Are axoplasmic microtubules necessary for membrane excitation?

Vascular mechanotransduction

This review aims to survey the current state of mechanotransduction in vascular smooth muscle cells (VSMCs) and endothelial cells (ECs), including their sensing of mechanical stimuli and transduction of mechanical signals that result in the acute functional modulation and longer-term transcriptomic and epigenetic regulation of blood vessels. The mechanosensors discussed include ion channels, plasma membrane-associated structures and receptors, and junction proteins. The mechanosignaling pathways presented include the cytoskeleton, integrins, extracellular matrix, and intracellular signaling molecules. These are followed by discussions on mechanical regulation of transcriptome and epigenetics, relevance of mechanotransduction to health and disease, and interactions between VSMCs and ECs. Throughout this review, we offer suggestions for specific topics that require further understanding. In the closing section on conclusions and perspectives, we summarize what is known and point out the need to treat the vasculature as a system, including not only VSMCs and ECs but also the extracellular matrix and other types of cells such as resident macrophages and pericytes, so that we can fully understand the physiology and pathophysiology of the blood vessel as a whole, thus enhancing the comprehension, diagnosis, treatment, and prevention of vascular diseases.

The living state: How cellular excitability is controlled by the thermodynamic state of the membrane

The thermodynamic (TD) properties of biological membranes play a central role for living systems. It has been suggested, for instance, that nonlinear pulses such as action potentials (APs) can only exist if the membrane state is in vicinity of a TD transition. Herein, two membrane properties in living systems - excitability and velocity - are analyzed for a broad spectrum of conditions (temperature (T), 3D-pressure (p) and pH-dependence). Based on experimental data from Characean cells and a review of literature we predict parameter ranges in which a transition of the membrane is located (15-35°C below growth temperature; 1-3pH units below pH7; at ∼800atm) and propose the corresponding phase diagrams. The latter explain: (i) changes of AP velocity with T,p and pH.(ii) The existence and origin of two qualitatively different forms of loss of nonlinear excitability ("nerve block", anesthesia). (iii) The type and quantity of parameter changes that trigger APs. Finally, a quantitative comparison between the TD behavior of 2D-lipid model membranes with living systems is attempted. The typical shifts in transition temperature with pH and p of model membranes agree with values obtained from cell physiological measurements. Taken together, these results suggest that it is not specific molecules that control the excitability of living systems but rather the TD properties of the membrane interface. The approach as proposed herein can be extended to other quantities (membrane potential, calcium concentration, etc.) and makes falsifiable predictions, for example, that a transition exists within the specified parameter ranges in excitable cells.

Ionic wave propagation and collision in an excitable circuit model of microtubules

In this paper, we report the propensity to excitability of the internal structure of cellular microtubules, modelled as a relatively large one-dimensional spatial array of electrical units with nonlinear resistive features. We propose a model mimicking the dynamics of a large set of such intracellular dynamical entities as an excitable medium. We show that the behavior of such lattices can be described by a complex Ginzburg-Landau equation, which admits several wave solutions, including the plane waves paradigm. A stability analysis of the plane waves solutions of our dynamical system is conducted both analytically and numerically. It is observed that perturbed plane waves will always evolve toward promoting the generation of localized periodic waves trains. These modes include both stationary and travelling spatial excitations. They encompass, on one hand, localized structures such as solitary waves embracing bright solitons, dark solitons, and bisolitonic impulses with head-on collisions phenomena, and on the other hand, the appearance of both spatially homogeneous and spatially inhomogeneous stationary patterns. This ability exhibited by our array of proteinic elements to display several states of excitability exposes their stunning biological and physical complexity and is of high relevance in the description of the developmental and informative processes occurring on the subcellular scale.

Lipid stress at play: mechanosensitivity of voltage-gated channels

Membrane stretch modulates the activity of voltage-gated channels (VGCs). These channels are nearly ubiquitous among eukaryotes and they are present, too, in prokaryotes, so the potential ramifications of VGC mechanosensitivity are diverse. In situ traumatic stretch can irreversibly alter VGC activity with lethal results but that is pathology. This chapter discusses the reversible responses of VGCs to stretch, with the general relation of stretch stimuli to other forms of lipid stress, and briefly, with some irreversible stretch effects (=stretch trauma). A working assumption throughout is that mechanosensitive (MS) VGC motions-that is, motions that respond reversibly to bilayer stretch-are susceptible to other forms of lipid stress, such as the stresses produced when amphiphilic molecules (anesthetics, lipids, alcohols, and lipophilic drugs) are inserted into the bilayer. Insofar as these molecules change the bilayer's lateral pressure profile, they can be termed bilayer mechanical reagents (BMRs). The chapter also discusses the MS VGC behavior against the backdrop of eukaryotic channels more widely accepted as "MS channels"--namely, the transient receptor potential (TRP)-based MS cation channels.

Studying the mechanosensitivity of voltage-gated channels using oocyte patches

The mechanosensitivity of voltage-gated (VG) channels is of biophysical, physiological. and pathophysiological interest. Xenopus oocytes offer a critical advantage for investigating the electrophysiology of recombinant VG channels subjected to membrane stretch, namely, the ability to monitor macroscopic current from membrane patches. High-density channel expression in oocytes makes for macroscopic current in conventional-size, mechanically sturdy patches. With the patch configuration, precisely the same membrane that is voltage-clamped is the membrane subjected to on-off stretch stimuli. With patches, meaningful stretch dose responses are possible. Experimental design should facilitate within-patch comparisons wherever possible. The mechanoresponses of some VG channels depend critically on patch history. Methods for minimizing and coping with interference from endogenous voltage-dependent and stretch-activated endogenous channels are described.

Stretch-activation and stretch-inactivation of Shaker-IR, a voltage-gated K+ channel

Mechanosensitive (MS) ion channels are ubiquitous in eukaryotic cell types but baffling because of their contentious physiologies and diverse molecular identities. In some cellular contexts mechanically responsive ion channels are undoubtedly mechanosensory transducers, but it does not follow that all MS channels are mechanotransducers. Here we demonstrate, for an archetypical voltage-gated channel (Shaker-IR; inactivation-removed), robust MS channel behavior. In oocyte patches subjected to stretch, Shaker-IR exhibits both stretch-activation (SA) and stretch-inactivation (SI). SA is seen when prestretch P(open) (set by voltage) is low, and SI is seen when it is high. The stretch effects occur in cell-attached and excised patches at both macroscopic and single-channel levels. Were one ignorant of this particular MS channel's identity, one might propose it had been designed as a sophisticated reporter of bilayer tension. Knowing Shaker-IR's provenance and biology, however, such a suggestion would be absurd. We argue that the MS responses of Shaker-IR reflect not overlooked "mechano-gating" specializations of Shaker, but a common property of multiconformation membrane proteins: inherent susceptibility to bilayer tension. The molecular diversity of MS channels indicates that susceptibility to bilayer tension is hard to design out of dynamic membrane proteins. Presumably the cost of being insusceptible to bilayer tension often outweighs the benefits, especially where the in situ milieu of channels can provide mechanoprotection.

Membrane stretch affects gating modes of a skeletal muscle sodium channel

The alpha subunit of the human skeletal muscle Na(+) channel recorded from cell-attached patches yielded, as expected for Xenopus oocytes, two current components that were stable for tens of minutes during 0.2 Hz stimulation. Within seconds of applying sustained stretch, however, the slower component began decreasing and, depending on stretch intensity, disappeared in 1-3 min. Simultaneously, the faster current increased. The resulting fast current kinetics and voltage sensitivity were indistinguishable from the fast components 1) left after 10 Hz depolarizations, and 2) that dominated when alpha subunit was co-expressed with human beta1 subunit. Although high frequency depolarization-induced loss of slow current was reversible, the stretch-induced slow-to-fast conversion was irreversible. The conclusion that stretch converted a single population of alpha subunits from an abnormal slow to a bona fide fast gating mode was confirmed by using gigaohm seals formed without suction, in which fast gating was originally absent. For brain Na(+) channels, co-expressing G proteins with the channel alpha subunit yields slow gating. Because both stretch and beta1 subunits induced the fast gating mode, perhaps they do so by minimizing alpha subunit interactions with G proteins or with other regulatory molecules available in oocyte membrane. Because of the possible involvement of oocyte molecules, it remains to be determined whether the Na(+) channel alpha subunit was directly or secondarily susceptible to bilayer tension.

Effects of neonatal attenuation of axoplasmic flow or transection of the rat's infraorbital nerve on the morphology of individual trigeminal primary afferent terminals in the brainstem

Attenuation of axoplasmic transport in the infraorbital nerve (ION), or transection of this trigeminal (V) branch at birth, results in degradation of the central cellular aggregates related to the mystacial vibrissae. However, blockade of axoplasmic transport does not result in the nearly 90% loss of ION ganglion cells that follows neonatal transection of this nerve. The present study was undertaken to further characterize the response of individual ION axons to attenuation of axoplasmic transport and to compare these effects to the changes observed following nerve transection. Neurobiotin injections were made into the V ganglion on postnatal day (P-) 6 in normal rats and animals that had vinblastine applied to the ION or received transection of the ION on P-0. Individual labeled fibers in the portions of V nucleus principalis (PrV) and subnucleus interpolaris (SpI) innervated by the ION were drawn from single sections with the aid of a computer. Morphological analysis of fibers drawn in SpI indicated no significant differences between axons from normal and vinblastine-treated animals. The fibers drawn from rats that sustained ION transection had significantly more branch points (P < 0.05) than those from either normal or vinblastine-treated animals. In PrV, fibers drawn from vinblastine-treated rats had a slightly, but significantly, larger total process length and cross-sectional area than those from the normal animals (P < 0.05). There were no other significant differences among the three groups of axons. These results support the conclusion that application of vinblastine to the developing ION does not dramatically alter the morphologic patterning of the central arbors of its axons.

Injury-induced vesiculation and membrane redistribution in squid giant axon

Injury of isolated squid giant axons in sea water by cutting or stretching initiates the following unreported processes: (i) vesiculation in the subaxolemmal region extending along the axon several mm from the site of injury, followed by (ii) vesicular fusions that result in the formation of large vesicles (20-50 micron diameter), 'axosomes', and finally (iii) axosomal migration to and accumulation at the injury site. Some axosomes emerge from a cut end, attaining sizes up to 250 microns in diameter. Axosomes did not form after axonal injury unless divalent cations (Ca2+ or Mg2+) were present (10mM) in the external solution. The requirement for Ca2+ and the action of other ions are similar to that for cut-end cytoskeletal constriction in transected squid axons (Gallant, P.E. (1988) J. Neurosci. 8, 1479-1484) and for electrical sealing in transected axons of the cockroach (Yawo, H. and Kuno, M. (1985) J. Neurosci. 5, 1626-1632). Axosomes probably consist of membrane from different sources (e.g., axolemma, organelles and Schwann cells); however, localization of axosomal formation to the inner region of the axolemma and the formation dependence on divalent cations suggest principal involvement of cisternae of endoplasmic reticulum. Patch clamp of excised patches from axosomes liberated spontaneously from cut ends of transected axons showed a 12-pS K+ channel and gave indications of other channel types. Injury-induced vesiculation and membrane redistribution seem to be fundamental processes in the short-term (minutes to hours) that precede axonal degeneration or repair and regeneration. Axosomal formation provides a membrane preparation for the study of ion channels and other membrane processes from inaccessible organelles.

Selective axonal transport in a single cholinergic axon of Aplysia--role of colchicine-resistant microtubules

Substance-specific selective axonal transport was examined in a single axon by injecting [3H]leucine and [14C]acetylcholine simultaneously into the cell body of a giant cholinergic neuron (R2) in the abdominal ganglion of Aplysia kurodai. The ganglion and attached nerves were cultured for several hours after the injection and the migration of radioactive substances along the axons of the injected neuron was examined. The substances examined were 3H labeled membrane proteins and soluble proteins synthesized in the cell body, 14C labeled bound acetylcholine formed in the cell, injected [3H]leucine and soluble [14C]acetylcholine. Membrane proteins and bound acetylcholine (plus a part of soluble acetylcholine) moved along the axon somatofugally at maximum velocities of 2.4 and 1.7 mm/h, respectively, at 25 degrees C. Soluble proteins, free leucine and most of the soluble acetylcholine did not move by fast axonal transport but diffused inside the axon of the neuron R2 at rates predicted from their expected diffusion constants in the axoplasm [Koike H. and Nagata Y. (1979) J. Physiol. 295, 397-417]. The diffusion kinetics of these substances were analysed and used for determination of true axon length, and to separate axonal transport components from diffusing components. An antimitotic drug, colchicine, selectively suppressed the axonal transport of membrane proteins but not of acetylcholine at 1-5 mM concentration, though it finally blocked the axonal transport of acetylcholine at 20 mM. When 1-5 mM colchicine was separately perfused only to the distal axon of the neuron R2, the migration of membrane proteins was stopped just proximal to the colchicine perfusion zone but acetylcholine migration was not disturbed by the drug. The moving component of acetylcholine was recovered by sucrose density centrifugation from a compartment previously reported as that of vesicular acetylcholine. As a possible mechanism of this selective axonal transport, it is proposed that there are two groups of microtubules: a colchicine-sensitive group of microtubules which may transport membrane proteins, and a colchicine-resistant group which may preferentially transport the transmitter substance acetylcholine at a slower rate.

The disturbance of the fast axonal transport of protein by passive stretching of an axon in Aplysia

1. Radioactive amino acid, either [3H]leucine or [3H]proline, was injected into neurone R2 in the abdominal ganglion of Aplysia kurodai to investigate the intra-axonal transport of protein in a single axon. 2. Some of the injected amino acid which was not utilized for protein synthesis diffused intra-axonally with a diffusion constant of 6.0 X 10(-6) cm2/s (25 degrees C), which is the value expected from the known diffusion constants of acetylcholine and gamma-aminobutyric acid in the axoplasm (Koike & Nagata, 1979). The true length of the zigzagging axon in the nerve bundle was measured by the diffusion distances. 3. The radioactive proteins newly synthesized in the cell body from either of the injected amino acid were transported axonally in a single axon of R2. 4. Elongation of the axon resulted in a suppression of the fast axonal transport of the proteins and the amount of protein transported. This contrasts with the observation that action potential propagation along the elongated axon never failed. 5. A possible site for the suppression of the axonal transport would be axonal microtubules whose structure is not likely to be resistant to the distortion caused by the elongation.

Cell membrane expansion and blockade of action potentials produced by 2-decenoic acid in cultured dorsal root ganglion neurons

2-Decenoic acid, a fatty acid having 10 carbon atoms, blocks the action potentials of cultured dorsal root ganglion (DRG) neurons and this effect of 2-decenoic acid is reversible. From the analysis of the video pictures from Nomarski optics, relative values of the diameter and the thickness of the neurons increased to 1.06 and 1.14, respectively, when 2.1 mM 2-decenoic acid was applied to the neurons. The relative value of cell surface area, which was calculated from the equation for a spheroid, increased to about 1.20. On the other hand, relative fluorescence intensity of the fluorescent probe F18 (5-(octadecylthiocarbamoylamino)fluorescein) labeled neurons decreased to 0.81, when 2.1 mM 2-decenoic acid was applied to the neurons. This indicates that the relative cell surface area increased to 1.23, a value similar to that calculated from the results of the measurement of cell size. The time course of blocking action potentials after treatment of the fatty acid was similar to that of the cell membrane expansion. These results show that the fatty acid perturbs the cell membrane and expands the cell surface area and this expansion might reduce the opening ability of the Na+-channels in the membrane.

Potential-dependent variations of the intracellular pressure in the intracellularly perfused squid giant axon

Intracellular pressure responses were recorded from squid giant axons after the axoplasm was removed by the intracellular perfusion technique. A glass tube was inserted into the axon and the movement of the air-water interface formed on the end of the tube was observed with a Y-shaped fibrescope. The intracellular pressure increased and decreased rapidly when an action potential was induced by electrical stimulation. The amplitude of the response was about 10 mPa (or 1 X 10(-3) mmH2O), which was very large in comparison with that observed in unperfused axons. It was sensitive to extracellular Ca2+. The pressure response appeared in an all-or-none manner and could be suppressed by tetrodotoxin. This excluded physicochemical processes on the stimulating electrode or current-supplying electrode as sources of the response. Various other sources of artifacts were also excluded. An extensive removal of the axoplasm by intracellular perfusion with a protease-containing solution and a KCl solution did not eliminate the pressure response. The intracellular pressure was membrane potential dependent, increasing upon depolarization and decreasing upon hyperpolarization of the membrane. Under voltage clamp, the relationship between the membrane potential and the pressure response was parabolic with a maximum at +109 mV (in reference to the resting level). The response did not depend on the membrane current. A much slower response due to electro-osmotic water flow was also detected. The pressure response induced by hyperpolarization of the membrane was suppressed by extracellular application of a lidocaine-containing solution, but not by a tetrodotoxin-containing solution. These results suggest that the pressure responses arise either from a change in electrostriction across the axolemma or from a change in charge-dependent tension along the axolemma.