Mechanosensitive ion channels MSL8, MSL9, and MSL10 have environmentally sensitive intrinsically disordered regions with distinct biophysical characteristics in vitro

Abstract

Intrinsically disordered protein regions (IDRs) are highly dynamic sequences that rapidly sample a collection of conformations over time. In the past several decades, IDRs have emerged as a major component of many proteomes, comprising ~30% of all eukaryotic protein sequences. Proteins with IDRs function in a wide range of biological pathways and are notably enriched in signaling cascades that respond to environmental stresses. Here, we identify and characterize intrinsic disorder in the soluble cytoplasmic N-terminal domains of MSL8, MSL9, and MSL10, three members of the MscS-like (MSL) family of mechanosensitive ion channels. In plants, MSL channels are proposed to mediate cell and organelle osmotic homeostasis. Bioinformatic tools unanimously predicted that the cytosolic N-termini of MSL channels are intrinsically disordered. We examined the N-terminus of MSL10 (MSL10^N) as an exemplar of these IDRs and circular dichroism spectroscopy confirms its disorder. MSL10^N adopted a predominately helical structure when exposed to the helix-inducing compound trifluoroethanol (TFE). Furthermore, in the presence of molecular crowding agents, MSL10^N underwent structural changes and exhibited alterations to its homotypic interaction favorability. Lastly, interrogations of collective behavior via in vitro imaging of condensates indicated that MSL8^N, MSL9^N, and MSL10^N have sharply differing propensities for self-assembly into condensates, both inherently and in response to salt, temperature, and molecular crowding. Taken together, these data establish the N-termini of MSL channels as intrinsically disordered regions with distinct biophysical properties and the potential to respond uniquely to changes in their physiochemical environment.

Keywords: Arabidopsis thaliana; circular dichroism; intrinsically disordered protein; ion channel; mechanobiology; phase separation; transmembrane protein.

FIGURE 1

The N‐termini of MSL family proteins are predicted to be disordered. (a) Topology and predicted disorder for MSL10. MSL8 and MSL9 present similar topologies. Residues predicted by IUPRED2A to be disordered are shown in black. The conserved MscS domain of MSL10 is dark gray. (b) Amino acid composition of the N‐termini of MSL8, MSL9, and MSL10. Calculations were performed using the PredictProtein webserver. Disorder‐promoting, order‐promoting, and neutral residues are shown in black, light gray, and dark gray, respectively. (c) Compositional profile of the MSL10 N‐terminus compared to the MSL10 C‐terminus (aa 562–734). Comparisons were performed using the composition profiler webtool. Enrichment of a particular amino acid is calculated as (composition N‐terminus–composition C‐terminus)/composition C‐terminus, where the composition is the fractional amount of an amino acid within the N‐ or C‐terminus. Color of the bars as in (b). (d) IUPRED2A disorder profiled for full‐length MSL8, MSL9, MSL10, and the median value of selected orthologs at a given aligned position of MSL10. Residues with a disorder score higher than .5 are predicted to be part of a disordered region, indicated by the dashed line. The black bar indicates N‐terminal regions.

FIGURE 2

The MSL10 N‐terminus is disordered and structurally responds to a variety of environments in vitro. (a) Representative coomassie‐stained 10% sodium dodecyl sulphate‐polyacrylamide gel electrophoresis (SDS‐PAGE) gel of his‐tagged MSL10^N. His‐tagged MSL10^N has an approximate molecular weight of 19 kDa. Circular dichroism spectra of his‐tagged MSL10^N when exposed to increasing temperatures (b), TFE (c), PEG 400 (e), and glycerol (f). Spectra were obtained at 20°C in 20 mM sodium phosphate buffer, pH 7.4 unless otherwise specified. (d) The ratio of the mean residue ellipticity value at 222 and 208 nm for the indicated TFE treatments.

FIGURE 3

Neither phosphomimetic nor phosphodead mutations affect the folding of MSL10 ^N . (a) Individual disorder scores for each amino acid were averaged for wild‐type, phosphomimetic, and phosphodead variants of MSL10^N using the indicated prediction algorithms. Binary disorder–order cutoffs for each prediction program are shown in parentheses. (b) Coomassie‐stained 10% sodium dodecyl sulphate‐polyacrylamide gel electrophoresis (SDS‐PAGE) gel. (c) Circular dichroism spectra of the indicated variants, obtained at 20°C in 20 mM sodium phosphate buffer, pH 7.4.

FIGURE 4

Self‐assembly behavior of MSL9 ^N and MSL10 ^N . (a) Key parameters of MSL8^N, MSL9^N, and MSL10^N protein sequences as calculated by the ProtParam webtool. (b) Fluorescence images of C‐terminally his‐tagged MSL9^N and MSL10^N in different salt concentrations at ~500 μM protein. (c) Phase diagram of MSL9^N as a function of protein and NaCl concentration. Black crosses denote tested conditions at which condensates were absent, whereas black circles are conditions that produced condensates. (d) Fluorescence images of MSL9^N wherein protein was treated with additional salt after droplet formation. Protein concentration before and after salt addition was ~550 μM. (e) Images of his‐tagged MSL9^N and MSL10^N in response to 30% PEG 400 treatment at ~500 μM protein. In b, d, and e, scale bars are 10 mm.

FIGURE 5

MSL8 ^N forms condensates in response to salt and to low temperatures. (a) Fluorescence images of his‐tagged MSL8^N and MSL9^N with increasing levels of salt in 20 mM sodium phosphate. Protein concentration is in the upper left side of each image. (b) Image of tubes containing purified MSL8^N at the indicated temperatures. (c) Fluorescence images of MSL9^N and MSL8^N at room temperature (upper) and 4°C (lower). The samples in these images are in 50 mM NaCl, 20 mM sodium phosphate. The protein concentrations of MSL9^N and MSL8^N preparations were approximately 10 and 2.5 μM, respectively.