Takeshi Nomura

The bacterial mechanosensitive channel of large conductance MscL is activated exclusively by increased tension in the membrane bilayer. Despite many proposed models for MscL opening, its precise mechano-gating mechanism, particularly how the received force at the tension sensor transmits to the gate remains incomplete. Previous studies have shown that along with amphipathic N-terminus located near the cytoplasmic surface of the membrane, Phe78 residue near the outer surface also acts as a "tension sensor," while Gly22 is a central constituent of the "hydrophobic gate." Present study focused on elucidating the force transmission mechanism from the sensor Phe78 in the outer transmembrane helix (TM2) to the gate in the inner transmembrane helix (TM1) of MscL by applying the patch clamp and molecular dynamics (MD) simulations to the wild type MscL channel and its single mutants at the sensor (F78N), the gate (G22N) and their combination (G22N/F78N) double mutant. F78N MscL resulted in a severe loss-of-function, while G22N MscL caused a gain-of-function channel exhibiting spontaneous openings at the resting membrane tension. We initially speculated that the spontaneous opening in G22N mutant might occur without tension acting on Phe78 residue. To test this hypothesis, we examined the (G22N/F78N) double mutant, which unexpectedly exhibited neither spontaneous activity nor activity by a relatively high membrane tension. To understand the underlying mechanism, we conducted MD simulations and analyzed the force transduction pathway. Results showed that the mutation at the tension sensor (F78N) in TM2 caused decreased interaction of this residue not only with lipids, but also with a group of amino acids (Ile32-Leu36-Ile40) in the neighboring TM1 helix, which resulted in an inefficient force transmission to the gate-constituting amino acids on TM1. This change also induced a slight tilting of TM1 towards the membrane plane and decreased the size of the channel pore at the gate, which seems to be the major mechanism for the inhibition of spontaneous opening of the double mutant channel. More importantly, the newly identified interaction between the TM2 (Phe78) and adjacent TM1 (Ile32-Leu36-Ile40) helices seems to be an essential force transmitting mechanism for the stretch-dependent activation of MscL given that substitution of any one of these four amino acids with Asn resulted in severe loss-of-function MscL as reported in our previous work.

Ion channel data recorded using the patch clamp technique are low-pass filtered to remove high-frequency noise. Almanjahie et al. (Eur Biophys J 44:545-556, 2015) based statistical analysis of such data on a hidden Markov model (HMM) with a moving average adjustment for the filter but without correlated noise, and used the EM algorithm for parameter estimation. In this paper, we extend their model to include correlated noise, using signal processing methods and deconvolution to pre-whiten the noise. The resulting data can be modelled as a standard HMM and parameter estimates are again obtained using the EM algorithm. We evaluate this approach using simulated data and also apply it to real data obtained from the mechanosensitive channel of large conductance (MscL) in Escherichia coli. Estimates of mean conductances are comparable to literature values. The key advantages of this method are that it is much simpler and computationally considerably more efficient than currently used HMM methods that include filtering and correlated noise.

Tension in cell membranes is closely related to various cellular events, including cell movement and morphogenesis. Therefore, modulation of membrane tension can be a new approach for manipulating cellular events. Here, we show that an amphipathic peptide derived from the influenza M2 protein (M2[45-62]) yields lamellipodia at multiple sites in the cell. Effect of M2[45-62] on cell membrane tension was evaluated by optical tweezer. The membrane tension sensor protein FBP17 was involved in M2[45-62]-driven lamellipodium formation. Lysine-to-arginine substitution in M2[45-62] further enhanced its activity of lamellipodium formation. M2[45-62] had an ability to reduce cell motility, evaluated by scratch wound migration and transwell migration assays. An increase in neurite outgrowth was also observed after treatment with M2[45-62]. The above results suggest the potential of M2[45-62] to modulate cell movement and morphology by modulating cell membrane tension.

Corynebacterium glutamicum is used in microbial biotechnology for the production of amino acids, in particular glutamate. The mechanism of glutamate excretion, however, is not yet fully understood. Recently, evidence was provided that the NCgl1221 gene product from C. glutamicum ATCC 13869, a MscS-type mechanosensitive efflux channel, is responsible for glutamate efflux [1]. The major difference of NCgl1221 and the homologous protein MscCG of C. glutamicum ATCC 13032 from Escherichia coli MscS and most other MscS-type proteins is the presence of an additional, 247 amino acid long C-terminal domain. By topology analysis, we show that this domain in MscCG carries a transmembrane segment. We have generated selected C-terminal truncations of MscCG, gain-of-function and loss-of-function constructs of both E. coli MscS and C. glutamicum MscCG, as well as fusion constructs of the two proteins. These mutant proteins were investigated for mechanosensitive efflux, MS channel activity, glutamate excretion and their impact on membrane potential. We provide evidence that the channel domain of MscCG mediates glutamate efflux in response to penicillin treatment, and that the E. coli MscS channel is to some extent able to function in a similar manner. We further show that the C-terminal domain of MscCG has a significant impact for function and/or regulation of MscCG. Significantly, a positive effect on glutamate efflux of the C-terminal extension of MscCG from C. glutamicum was also observed when fused to the E. coli MscS channel.

Ion channel studies have been focused on ion channels from animal and human cells over many years. Based on the knowledge acquired, predominantly over the last 20 years, a large diversity of ion channels exists in cellular membranes of prokaryotes as well. Paradoxically, most of what is known about the structure of eukaryotic ion channels is based on the structure of bacterial channels. This is largely due to the suitability of bacterial cells for functional and structural studies of biological macromolecules in a laboratory environment. Development of the "giant spheroplast" preparation from E. coli cells was instrumental for functional studies of ion channels in the bacterial cell membrane. Here we describe detailed protocols used for the preparation of giant spheroplasts as well as protocols used for the patch-clamp recording of native or heterologously expressed ion channels in E. coli spheroplast membrane.

Every cell in the phylogeny from bacteria to human being displays various reactions in response to mechanical stimuli. We named this cellular function, 'Saibou-Rikikaku', however, little is known about its molecular mechanisms and physiological meanings. One of the most important issues of 'Saibou-Rikikaku' is to identify the molecular entity of cell mechanosensors, which are the major player in 'Saibou-Rikikaku', and to reveal their operation mechanisms on the basis of 3D molecular structures. The ion channel-type mechanosensors called MS channel is the only class of mechanosensors that have been identified to date. Among them the best studied MS channels are those from bacteria whose crystal structures have been resolved. In this article, we review the molecular mechanisms how bacterial MS channels are activated and briefly touch on the evolution of the MS channel through the comparison of activation mechanisms in MS channels between bacteria and higher organisms.

Effects of 9-week hindlimb suspension and 8-week recovery on locomotor performance and electromyogram (EMG) activities of soleus (Sol), plantaris (Pl), lateral gastrocnemius (LG), and tibialis anterior (TA), were studied in adult rats. Hyperextension of knee and ankle joints, noted after nine weeks of suspension, did not recover during 8-week ambulation. Growth of Sol was fully inhibited by suspension and did not recover completely within 8 weeks of ambulation. The EMG levels in Sol, Pl, and LG 1 day after suspension were 52-76% less than the pre-suspension level (resting on the floor). These activities were recovered to or near the pre-suspension level around 1 week, but decreased again to 10-29% of controls from 7 to 9 weeks. The integrated EMG of TA was elevated during the first week of suspension but then gradually declined to control levels within four weeks. At the end of suspension, the Sol and Pl, not the LG, EMGs remained reduced and the TA EMG remained hyperactive. Co-activation of dorsi- and plantar- flexors occurred often during quadripedal walking following suspension. Such a phenomenon was not observed in the control rats. These phenomena were recovered within 1 week. It is suggested that the unloading-related alterations of neuromuscular activities and/or locomotion, but not the hyperextension of knee and ankle joints, in rats are reversible.

Effects of hindlimb unloading and reloading on the patterns of landing and posture adjustment in response to head-down drop from a height of approximately 30 cm were investigated in rats. Seven weeks old male Wistar rats were hindlimb-unloaded by tail suspension for 9 consecutive weeks. Motor tests were performed immediately after the termination of suspension and recovery patterns were checked during 8 weeks of ambulation recovery. Although all of the control rats were able to land smoothly by using the four limbs as the shock absorber, the unloaded rats landed by hitting their abdomen. The hindlimb-unloaded, but not control, rats dorsi-flexed their trunk during fall. The mean angle of abdominal side was approximately 145 degrees in control and approximately 215 degrees in unloaded rats. Even though such phenomena were maintained for approximately 12 hours, the response of the trunk angle recovered significantly 2 days later. However, it was not normalized completely even after 8 weeks. Hyper-extension of ankle joints and eversion of hindlimbs at landing were also noted in the unloaded rats. These phenomena were not recovered at all. It was generally suggested that severe detrimental effects on the landing performance of rats are induced following 9-weeks of suspension. And some of the responses are irreversible.