Survey
* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project
* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project
Cytokinesis wikipedia , lookup
Signal transduction wikipedia , lookup
Theories of general anaesthetic action wikipedia , lookup
List of types of proteins wikipedia , lookup
Endomembrane system wikipedia , lookup
Cell membrane wikipedia , lookup
Ethanol-induced non-lamellar phases in phospholipids wikipedia , lookup
electronic-Liquid Crystal Communications April 10, 2007 Piezoelectricity of phospholipids: A possible mechanism for mechano-, and magneto-receptions in biology A. Jákli, J. Harden, C. Notz, C. Bailey Liquid Crystal Institute and Chemical Physics Interdisciplinary Program, Kent State University, Kent, OH 44242, U.S.A. Abstract Classically, the phospholipid bilayer within the cell membrane has been considered as a flexible and self-healing barrier between the inside and outside of the cell, as well as a structural unit to support functional proteins. Here we show that the phospholipids may not play just a passive role, but may act as active transducers. By periodically shearing and compressing films of hydrated L-α-Phosphatidylcholine, we induced tilt of the molecules with respect to the bilayer’s normal and produced electric current perpendicular to the tilt plane. This effect occurs due to the spontaneous assembly of the chiral phospholipid molecules to stack in bilayers, corresponding to a SmA* liquid crystal phase with D∞ symmetry. Tilting the molecules with respect to these layers induces a transition to a ferroelectric SmC* phase with C2 polar symmetry where the polarization is normal to the tilt plane. We find that a 5 degree tilt yields a polarization of 300 nC/cm2 and propose that this piezoelectric effect can couple with functional units and create new ways to convert mechanical stimuli into electric signals within the membrane. We suggest that this coupling allows for a wide variety of sensory possibilities such as mechano-reception (the sensing of mechanical distortions). We also hypothesize that rotation of magnetic particles found in migratory animals can induce local reorientation of the lipid molecules, which via this piezoelectric effect can produce electric signals that may trigger firing of nerve cells, thus allowing navigation based on magnetic information. We demonstrate this hypothesis by generating electric currents in hydrated phospholipids doped with 0.5wt% of ferrofluid of magnetite (Fe3O4) nanoparticles when less than a 100G alternating magnetic field. Introduction Lipids are important components in biological systems1, specifically when dealing with the structure of cellular membranes. The cellular membrane fulfills the following functions critical to cellular survival; it acts as a flexible, self-healing barrier between the cell and its environment and it also acts as a structural unit for functional proteins.2,3 The membrane, however, does not have a purely passive role. There is now recognition of the importance that lipid organization within the cell membranes plays in controlling protein function4 and many disease states have been associated with aberrations of these lipid/protein interactions. A number of these interactions occur via electric signals. For example some membranes swell5 or become birefringent6,7 in response to voltage changes, which was interpreted as a converse piezoelectric8 or electroclinic effect9, respectively. Ion channels sensitive to membrane stretch have been observed in muscle cells10 and piezoelectric models of the outer hair cell composite membranes have been considered.11 These latter models are based on the flexoelectric properties of the lipid bilayers,12 which are related to membrane curvature resulting in polarization normal to the lipid bilayers as discussed extensively by Petrov.3 Using liquid crystal terminology, a lipid bilayer can 1 http://www.e-lc.org/docs/2007_04_09_12_01_56 electronic-Liquid Crystal Communications April 10, 2007 be considered as a SmA* phase (Sm stands for smectic, which indicates layered structure; A means that the average molecular orientation (director) is normal to the layers, and * indicates that the constituent molecules are chiral). Such a phase has D∞ symmetry and therefore must be piezoelectric, because by tilting the molecules, induced by shear and/or layer compression, one induces a SmC* phase, which has polar C2 symmetry with the polar axis normal to the tilt plane.13 In this paper we concentrate on this piezoelectric property of bare lipid bilayers. We show that piezoelectricity of lipid bilayers might explain previously observed ferroelectric-like behavior14 of ion channels and may have consequences in various sensory mechanisms. As opposed to the flexoelectric polarization, here the tilt induced polarization occurs within the insulating chains of the bilayers and therefore cannot be screened out by free ions of the surrounding aqueous plasma. We studied these properties using a hydrated phospholipid extract of egg yolk from Avanti Inc., which forms a stable SmA* liquid crystal phase in bulk over a wide temperature range including the room temperature.15 The pictorial representation of the physical mechanism of the piezoelectricity, and the molecular structure of the major phospholipid component of the mixture, L-α-Phosphatidylcholine, are illustrated in Figure 1. Figure 1: Illustration of the molecular structure of phospholipid L-α-Phosphatidylcholine and of the piezoelectricity of a lipid bilayer. A tilt of the average molecular orientation (director) with respect to the layer normal, induced by mechanical shear and/or layer compression, leads to a SmC* configuration with polarization normal to the tilt (shear) plane. We assert that the lipid bilayers of cell membranes have this local SmA* structure, suggesting that real biological membranes are piezoelectric and electric charges are generated along the membrane when the lipids become tilted due to mechanical stimuli. Furthermore, we believe that cell membrane piezoelectricity may have numerous applications in biological processes. For example, it could explain communication between proteins embedded in cell membranes, or allow the conversion of external stimuli to electric signals within sensory proteins. One specific sensory mechanism that we suspect might be possible is magneto-reception, where animals use ferromagnetic particles to sense local changes in magnetic fields. Homing pigeons, as well as many other animals have the ability to travel long distances without landmarks and arrive at their destinations with very high accuracy.16 For example, salmon, having spent most of their life in 2 http://www.e-lc.org/docs/2007_04_09_12_01_56 electronic-Liquid Crystal Communications April 10, 2007 the ocean will one day return to their place of birth in a mountain stream far away. The artic tern travels 20,000km one way from Antarctica to their breeding grounds in the Artic. The fact that magnetite (Fe3O4) is found within all of these animals, indicate that the Earth’s magnetic fields might play some role in their ability to navigate (magnetoreception).17 Magnetoreception can provide an animal with both orientational and positional information. Although the use of magnetoreception for directional information in migratory animals is well established experimentally, it is not clear by which biophysical mechanism magnetoreception is achieved. However it is known that the magnetic particles form chains18 linked by microtubule-like strands to a few ion-channels in the membrane of the receptor cell. We hypothesize therefore, that the rotation of magnetic particles cause mechanical deformations of the membrane by selectively pulling the microtubules, thus inducing piezoelectric signals, which in turn, triggers the ion channels that fire nerve cells. Piezoelectricity is sensitive to both the direction and magnitude of the mechanical strains, thus by comparing the electric signals in different channels the brain can calculate the direction and magnitude of the local magnetic fields. Experiments To test these ideas we have carried out measurements using two experimental setups shown in Figure 2a and 2b for piezoelectricity and magneto-reception measurements, respectively. a b Figure 2: Illustration of the experiment arrangements used in this paper. (a): A setup used for piezoelectric measurements. The upper part illustrates the sample holder with the piezo-actuators and piezo-detectors. The lower part illustrates the alignment and electrode arrangement compared to the smectic layers and mechanical deformation directions. The electrodes running in the y direction have diameters of 60μm (this determines the film thickness) and are separated by 2mm. The lateral dimension in y direction is 1cm. This means that the area where the induced charge is collected is A=60μm x 1 cm= 0.6mm2. (b): Schematics of experimental setup to detect magnetic field – induced electric signals in L-α Phosphatidylcholine doped with 0.5wt% ferrofluid. The magnetic field, shear direction and the electrode wires are all parallel to each other. The top wire from the temperature controller is a thermistor. The circle in the center of the magnet represents the bottom heating element (top not shown). 3 http://www.e-lc.org/docs/2007_04_09_12_01_56 electronic-Liquid Crystal Communications April 10, 2007 For the piezoelectric measurements (Figure 2a) the liquid crystal material is sandwiched between two glass plates (2 and 2’) that are firmly fixed to temperature stabilized heaters (1 and 1’). A pair of piezoelectric plates (5,6), sensitive to forces in the lateral directions, is attached to the bottom of the lower heated plate, which is connected to a rigid frame (8) via three layers of piezo sensors (5,6,7) sensitive in orthogonal directions. The frame also holds a piezoelectric actuator (9) ((PSt 500/10/5 from Piezomechanik GmbH), which can shift the top plate with maximum amplitude of 5μm. The actuator is driven by a high voltage amplifier (LE 430/015 from Piezomechanik GmbH). The motion of the top plate is monitored by a piezoelectric accelerometer (BK 4375 from Bruel & Kjaer, sensitivity 0.1mm/s2). The sample holder is placed in a polarizing microscope (10), which enables textural observations of the sample during the measurements. One can make controlled periodic shear deformations of a given frequency and detect the induced corresponding current using a lock-in amplifier (7265 DSP from PerkinElmer). As shown in Figure 3a, an important feature of the vibration created by this setup is that it is along the y direction only at frequencies below 200 Hz, and at higher frequencies there are number of resonances where the vertical vibration becomes generated with basically similar amplitudes. 400 vertical horizontal 300 0.4 0 o 71 C o 62 C o 53 C o 23 C 2 0.8 P (nC/cm ) Dispalcement (μm) 1.2 200 100 0 200 400 600 Frequency (Hz) 800 1000 0 0 200 400 600 Frequency (Hz) 800 1000 a b Figure 3: (a): Frequency dependences of the amplitudes of the vibrations of the top plate in vertical (Δx) and horizontal (Δy) displacements. (b): The frequency dependence of the induced electric polarization P calculated from the piezo-current I as P = I /( A ⋅ ω ) , where A~0.6mm2 is the area of the electrode and ω is the angular frequency. Comparing Figure 3a and b, one can see that the peak positions measured in the vibration of the top plate and of the induced currents correspond to each other (the correspondence is more evident at low frequencies). The frequency dependence of the induced polarization (see Figure 3/b) shows that the response is decreasing toward lower temperatures as the material becomes stiffer. We note that the modulus of the piezo actuator is much larger than of the lipid even at room temperature and the measured displacement of the top plate is basically temperature independent. The amplitude of the current measured at different runs increases after the initial frequency scans. Simultaneous textural observations (see Figure 4) revealed improving homeotropic alignment during the vibration of the upper plate. 4 http://www.e-lc.org/docs/2007_04_09_12_01_56 electronic-Liquid Crystal Communications April 10, 2007 It is important to note that the induced polarization at low frequencies is much smaller than at the resonances, where in addition to the horizontal vibration, vertical motion takes place, too. This suggests that the induced tilt angle is determined more by the vertical than the horizontal vibrations. Indeed a vertical displacement Δx causes a decrease of the layer spacing l by Δl = l ⋅ Δx / d , where d =60μm is the film thickness. Assuming rigid molecules the induced tilt angle θ is determined from the relation cos θ = 1 − Δl / l = 1 − Δx / d . Taking the largest Δx~0.5μm measured at 200Hz (see Figure 3/a) this provides θ~7o. However, the maximum contribution of the shear (horizontal vibration by Δy) to the induced tilt would be θy~Δy/d, which is 1.2μm/60μm=1/50 ~1.1o. The effect of the horizontal shear therefore is more to uniformly direct the buckling of the molecules induced by the vertical vibration. a b Figure 4: Texture of a 60μm film seen in polarizing microscope between crossed polarizers. (a): Prior to the first frequency scan. (b): After the third frequency scan. Uniform dark texture indicates homeotropic alignment (lipid molecules perpendicular and smectic layers parallel to the film substrates). Length of bar corresponds to 100μm. 20 2 P (nC/cm ) 15 10 5 0 0 1 2 3 Tilt angle (degrees) 4 5 Figure 5: Tilt angle dependence of the induced polarization at room temperature. 5 http://www.e-lc.org/docs/2007_04_09_12_01_56 electronic-Liquid Crystal Communications April 10, 2007 The tilt angle dependence of the induced polarization at room temperature is plotted in Figure 5. It can be seen that there is a hysteresis, which probably is related to the effect of the vertical vibration on the homeotropic alignment. In increasing the amplitude of vibrations the alignment becomes more and more uniform resulting in the increasing slope at the function of tilt angle. In decreasing amplitudes the slope is almost constant below 3 degrees showing that the homeotropic alignment basically stays. The saturation above 4 degrees indicates that in this range the polarization is not a linear function of the tilt angle anymore. After achieving these unambiguous piezoelectric responses we tested the hypothesis about the magneto-reception mentioned in the introduction. For this we have measured the generation of an electric current by using a periodically varying magnetic field (see setup in Figure 2/b) with L-α –Phosphatidylcholine doped with 0.5wt% of ferrofluid which contain magnetite (Fe3O4) nanoparticles in oleic acid surfactant solution. The experimental arrangement contains two tungsten shims enclosing the phospholipids sandwiched between two glass slides. The tungsten acts as both 25µm spacers and as the electrodes. This forms a cell with and effective area of 2cm x 25µm (0.5mm2). For control measurements empty cells, pure ferrofluid and pure phospholipid were also tested. The measurements could be carried out at various temperatures controlled by an Omega CN8500 temperature controller, and the setup allows shearing of the cell for alignment purposes. The electromagnet was made by reworking a 110V to 220V transformer by removing the iron core and reconfiguring the primary and secondary coils to act as one large coil. The electromagnet was driven by a Regent home theater system model HT-391 amplifier allowing the magnetic field to vary up to 100G. Similar to the piezoelectric measurements, the input signal was provided and the current produced by the sample was measured by the lock-in Amplifier 7265 DSP from PerkinElmer allowing phase sensitive detecting of electric current at the frequency of the magnetic field. The main results of the magnetic measurements are summarized in Figure 6, where the magnetic field dependence of the induced currents are shown for the shear aligned and nonsheared phospholipid doped with 0.5wt% ferrofluid in comparison with the control groups. One sees the effect is largest for the aligned mixture and is much smaller for the control group. All the phospholipid containing samples have a constant background of about 3pA which is most likely due to thermally driven director fluctuations. Not surprisingly, the response is independent of the magnetic field in the pure phospholipids and increases with the magnetic field in the mixtures doped with ferrofluid. We also note that the slight linear increase for the pure ferrofluid up to about 2 pA at 100G is probably due to magnetic induction. The response in the aligned mixture is about 1 order of magnitude larger than the background and it can be related to the magnetic particles - induced director realignment, which was optically detected. The director tilt induces the electric current by the same piezoelectric mechanism we have illustrated in Figure 1 and measured above. The temperature dependence of the induced electric signal is shown in Figure 6/b. Similar to the piezoelectric observations on the pure phospholipids (see Figure 3/b), the response decreases at lower temperatures, indicating that the material becomes more and more rigid against mechanical deformations. 6 http://www.e-lc.org/docs/2007_04_09_12_01_56 electronic-Liquid Crystal Communications April 10, 2007 20 20 Shear aligned lipid+ferrofluid Not Sheared lipid+ferrofluid pure phospholipid empty cell ferrofluid 15 Current (pA) Current (pA) 15 10 5 0 10 5 0 20 40 60 80 Magnetic Field (Gauss) 100 0 60 80 100 120 o Temperature ( C) a b Figure 6 : Electric current induced by 1.1Hz periodic magnetic field that causes reorientation of the director in the plane along the electrode wires. (a): Magnetic field dependence of the phospholipid – 0.5wt% ferrofluid mixtures in different alignments and of the control groups; (b): The temperature dependence of the current of the aligned mixture at 100G. To summarize, we have unambiguously demonstrated the piezoelectricity of bare lipid bilayers. In addition this current appears within the structured part of the membrane, so they are normally not screened out by ions surrounding the membrane; therefore the lateral background noise level is lower as indeed found on the control phospholipid samples. We have also clearly shown the effect of magnetic particles on electric responses of bilayer systems, supporting the magnetite model for magneto-reception. Although our magnetically induced signals are small, we emphasize that in real receptor cells the deformations are much larger than in the experiments we have used, so they may result in stronger localized signals. Accordingly, we are convinced that study of the piezoelectric signals within the cell membrane is worthwhile and may shed light on a number of biophysical processes involving signaling within the cell membranes. Acknowledgement: The work was partially supported by NSF DMS 0456221. We thank to Professor Philip Westerman for helpful discussions, Dr. G. Liao, T. Heatdrech and C. Braganza for their involvement in early stages of the experiments. 7 http://www.e-lc.org/docs/2007_04_09_12_01_56 electronic-Liquid Crystal Communications April 10, 2007 References: 1 H. Ellens, J. Bentz., F.C. Szoka, Biochem. 25, 4141-4147, (1986); V. Vill, H.M. von Minden, M.H.J. Koch, U. Seydel, K. Brandenburg, Chem. Phys. Lipids 104, 75, (2000); W.Curatolo, Biochim/Biophys Acta 906, 111, (1987) 2 I.W. Hamley, “Introduction to soft matter”, Wiley, Chichester (2000) 3 A.G. Petrov, “The lyotropic state of matter”, Gordon and Breach, Singapore (1999); Biochimica et Biophysica Acta 1561 1 (2001) and Analytica Chimica Acta 568, 70 (2006) 4 F. R. Maxfield, I. Tabas, Nature 438, 612 (2005) 5 K. Iwasa, I. Tasaki, R.C. Gibbons, Science, 210, 338 (1980) 6 M.E. Lines, A.M. Glass, “Principles and applications of ferroelectrics and related materials”, Clarendon Press, Oxford, (1977) 7 L.B. Cohen, B. Hille, R.D. Keynes, J. Physiol., 211, 495 (1971) 8 H.R. Leuchtag, J. Theor. Biol., 127, 321 (1987) 9 I. Ermolina, A. Strinskovski, A. Lewis, Y. Feldman, Jour. Phys. Chem., B 105 (14), 2673 (2001) 10 B. Hille, “Ionic channels of excitable membranes”, Sinauer, Sunderland (1992) 11 A.A. Spector, N. Deo, K. Grosh, J.T. Ratnanather, R.M. Raphael, J. Membrane Biol., 209, 135 (2006) 12 A.G. Petrov, “Physical and chemical bases of biological information transfer”, (J. Vassileva, ed), Plenum Press, New York, pp. 167 (1975) 13 A. Jákli, A. Saupe, One and Two Dimensional Fluids: Properties of Smectic, Lamellar, and Columnar Liquid Crystals, 74-80, Taylor & Francis (2006) 14 L. Beresnev, L.M. Blinov, Mendellev J. All-Union Chem Soc., 28, 149 (1982) 15 J. W. Goodby, “Liquid Crystals and Life”, Liquid Crystals, 24, 25 (1998) 16 Diebel, C.E., Proksch, R., Green, C.R., et al., Nature 406 299-302 (2000). 17 Wiltschko, R., Wiltschko, W., J. Comp. Physiol A, 191, 675 (2005); Magnetoreception, BioEssays 28 (2), 157 (2006.) 18 M.M. Walker, T.E. Dennis, J.L. Kirschvink, Neurobiology, 12, 735 (2002) 8 http://www.e-lc.org/docs/2007_04_09_12_01_56