Lars D. Mosgaard

Lipid Ion Channels and the Role of Proteins

In the absence of proteins, synthetic lipid membranes can display quantized conduction events for ions that are virtually indistinguishable from those of protein channels. The phenomenological similarities between typical conductances are striking: they are of equal order and show similar lifetime distributions and current histograms. They can include conduction bursts, flickering, and multistep conductance. Lipid channels can be gated by voltage and blocked by drugs. They respond to changes in lateral membrane tension and temperature. Thus, they behave like voltage-gated, temperature-gated, and mechano-sensitive protein channels, or like receptors. The similarity between lipid and protein channels poses an important problem for the interpretation of protein channel data. For example, the Hodgkin-Huxley theory for nerve pulse conduction requires a selective mechanism for the conduction of sodium and potassium ions. To this end, the lipid membrane must act both as a capacitor and as an insulator. Nonselective ion conductance by mechanisms other than the gated protein channels challenges the proposed mechanism for pulse propagation. Nevertheless, textbooks rarely describe the properties of the lipid membrane surrounding the proteins in their discussions of membrane models. These similarities lead to important questions: Do these similarities in lipid and protein channels result from a common mechanism, or are these similarities fortuitous? What distinguishes protein channels from lipid channels, if anything? In this Account, we document experimental and theoretical findings that show the similarity between lipid and protein channels. We discuss important cases where protein channel function strongly correlates with the properties of the lipid. Based on statistical thermodynamics simulations, we discuss how such correlations could come about. We suggest that proteins can act as catalysts for lipid channel formation and that this hypothesis can explain some of the unexplained correlations between protein and lipid membrane function.

Penetration of Action Potentials During Collision in the Median and Lateral Giant Axons of Invertebrates

The collisions of two simultaneously generated impulses in the giant axons of both earthworms and lobsters propagating in orthodromic and antidromic direction are investigated. The experiments have been performed on the extracted ventral cords of Lumbricus terrestris and the abdominal ventral cord of a lobster, Homarus americanus, by using external stimulation and recording. The collision of two nerve impulses of orthodromic and antidromic propagation did not result in the annihilation of the two signals, contrary to the common notion that is based on the existence of a refractory period in the well-known Hodgkin-Huxley theory. However, the results are in agreement with the electromechanical soliton theory for nerve-pulse propagation, as suggested by Heimburg and Jackson [On Soliton Propagation in Biomembranes and Nerves, Proc. Natl. Acad. Sci. U.S.A. 102, 9790 (2005).].

Non-invasive detection of animal nerve impulses with an atomic magnetometer operating near quantum limited sensitivity

Magnetic fields generated by human and animal organs, such as the heart, brain and nervous system carry information useful for biological and medical purposes. These magnetic fields are most commonly detected using cryogenically-cooled superconducting magnetometers. Here we present the first detection of action potentials from an animal nerve using an optical atomic magnetometer. Using an optimal design we are able to achieve the sensitivity dominated by the quantum shot noise of light and quantum projection noise of atomic spins. Such sensitivity allows us to measure the nerve impulse with a miniature room-temperature sensor which is a critical advantage for biomedical applications. Positioning the sensor at a distance of a few millimeters from the nerve, corresponding to the distance between the skin and nerves in biological studies, we detect the magnetic field generated by an action potential of a frog sciatic nerve. From the magnetic field measurements we determine the activity of the nerve and the temporal shape of the nerve impulse. This work opens new ways towards implementing optical magnetometers as practical devices for medical diagnostics.

Solitary electromechanical pulses in lobster neurons.

Investigations of nerve activity have focused predominantly on electrical phenomena. Nerves, however, are thermodynamic systems, and changes in temperature and in the dimensions of the nerve can also be observed during the action potential. Measurements of heat changes during the action potential suggest that the nerve pulse shares many characteristics with an adiabatic pulse. First experiments in the 1980s suggested small changes in nerve thickness and length during the action potential. Such findings have led to the suggestion that the action potential may be related to electromechanical solitons traveling without dissipation. However, there have been no modern attempts to study mechanical phenomena in nerves. Here, we present ultrasensitive AFM recordings of mechanical changes on the order of 2-12Å in the giant axons of the lobster. We show that the nerve thickness changes in phase with voltage changes. When stimulated at opposite ends of the same axon, colliding action potentials pass through one another and do not annihilate. These observations are consistent with a mechanical interpretation of the nervous impulse.

Whole-GUV patch-clamping

Significance Although membrane composition and tension modulate the activity of ion channels and transporters, this protein–membrane coupling has been challenging to study due to the difficulty of controlling membrane properties in cells and technical limitations of existing in vitro systems. This work demonstrates that the whole-cell patch-clamp technique is stabilized by a dynamic passivation mechanism that can be used to control and measure the current and voltage of intact giant unilamellar vesicles (GUVs), a cell-sized model biomimetic system in which the membrane composition, tension, and shape can be readily controlled. The resulting “whole-GUV” configuration will permit electrophysiological studies of ion channels and transporters in a membrane with a defined composition and physiologically relevant range of tensions. Studying how the membrane modulates ion channel and transporter activity is challenging because cells actively regulate membrane properties, whereas existing in vitro systems have limitations, such as residual solvent and unphysiologically high membrane tension. Cell-sized giant unilamellar vesicles (GUVs) would be ideal for in vitro electrophysiology, but efforts to measure the membrane current of intact GUVs have been unsuccessful. In this work, two challenges for obtaining the “whole-GUV” patch-clamp configuration were identified and resolved. First, unless the patch pipette and GUV pressures are precisely matched in the GUV-attached configuration, breaking the patch membrane also ruptures the GUV. Second, GUVs shrink irreversibly because the membrane/glass adhesion creating the high-resistance seal (>1 GΩ) continuously pulls membrane into the pipette. In contrast, for cell-derived giant plasma membrane vesicles (GPMVs), breaking the patch membrane allows the GPMV contents to passivate the pipette surface, thereby dynamically blocking membrane spreading in the whole-GMPV mode. To mimic this dynamic passivation mechanism, beta-casein was encapsulated into GUVs, yielding a stable, high-resistance, whole-GUV configuration for a range of membrane compositions. Specific membrane capacitance measurements confirmed that the membranes were truly solvent-free and that membrane tension could be controlled over a physiological range. Finally, the potential for ion transport studies was tested using the model ion channel, gramicidin, and voltage-clamp fluorometry measurements were performed with a voltage-dependent fluorophore/quencher pair. Whole-GUV patch-clamping allows ion transport and other voltage-dependent processes to be studied while controlling membrane composition, tension, and shape.

Fluctuations of systems in finite heat reservoirs with applications to phase transitions in lipid membranes

In an adiabatically shielded system, the total enthalpy is conserved. Enthalpy fluctuations of an arbitrarily chosen subsystem must be buffered by the remainder of the total system which serves as a heat reservoir. The magnitude of these fluctuations depends on the size of the reservoir. This leads to various interesting consequences for the physical behavior of the subsystem. As an example, we treat a lipid membrane with a phase transition that is embedded in an aqueous reservoir. We find that large fluctuations are attenuated when the reservoir has finite size. This has consequences for the compressibility of the membrane since volume and area fluctuations are also attenuated. We compare the equilibrium fluctuations of subsystems in finite reservoirs with those in periodically driven systems. In such systems, the subsystem has only finite time available to exchange heat with the surrounding medium. A larger frequency therefore reduces the volume of the accessible heat reservoir. Consequently, the fluctuations of the subsystem display a frequency dependence. While this work is of particular interest for a subsystem displaying a transition such as a lipid membrane, some of the results are of a generic nature and may contribute to a better understanding of relaxation processes in general.

Solitary Electromechanical Pulses in Lobster Neurons

Investigations of nerve activity have been focused predominantly on electrical phenomena. It is to be expected that the state of the nerve cell depend not only on electrochemical potentials and the conjugated flux of ions but also on all other thermodynamic forces including variations in lateral pressure, resulting in changes of membrane area and thickness (1-2) and temperature, resulting in heat flux (3).While both mechanical and thermal signals are very small, they are found to be in phase with voltage changes. Such findings have led to the suggestion that the action potential may be related to electromechanical solitons traveling without dissipation (4). A condition for the existence of such a soliton is the existence of an order transition in the membrane from solid to liquid lightly below physiological temperature.Here, we present ultrasensitive AFM recordings of mechanical changes on the order of 0.2 - 1.2 nm in the giant axons of the lobster. Also, when stimulated at opposite ends of the same axon of lobster, colliding action potentials pass through one another and do not annihilate, as was shown in (5). These results are submitted in Physical Review Letters X (2015) and are not consistent with expectations from the established Hodkin-Huxley model. Findings are consistent with a mechanical interpretation of the nervous impulse.1. Iwasa, K., and I. Tasaki. 1980. Biochem. Biophys. Research Comm. 95:1328-31.2. Tasaki, I., K. Kusano, and M. Byrne. 1989. Biophys. J. 55:1033-40.3. Ritchie, J. M., and R. D. Keynes. 1985. Quart. Rev. Biophys. 18:451-76.4. Heimburg, T., and A. D. Jackson. 2005. Proc. Natl. Acad. Sci. USA 102:9790-95.5. Gonzalez-Perez, A., R. Budvytyte, L. D. Mosgaard, S. Nissen, and T. Heimburg. 2014. Phys. Rev. X 4:031047.

The Effect of the Nonlinearity of the Response of Lipid Membranes to Voltage Perturbations on the Interpretation of Their Electrical Properties. A New Theoretical Description

Our understanding of the electrical properties of cell membranes is derived from experiments where the membrane is exposed to a perturbation (in the form of a time-dependent voltage or current change) and information is extracted from the measured output. The interpretation of such electrical recordings consists in finding an electronic equivalent that would show the same or similar response as the biological system. In general, however, there is no unique circuit configuration, which can explain a single electrical recording and the choice of an electric model for a biological system is based on complementary information (most commonly structural information) of the system investigated. Most of the electrophysiological data on cell membranes address the functional role of protein channels while assuming that the lipid matrix is an insulator with constant capacitance. However, close to their melting transition the lipid bilayers are no inert insulators. Their conductivity and their capacitance are nonlinear functions of both voltage, area and volume density. This has to be considered when interpreting electrical data. Here we show how electric data commonly interpreted as gating currents of proteins and inductance can be explained by the nonlinear dynamics of the lipid matrix itself.

Electromechanical properties of biomembranes and nerves

Lipid membranes are insulators and capacitors, which can be charged by an external electric field. This phenomenon plays an important role in the field of electrophysiology, for instance when describing nerve pulse conduction. Membranes are also made of polar molecules meaning that they contain molecules with permanent electrical dipole moments. Therefore, the properties of membranes are subject to changes in trans-membrane voltage. Vice versa, mechanical forces on membranes lead to changes in the membrane potential. Associated effects are flexoelectricity, piezoelectricity, and electrostriction. Lipid membranes can melt from an ordered to a disordered state. Due to the change of membrane dimensions associated with lipid membrane melting, electrical properties are linked to the melting transition. Melting of the membrane can induce changes in trans-membrane potential, and application of voltage can lead to a shift of the melting transition. Further, close to transitions membranes are very susceptible to piezoelectric phenomena. We discuss these phenomena in relation with the occurrence of lipid ion channels. Close to melting transitions, lipid membranes display step-wise ion conduction events, which are indistinguishable from protein ion channels. These channels display a voltage-dependent open probability. One finds asymmetric current-voltage relations of the pure membrane very similar to those found for various protein channels. This asymmetry falsely has been considered a criterion to distinguish lipid channels from protein channels. However, we show that the asymmetry can arise from the electromechanical properties of the lipid membrane itself. Finally, we discuss electromechanical behavior in connection with the electromechanical theory of nerve pulse transduction. It has been found experimentally that nerve pulses are related to changes in nerve thickness. Thus, during the nerve pulse a solitary mechanical pulse travels along the nerve. Due to electromechanical coupling it is unavoidable that this pulse generates a trans-membrane voltage. In the past, we have proposed that this electromechanical pulse is the origin of the action potential in nerves.

Sound Propagation in Lipid Membranes

It has recently been shown that in-plane sound waves can propagate over long distances in lipid monolayers [1]. Earlier it has been proposed that the propagation of nerve signals can be described by sound phenomena called solitons [2]. The implications of sound propagation in lipid membranes for signaling in biology are far reaching and insight into this is essential for further investigation. Particularly interesting are propagation properties in the vicinity of the biologically relevant lipid melting transition, where mechanical and thermodynamical properties of the system change drastically. We have theoretically addressed the properties of sound propagation in lipid membranes throughout the lipid melting transition. We explored dispersion and attenuation for low frequency sound propagation, a regime previously unexplored. We find that dispersion and attenuation is closely related to the relaxation and the state of lipid membranes [3]. Interestingly, the vast significant changes of dispersion and attenuation occur on timescales similar to ion channel open times and the temporal length of the nerve pulse.[1] Griesbauer et al., PRL, 108, 198103 (2012).[2] Heimburg & Jackson, PNAS, 102, 9790 (2005).[3] Mosgaard et al., Adv. Planar Lipid Bilayers Liposomes, 16 (2012).