Patch-clamp characterization of the MscS-like mechanosensitive channel from Silicibacter pomeroyi

Abstract

Based on sequence similarity, the sp7 gene product, MscSP, of the sulfur-compound-decomposing Gram-negative marine bacterium Silicibacter pomeroyi belongs to the family of MscS-type mechanosensitive channels. To investigate MscSP channel properties, we measured its response to membrane tension using the patch-clamp technique on either a heterologous expression system using giant spheroplasts of MJF465 Escherichia coli strain (devoid of mechanosensitive channels MscL, MscS, and MscK), or on purified MscSP protein reconstituted in azolectin liposomes. These experiments showed typical pressure-dependent gating properties of a stretch-activated channel with a current/voltage plot indicating a rectifying behavior and weak preference for anions similar to the MscS channel of E. coli. However, the MscSP channel exhibited functional differences with respect to conductance and desensitization behavior, with the most striking difference between the two channels being the lack of inactivation in MscSP compared with MscS. This seems to result from the fact that although MscSP has a Gly in an equivalent position to MscS (G113), a position that is critical for inactivation, MscSP has a Glu residue instead of an Asn in a position that was recently shown to allosterically influence MscS inactivation, N117. To our knowledge, this study describes the first electrophysiological characterization of an MscS-like channel from a marine bacterium belonging to sulfur-degrading α-proteobacteria.

Figure 1

MscSP versus MscS sequence alignment. Comparison of the primary amino acid sequence between MscS of E. coli and its homolog from S. pomeroyi, MscSP. The alignment reveals ∼40% sequence identity. Identical residues are marked with a black box and TM domains (6) TM1, TM2, and TM3 are represented by black bars.

Figure 2

Comparison of MscS and MscSP structures. (A) Corresponding hydropathy plots for MscS and MscSP were determined using the Kyte-Doolittle algorithm. Gray bars beneath represent the respective TM helices and cytoplasmic domains. (B) Pairwise alignment of MscS with MscSP showing percentage identity and similarity of the full-length proteins, cytoplasmic domains, and pore-forming helices. (C) Pairwise sequence alignment of the MscS pore-forming helix and the putative pore-forming helix of MscSP further illustrates the high level of similarity between these homologs. (D) Comparison of a homology model of MscSP created using the Swiss-Model program with the structure of MscS (PDB: 2VV5). Important functional residues in MscS are highlighted on the left and their corresponding residues in MscSP are shown on the right.

Figure 3

Activation of MscSP by membrane tension. MscSP protein was successfully expressed in the native E. coli membrane and reconstituted into liposome bilayers. (A and B) In both cases, MscSP exhibited activities of a functional MS channel: (A) giant spheroplasts and (B) azolectin liposomes. The activity of tens of active channels could be recorded in both preparations. Arrowheads point to the first channel opening observed in the current traces shown.

Figure 4

Boltzmann distribution curves for MscSP and MscS. (A) Mechanosensitivity profile plotted as a function of open probability P_o and negative pressure. P_o-values for MscS reconstituted into azolectin liposomes (open symbols), for MscSP reconstituted into azolectin liposomes (solid symbols; Boltzmann curves on the left), and for MscS (gray symbols) and MscSP (black symbols) in spheroplast membrane, respectively (Boltzmann curves on the right), are shown with their best-fit plots as indicated in the figure above each set of Boltzmann curves. Note the similarity in mechanosensitivity between MscS and MscSP in azolectin liposomes, and the small difference between MscSP in azolectin and MscSP in spheroplast membrane. (B) Activation threshold of MscS and MscSP reconstituted into azolectin liposomes and MscSP expressed in spheroplasts, respectively. The asterisk, dagger, and double dagger indicate that the value is significantly different from MscS and MscSP reconstituted into liposomes, respectively (^∗∗∗p < 0.001 vs. MscS liposomes, ^†††p < 0.001 vs. MscSP liposomes, ^‡‡‡p < 0.001 vs. MscS spheroplasts by t-test); n indicates the number of different patches tested. All values are represented as mean ± SE.

Figure 5

MscSP and MscL channels coreconstituted into azolectin liposomes. (A) Channel activities of MscS and MscL upon activation by a pressure ramp in azolectin liposomes. The arrowhead and arrow indicate the first opening of MscSP and MscL, respectively. The pipette potential was +30 mV. (B) Current (upper) trace shows opening of three MscSP (arrow) and four MscL (^∗) channels upon activation by pressure steps. Note that MscSP channels have a lower opening threshold than MscL and do not desensitize like MscS (see Discussion). The lower trace shows negative pressure applied to the patch electrode. The voltage applied to the patch pipette is shown in the upper-left corner, and the pressure-current timescale is shown in the bottom-left corner.

Figure 6

Single-channel activity and current-voltage plots for MscSP and MscS. (A) Substates of MscSP. Three channels are active in the patch. The inset shows a scaled fragment of the open state of one channel (at +50 mV). (B) Amplitude histogram shows several substates near the open state. The inset shows the same histogram, with N shown on the logarithmic scale. Note the single substate (S) and multiple substates between S and the open state (O). (C) Current-voltage plots show the single-channel conductance of MscSP (□) and MscS (○) (see Results).

Figure 7

Ion selectivity of MscSP. (A) Current-voltage plot of the MscSP single channel recorded in high potassium (750 mM KCl) bath solution. The reversal potential is ∼−4 mV. This is an indication of weak preference of MscSP channel for Cl⁻ ions. (B) Sequence alignment of vestibular portal residues of MscS and MscSP are shown for comparison. The residues circled are those contributing to weak anionic selectivity in MscS.