, 347, 231. 19 H. Takahashi, A. Matsuo, I. Hatta, Phys. Chem. Chem. Phys. 2002, 4, 2365. 20 S. Abe, H. Takahashi, J. Appl. Crystallogr. 2003, 36, 515. 21 S. Abe, H. Takahashi, Chem. Phys. Lipids 2007, 147, 59. 22 Y. Amemiya, K. Wakabayashi, T. Hamanaka, T. Wakabayashi, T. Matsushita, H. Hashizume, Nucl. Instrum. Methods Phys. Res. 1983, 208, 471. 23 H. Qiu, M. Caffrey, Biomaterials 2000, 21, 223. 24 C. Czeslik, R. Winter, G. Rapp, K. Bartels, Biophys. J. 1995, 68, 1423. 25 J. M. Seddon, K. Harlos, D. Marsh, J. Biol. Chem. 1983, 258, 3850. 26 M. Caffrey, Biochemistry 1987, 26, 6349. 27 A. Tilley, Y.-D. Dong, H. Amenitsch, M. Rappolt, B. J. Boyd, Phys. Chem. Chem. Phys. 2011, 13, 3026. 28 P. W. Sanderson, L. J. Lis, P. J. Quinn, W. P. Williams, Biochim. Biophys. Acta, Biomembr. 1991, 1067, 43. 29 B. de Kruijff, Curr. Opin. Chem. Biol. 1997, 1, 564.

© 2012 The Chemical Society of Japan

www.csj.jp/journals/chem-lett/