pubs.acs.org/Langmuir © 2009 American Chemical Society
Phospholipid-Induced Fibrillation of a Prion Amyloidogenic Determinant at the Air/Water Interface Jerzy Dorosz, Roman Volinsky, Ehud Bazar, Sofiya Kolusheva, and Raz Jelinek* Department of Chemistry and Ilse Katz Institute of Nanotechnology, Ben Gurion University, Beer Sheva 84105, Israel Received May 18, 2009. Revised Manuscript Received June 16, 2009 The peptide fragment 106-126 of prion protein [PrP(106-126)] is a prominent amyloidogenic determinant. We present analysis of PrP(106-126) fibrillation at the air/water interface and, in particular, the relationship between the fibrillation process and interactions of the peptide with phospholipid monolayers. We find that lipid monolayers deposited at the air/water interface induce rapid formation of remarkably highly ordered fibrils by PrP(106-126), and that the extent of fibrillation and fiber organization were dependent upon the presence of negatively charged and unsaturated phospholipids in the monolayers. We also observe that fibrillation was enhanced when PrP(106-126) was injected underneath preassembled phospholipid monolayers, compared to deposition and subsequent compression of mixed monolayers of the peptide and phospholipids. In a broader context, this study demonstrates that Langmuir systems constitute a useful platform for studying lipid interactions of amyloidogenic peptides and lipid-induced fibrillation phenomena.
Introduction Prion diseases encompass several devastating neurological conditions which are believed to be caused by infectious misfolded protein species 1. The primary pathological feature of prion diseases is the conversion of the normal cellular prion protein (PrPc) to the aberrant isoform (PrPsc) in a post-translational process involving significant secondary structure changes 2. The abnormal, protease-resistant isoform is rich in beta-sheet and aggregates into amyloid fibrils which are often found in brains of affected persons 2. A fragment of human prion protein spanning residues 106-126 [amino-acid sequence KTNMKHMAGAAAAGAVVGGLG, referred to as PrP(106-126)] has been a widely used model peptide for studying the pathogenic characteristics of full-length PrP because it retains important features of the full-length protein. In particular, PrP(106-126) is highly amyloidogenic 3 and neurotoxic 4-6. While the occurrence of amyloid fibrils in prion diseases is common, the exact functions of these structures as toxic factors or their direct pathological consequences are still unclear 7. In recent years it has become increasingly apparent that membrane interactions of prion proteins and other amyloidogenic *To whom correspondence should be addressed. E-mail:
[email protected].
(1) DeArmond, S. J.; Prusiner, S. B. Am. J. Pathol. 1995, 146, 785–811. (2) Prusiner, S. B. Proc. Natl. Acad. Sci. U. S. A. 1998, 95, 13363–13383. (3) Tagliavini, F.; Prelli, F.; Verga, L.; Giaccone, G.; Sarma, R.; Gorevic, P.; Ghetti, B.; Passerini, F.; Ghibaudi, E.; Forloni, G..; et al. Proc. Natl. Acad. Sci. U. S. A. 1993, 90, 9678–9682. (4) Brown, D. R.; Schmidt, B.; Kretzschmar, H. A. Nature 1996, 380, 345–347. (5) Forloni, G.; Angeretti, N.; Chiesa, R.; Monzani, E.; Salmona, M.; Bugiani, O.; Tagliavini, F. Nature 1993, 362, 543–546. (6) Perez, M.; Wandosell, F.; Colac-o, C.; Avila, J. Biochem. J. 1998 335 (Pt 2), 369-374. (7) Caughey, B.; Lansbury, P. T. Annu. Rev. Neurosci. 2003, 26, 267–298. (8) Baron, G. S.; Wehrly, K.; Dorward, D. W.; Chesebro, B.; Caughey, B. EMBO J. 2002, 21, 1031–1040. (9) Pinheiro, T. J. T. Chem. Phys. Lipids 2006, 141, 66–71. (10) Wang, F.; Yang, F.; Hu, Y.; Wang, X.; Wang, X.; Jin, C.; Ma, J. Biochemistry 2007, 46, 7045–7053. (11) Critchley, P.; Kazlauskaite, J.; Eason, R.; Pinheiro, T. J. T. Biochem. Biophys. Res. Commun. 2004, 313, 559–567.
Langmuir 2009, 25(21), 12501–12506
proteins play important roles in fibrillation and toxicity 8-13. Indeed, toxic effects of amyloidogenic peptides are increasingly ascribed to pore formation and other bilayer disruption events 14. There are however indications that fibrillar aggregates or the fibrillation process itself may also be cytotoxic 15-17. Previous studies have revealed the lipid molecules intimately affect the misfolding pathways of amyloidogenic proteins 18,19. Binding of amyloidogenic peptides to specific lipid molecules and membrane domains has been widely observed 9,20-22, and several theories propose that membrane-induced fibrillation is a critical initial step in amyloidogenesis 23-25. In particular, a number of studies point to the central role of negatively charged phospholipids in fibrillation and aggregation of amyloid fibrils 23,26,27. (12) Kazlauskaite, J.; Sanghera, N.; Sylvester, I.; Venien-Bryan, C.; Pinheiro, T. J. T. Biochemistry 2003, 42, 3295–3304. (13) Matsuzaki, K. Biochim. Biophys. Acta 2007, 1768, 1935–1942. (14) Valincius, G.; Heinrich, F.; Budvytyte, R.; Vanderah, D. J.; McGillivray, D. J.; Sokolov, Y.; Hall, J. E.; L€osche, M. Biophys. J. 2008, 95, 4845–4861. (15) Okada, T.; Wakabayashi, M.; Ikeda, K.; Matsuzaki, K. J. Mol. Biol. 2007, 371, 481–489. (16) Novitskaya, V.; Bocharova, O. V.; Bronstein, I.; Baskakov, I. V. J. Biol. Chem. 2006, 281, 13828–13836. (17) Engel, M. F. M.; Khemtemourian, L.; Kleijer, C. C.; Meeldijk, H. J. D.; Jacobs, J.; Verkleij, A. J.; de Kruijff, B.; Killian, J. A.; H€oppener, J. W. M. Proc. Natl. Acad. Sci. U. S. A. 2008, 105, 6033–6038. (18) Jayasinghe, S. A.; Langen, R. Biochemistry 2005, 44, 12113–12119. (19) Gorbenko, G. P.; Kinnunen, P. K. J. Chem. Phys. Lipids 2006, 141, 72–82. (20) Meng, X.; Fink, A. L.; Uversky, V. N. J. Mol. Biol. 2008, 381, 989–999. (21) Avdulov, N. A.; Chochina, S. V.; Igbavboa, U.; Warden, C. S.; Vassiliev, A. V.; Wood, W. G. J. Neurochem. 1997, 69, 1746–1752. (22) Micelli, S.; Meleleo, D.; Picciarelli, V.; Gallucci, E. Biophys. J. 2004, 86, 2231–2237. (23) Aisenbrey, C.; Borowik, T.; Bystr€om, R.; Bokvist, M.; Lindstr€om, F.; Misiak, H.; Sani, M.; Gr€obner, G. Eur. Biophys. J. 2008, 37, 247–255. (24) Bystr€om, R.; Aisenbrey, C.; Borowik, T.; Bokvist, M.; Lindstr€om, F.; Sani, M.; Olofsson, A.; Gr€obner, G. Cell. Biochem. Biophys. 2008, 52, 175–189. (25) Murray, I. V. J.; Liu, L.; Komatsu, H.; Uryu, K.; Xiao, G.; Lawson, J. A.; Axelsen, P. H. J. Biol. Chem. 2007, 282, 9335–9345. (26) Zhao, H.; Tuominen, E. K. J.; Kinnunen, P. K. J. Biochemistry 2004, 43, 10302–10307. (27) Olofsson, A.; Borowik, T.; Gr€obner, G.; Sauer-Eriksson, A. E. J. Mol. Biol. 2007, 374, 186–194.
Published on Web 07/09/2009
DOI: 10.1021/la901750v
12501
Article
Dorosz et al.
PrP(106-126), in particular, has been shown to exhibit significant membrane interactions 28-30. These interactions have been implicated as a prominent factor in the neurotoxicity of the peptide 30-32; however, the exact mechanisms of membrane binding of PrP(106-126) and the structural outcome of lipid interactions of this fragment are still elusive 29. In particular, the relationship between membrane binding of PrP(106-126) and the fibrillation process has not been deciphered. Indeed, a core question pertaining to PrP(106-126) and amyloidogenesis is whether membranes constitute mere platforms for amyloid fibril accumulation, or whether lipid molecules function as initiators for fibrillation. Varied biophysical techniques have been applied for the study of amyloid protein fibrillation and the contribution of lipids to fibrillation phenomena 33,34. Monolayer techniques have been used for analysis of fibrillation of amyloidogenic peptides and proteins. Supported lipid monolayers, for example, have been shown to significantly enhance fibrillation of amyloidogenic proteins associated with type II diabetes 35. Studies of lipidassociated fibrillation phenomena at the air/water interface (i.e., Langmuir monolayers), however, have been rare 36. The scarcity of studies analyzing lipid-associated fibrillation at the air/water interface is somewhat surprising, since lipid Langmuir monolayers constitute a useful platform for mimicking membrane surfaces and for investigation of membrane docking by varied guest molecules 37-40. In particular, numerous reports have analyzed interactions of membrane-associated peptides with Langmuir lipid monolayers 41. Most such studies have explored the interplay between lipid structures and organization on the one hand, and binding/insertion of amphiphilic and hydrophobic peptides into the lipid monolayers on the other hand. This work presents an investigation of lipid-modulated fibrillation of PrP(106-126) at the air/water interface. We observe remarkable macroscopic organization of prion peptide fibrils, highly dependent upon the phospholipids’ charge, saturation, and surface pressure. Furthermore, the experiments show that PrP(106-126) fibrillation is promoted by organized lipid layer, rather than the presence of individual lipid molecules. The monolayer experiments reported here demonstrate that Langmuir analysis is a powerful tool for visualizing the relationship between lipid assemblies and amyloid peptide fibrillation.
of 95% purity. The peptide was dissolved in DMSO and the stock solution at a 5 mM concentration was stored at -20 °C. All phospholipids, including: 1,2-dimyristoyl-sn-glycerophosphocholine (DMPC), 1,2-dimyristoyl-sn-glycero-3-phospho-(10 -rac-glycerol) (DMPG) sodium salt, 1-palmitoyl-2-oleoyl-sn-glycero-3-phosphocholine (POPC), 1-palmitoyl-2-oleoyl-sn-glycero-3-phospho-(10 rac-glycerol) (POPG) sodium salt and L-R-phosphatidilserine from porcine brain (brain PS) sodium salt were purchased from Avanti Polar Lipids, Inc. (Alabster, AL, USA). One millimolar stock solutions were prepared by dissolution of the phospholipids in chloroform/ethanol 1:1 mixtures and were kept at -20 °C. Compression and Adsorption Isotherms. The experiments were performed at 25 °C using a Nima 312D Teflon trough (Nima Technology Ltd., Coventry, UK). The absorption isotherms (Δπ - time) were monitored throughout the duration of the experiment using a Nima PS4, Wilhelmy plate sensor. Lipid monolayers at different surface pressures were formed by deposition of the lipid solutions at the air-water interface of the dipping well (total volume of 50 mL) in the absorption variant of the experiment, whereas in the compression variant the lipid monolayer was deposited on the whole area of the trough (250 cm2). After 15 min of solvent evaporation and equilibration, the peptide was injected to the water subphase, below the preformed lipid monolayer through a thin, vertical tube, to reach a concentration of 500 nM followed by 3-4 h incubation with gentle stirring. The monolayer was compressed at 8 cm2/min speed in compressionvariant only. Film samples were transferred onto a glass surface (for AFM experiments) or copper Formvar/carbon grid (for TEM experiments). Transmission Electron Microscopy (TEM). Samples for TEM analysis were transferred on the copper Formvar/carbon grids (Electron Microscopy Sciences, Hatfield, PA, USA) from the air-water interface of the Langmuir trough using LangmuirBlodgett technique. The grids were dried and negatively stained with a 1% solution of uranyl acetate. TEM images were obtained using Jeol JEM-1230 electron microscope operating at 100 kV. Atomic Force Microscopy (AFM). Samples were collected by transferring the material from the air-water interface onto a glass surface and subsequent drying. AFM measurements were performed at ambient conditions using a Digital Instrument Dimension3100 (Veeco, Santa Barbara, CA, USA) mounted on an active antivibration table. A 100 μm scanner was used. Microfabricated Si oxide NSC11\50 type Ultralsharp with integrated pyramidal tip was used. The 512 512 pixel images were recorded in the tapping mode with a scan size of up to 5 μm at a scan rate of 1 Hz.
Materials and Methods
Attenuated Total Reflection-Fourier Transform Infrared Spectroscopy (ATR-FTIR). Spectra were obtained at 4 cm-1
Materials. PrP(106-126) was obtained from Alpha Diagnostic International Inc. (Dallas, TX, USA) as a lyophilized powder (28) Salmona, M.; Forloni, G.; Diomede, L.; Algeri, M.; De Gioia, L.; Angeretti, N.; Giaccone, G.; Tagliavini, F.; Bugiani, O. Neurobiol. Dis. 1997, 4, 47–57. (29) Miura, T.; Yoda, M.; Takaku, N.; Hirose, T.; Takeuchi, H. Biochemistry 2007, 46, 11589–11597. (30) Dupiereux, I.; Zorzi, W.; Lins, L.; Brasseur, R.; Colson, P.; Heinen, E.; Elmoualij, B. Biochem. Biophys. Res. Commun. 2005, 331, 894–901. (31) Dupiereux, I.; Zorzi, W.; Rachidi, W.; Zorzi, D.; Pierard, O.; Lhereux, B.; Heinen, E.; Elmoualij, B. J. Neurosci. Res. 2006, 84, 637–646. (32) Kourie, J. I.; Culverson, A. J. Neurosci. Res. 2000, 62, 120–133. (33) Komatsu, H.; Liu, L.; Murray, I. V. J.; Axelsen, P. H. Biochim. Biophys. Acta 2007, 1768, 1913–1922. (34) Munishkina, L. A.; Fink, A. L. Biochim. Biophys. Acta 2007, 1768, 1862– 1885. (35) Domanov, Y. A.; Kinnunen, P. K. J. J. Mol. Biol. 2008, 376, 42–54. (36) Lopes, D. H. J.; Meister, A.; Gohlke, A.; Hauser, A.; Blume, A.; Winter, R. Biophys. J. 2007, 93, 3132–3141. (37) Maget-Dana, R. Biochim. Biophys. Acta 1999, 1462, 109–140. (38) Leblanc, R. M. Curr. Opin. Chem. Biol. 2006, 10, 529–536. (39) Volinsky, R.; Kolusheva, S.; Berman, A.; Jelinek, R. Biochim. Biophys. Acta 2006, 1758, 1393–1407. (40) Brockman, H. Curr. Opin. Struct. Biol. 1999, 9, 438–443. (41) Vie, V.; Van Mau, N.; Chaloin, L.; Lesniewska, E.; Le Grimellec, C.; Heitz, F. Biophys. J. 2000, 78, 846–856.
12502 DOI: 10.1021/la901750v
resolution comprising 100 coadded scans at room temperature using a Jasco FT/IR 6300 spectrophotometer (Ophir Analytical Ltd., Israel). The samples were prepared by placing a small volume (1 mL) of 5 μM aqueous solution of the peptide on a ZnSe crystal with an aperture angle of 45°, followed by deposition of the lipid film at the air-water interface of the solution. The crystal was incubated overnight at 25 °C until the lipid-peptide film appeared completely dry. To eliminate carbon monoxide and water vapor signals, the sample was purged with nitrogen for 15 min before and during the measurement. Spectral background was recorded using a clean ATR crystal and the corresponding signal was subtracted from the sample spectrum. The resulting spectrum was then baseline-corrected and further smoothed using a Savitzkey-Golay algorithm with a convolution width of 15.
Results Figure 1 depicts the isothermal adsorption data recorded following injection of PrP(106-126) into the water subphase underneath preformed lipid monolayers. The control experiment in which no phospholipids were deposited prior to injection of PrP(106-126) featured a negligible increase of surface pressure Langmuir 2009, 25(21), 12501–12506
Dorosz et al.
Figure 1. Isothermal adsorption of PrP(106-126) onto phospholipid monolayers. (A) Adsorption isotherms of the peptide injected into the water subphase underneath different lipid monolayers: i. Control (no lipids); ii. DMPC monolayer; iii. DMPC/DMPG (7:3 mol ratio); iv. DMPC/PS (7:3); v. pure DMPG; vi. PS. (B) Absolute change of surface pressure recorded after injection, plotted as a function of initial monolayer pressure when the peptide was injected. X: DMPC monolayer; circles: DMPC: DMPG (7:3 mol ratio); diamonds: DMPG; triangles: PS.
(Figure 1Ai), indicating that the peptide hardly migrated to the air/water interface and remained soluble in water. Very small adsorption of PrP(106-126) was apparent when the peptide was injected underneath a monolayer of dimyristoylphosphatydilcholine (DMPC) (Figure 1Aii), pointing to insignificant attraction of the prion fragment to the zwitterionic DMPC monolayer. The adsorption isotherms of PrP(106-126) injected underneath monolayers containing dimyristoylphosphatydilglycerole (DMPG) or phosphatydilserine (DMPS) indicate that the presence of the negatively charged phospholipids significantly promoted incorporation of the peptide into the phospholipid monolayers. Specifically, following injection of the prion fragment underneath monolayers comprising DMPC/DMPG (Figure 1Aiii), DMPC/PS (Figure 1Aiv), and in particular DMPG (Figure 1Av), and PS (Figure 1Avi), the absorption isotherms featured noticeable increases in surface pressures due to insertion of the peptide into the monolayers. Figure 1B illustrates the relationship between the absolute increase in surface pressure following injection of PrP(106-126) into the water subphase and the initial monolayer pressure. This graphical relationship generally reflects the relative affinities to Langmuir 2009, 25(21), 12501–12506
Article
the Langmuir monolayer; higher values of the absolute increase in surface pressure indicate more pronounced penetration to the lipid monolayer. The broken lines, which were calculated from the experimental data-points, further highlight the affinity of the prion peptide to the negatively charged phospholipids. Specifically, both the extrapolated intersections with the X axis which correspond to the exclusion pressures and the slopes of the lines mirror the extent of adsorption and incorporation of guest molecules into the phospholipid monolayers. The greater the exclusion pressure, the higher the affinity of the tested molecules into the monolayer 24, while steeper slopes point to higher sensitivity of monolayer binding to the surface pressure. The graphic representation in Figure 1B attests to the direct dependence of PrP(106-126) monolayer adsorption upon the presence of negatively charged phospholipids. DMPS/prion interactions appear more pronounced at lower pressures (
