SWI/SNF senses carbon starvation with a pH-sensitive low-complexity sequence
- Department of Molecular and Cell Biology, University of California, Berkeley, United States
- Institute for Systems Genetics, New York University Grossman School of Medicine, United States
- Program in Molecular Medicine, University of Massachusetts Medical School, United States
- Department of Cell and Molecular Biology, University of Rhode Island, United States
- Center for Science and Engineering of Living Systems (CSELS), Washington University in St. Louis, United States
- Department of Biochemistry and Molecular Biophysics, Washington University School of Medicine, United States
Figures
Figure 1 with 9 supplements
Efficient induction of ADH2 upon glucose starvation requires the SNF5 glutamine-rich low-complexity sequence with native histidines.
(A) Sequence of the N-terminal low-complexity domain of SNF5. This domain was deleted in the ΔQsnf5 strain. The glutamine-rich domain is highlighted in orange. The 4/7 histidines that were mutated …
Figure 1—figure supplement 1
The SWI/SNF complex has 10/11 subunits with significant disorder.
Fractional disorder in each of the core 11 SWI/SNF components. Dashed red lines represent 25% and 50% disorder. 5 of the 11 components contain over 25% disorders. Disorder prediction performed using …
Figure 1—figure supplement 2
Identification and analysis of glutamine-rich low-complexity sequences (QLCs).
(A) Example of a QLC region with the criteria that define QLCs annotated: QLCs were defined as subregions of the proteome in which they have an average fraction of glutamine residues of 25% or …
Figure 1—figure supplement 3
Histidines are enriched in glutamine-rich low-complexity sequences.
Amino acid frequencies within glutamine-rich low-complexity sequences (QLCs) in S. cerevisiae (yeast), Dictyostelium discoides, Drosophila melanogaster, and humans. (A) Enrichment of each amino acid …
Figure 1—figure supplement 4
The SNF5 N-terminal glutamine-rich low-complexity domain (with embedded histidines) is broadly conserved across Ascomycota.
(A) Analysis of SNF5 N-terminal region showing conservation (black), disorder (red), glutamine positions (green), histidine positions (blue), histidines that are mutated (orange), and the QLC …
Figure 1—figure supplement 5
The SNF5 N-terminal glutamine-rich low-complexity domain was probably gained in the fungal lineage.
Broad orthologs of SNF5 were determined using the more conserved C-terminal domain. These orthologs were analyzed, and if QLCs were detected, the number of residues within is plotted in the bar …
Figure 1—figure supplement 6
The SNF5 QLC is important for recovery from carbon starvation.
Growth rate was assessed in a plate reader in various conditions. (A) Comparison of growth rate of WT, ΔQsnf5, HtoAsnf5, and snf5Δ strains in synthetic complete media with glucose. (B) Cells were …
Figure 1—figure supplement 7
Mutation of the SNF5 QLC does not lead to protein degradation or loss of SWI/SNF complex integrity.
(A) The entire SWI/SNF complex copurifies with SNF2 in all strains and conditions. The endogenous SNF2 gene was tandem affinity purification (TAP)-tagged at the C-terminus and used to …
-
Figure 1—figure supplement 7—source data 1
The entire SWI/SNF complex copurifies with SNF2 in all strains and conditions.
Figure 1—figure supplement 7A: annotated. The endogenous SNF2 gene was tandem affinity purification (TAP)-tagged at the C-terminus and used to immunoprecipitate the SWI/SNF complex from WT, ΔQsnf5, or HtoAsnf5 strains either exponentially growing in glucose or after 4 hr acute carbon starvation in media titrated to pHe 5 or 7.4 (indicated at bottom). A silver stain of an SDS-PAGE analysis is shown. Figure 1—figure supplement 7A: unannotated. Silver-stained SDS-PAGE gel with no annotation. Figure 1—figure supplement 7B: annotated. Neither SNF5 nor its mutant alleles are degraded upon glucose starvation. Western blots of the TAP-tagged SNF5 alleles in various conditions (indicated at bottom). TAP-tagged ΔQ-snf5 runs at ~110 kDa, 288 amino acids smaller than WT (~160 kDa). An anti-glucokinase antibody was used as a loading control (bottom band at ~50 kDa). The SNF5-TAP bands are indicated by red boxes. Figure 1—figure supplement 7B: unannotated. Western blot with no annotation.
- https://cdn.elifesciences.org/articles/70344/elife-70344-fig1-figsupp7-data1-v2.zip
Figure 1—figure supplement 8
Efficient recruitment of the SWI/SNF complex to the ADH2 promoter depends upon pH, the SNF5 QLC and histidines within.
The endogenous SNF2 gene was tandem affinity purification (TAP)-tagged at the C-terminus and used to immunoprecipitate the SWI/SNF complex from WT, ΔQsnf5, or HtoAsnf5 strains. Prior to …
Figure 1—figure supplement 9
The SNF5 QLC and embedded histidines are required for efficient ADH2 induction upon carbon starvation.
(A) Schematic of the PADH2-mCherry reporter gene: the reporter construct was integrated into the endogenous ADH2 locus, resulting in a tandem repeat of the reporter gene followed and an intact ADH2 …
Figure 2 with 5 supplements
The SNF5 QLC is required for ADH2 expression and recovery of neutral pH.
(A) Representative flow cytometry for WT, ΔQsnf5, or HtoAsnf5 strains: the x-axis shows nucleocytoplasmic pH (pHi), while the y-axis shows fluorescence from the PADH2-mCherry reporter. Panels show …
Figure 2—figure supplement 1
Examples of calibration curves to measure cytosolic pH using pHluorin.
(A–C) Representative calibration curves to determine the ratio of fluorescence intensities at 405 and 488 nm in cells adjusted to a known pH by ATP depletion and permeabilization in buffers. The …
Figure 2—figure supplement 2
Cells that fail to induce PADH2-mCherry had lower fitness relative to the inducing population.
6 hr after acute carbon starvation, we used fluorescence-activated cell sorting (FACS) to separate equal numbers of cells with high (induced) and low (uninduced) mCherry fluorescence. (A, B) …
Figure 2—figure supplement 3
All strains ultimately express some amount of PADH2-mCherry reporter.
Cytometry data showing PADH2-mCherry induction either in glucose (light gray peaks to left) or after 24 hr of carbon starvation (dark lines, and color coded by strain).
Figure 2—figure supplement 4
snf5∆ strains only had a slight delay in expression of the PADH2-mCherry reporter.
Cytometry data showing PADH2-mCherry induction (y-axis) and nucleocytoplasmic pH (pHi), calculated using the ratiometric pHluorin probe (x-axis), in WT (left) and snf5∆ (right) strains. Percentage …
Figure 2—figure supplement 5
Recovery of pHi requires new protein translation.
(A) Cytometry data showing nucleocytoplasmic pH (pHi), calculated using the ratiometric pHluorin probe. (B) Quantification of pHi data (see Materials and methods), orange and gray lines are from …
Figure 3 with 4 supplements
Transient acidification is required for ADH2 induction upon carbon starvation.
(A) Expression of PADH2-mCherry reporter gene in WT, ΔQsnf5, or HtoAsnf5 strains 8 hr after acute carbon starvation in media titrated to various pH (pHe, see legend, right). Bar height indicates the …
Figure 3—figure supplement 1
Deletion of the N-terminal glutamine-rich domain of SNF5 renders cells hypersensitive to starvation at suboptimal extracellular pH.
Cells were grown to log phase and then subjected to acute carbon starvation in media titrated to various pHe values (see legend). After 24 hr starvation, cells were plated to determine the number of …
Figure 3—figure supplement 2
PADH2-mCherry induction requires an acidic extracellular environment and the SNF5 QLC.
Cytometry data showing expression levels of the PADH2-mCherry reporter from WT, ΔQsnf5, or HtoAsnf5 cells either growing in glucose (Glu), or 6 hr after acute carbon starvation in media titrated to …
Figure 3—figure supplement 3
Expression of the endogenous ADH2 mRNA requires an acidic extracellular environment and the SNF5 QLC.
RT-qPCR data showing ADH2 mRNA levels. The ratio of ADH2 levels in carbon-starved cells to cells growing in glucose is shown. ACT1 was used as an internal control to normalize ADH2 values. WT and ΔQs…
Figure 3—figure supplement 4
Transient acidification of cells requires an acidic extracellular environment.
Flow cytometry for WT, ΔQsnf5, or HtoAsnf5 strains: the x-axis shows nucleocytoplasmic pH (pHi), while the y-axis shows fluorescence from the PADH2-mCherry reporter. Panels show cells grown in …
Figure 4
The SNF5 QLC and acidification of the nucleocytoplasm are required for efficient widespread transcriptional reprogramming upon carbon starvation.
(A) Principal component (PC) analysis of three RNA-seq biological replicates for each condition tested. (B) Expression levels of genes that were greater than threefold induced or repressed upon …
Figure 5 with 2 supplements
The SNF5 QLC mediates a pH-sensitive transcription factor interaction in vitro.
(A) Schematic: a Cy3 donor fluorophore was attached to one end of the DNA, and the histone H2A C-termini were labeled with a Cy5 acceptor fluorophore. ATP-dependent mobilization of the nucleosome to …
Figure 5—figure supplement 1
QLCs of SWI/SNF cluster around putative transcription factor interaction sites, as do low-complexity sequences of human BAF complex.
(A) Electron microscopy structure of SWI/SNF (gray) bound to a nucleosome (DNA blue, histones green; PDB ID: 7C4J). The position of SNF5 is highlighted in coral. Rough positions of QLCs are depicted …
Figure 5—figure supplement 2
Basal ATPase activity is not affected by pH, and Förster resonance energy transfer (FRET) changes require ATP hydrolysis.
(A) Representative trace of ATPase activity for WT and ΔQsnf5 mutant SWI/SNF complexes in response to varied environmental pH. WT and mutant complexes do not show significant changes in ATPase …
Figure 6 with 1 supplement
Protonation of histidines leads to conformational expansion of the SNF5 QLC.
(A) Schematic of the SNF5 gene (center) with the N-terminal QLC in orange and the two simulated peptides in dark orange. Sequences of the simulated peptides and identities of histidines mutated in …
Figure 6—figure supplement 1
A second peptide within the N-terminal QLC of SNF5 undergoes conformational expansion upon protonation.
(A) Schematic of the SNF5 gene, with the sequence and location of the simulated peptide indicated. (B) Radius of gyration (Rg, y-axis) of all-atom Monte Carlo simulations of amino acids 195–233 of …
Author response image 1
Tables
Key resources table
| Reagent type (species) or resource | Designation | Source or reference | Identifiers | Additional information |
|---|---|---|---|---|
| Gene (Saccharomyces cerevisiae) | SNF5 | https://www.yeastgenome.org/ | SGD:S000000493 | |
| Gene (S. cerevisiae) | SNF2 | https://www.yeastgenome.org/ | SGD:S000005816 | |
| Gene(pHluorin) | pHluorin | doi:10.1099/mic.0.022038-0 | ||
| Strain, strain background (S. cerevisiae S288c) | BY4741 | doi:https://doi.org/10.1002/(SICI)1097-0061(19980130)14:2<115::AID-YEA204>3.0.CO;2-2 | All strains used in this study are derived form BY4741 | |
| Other | LH3647 | ADH2::PADH2-mCherry-URA3 snf2::SNF2-TAP-His3MX6 | Yeast strain used to purify SWI/SNF complex | |
| Other | LH3649 | ΔQsnf5-HIS3 ADH2::PADH2-mCherry-URA3 snf2::SNF2-TAP-kanMX6 | Yeast strain used to purify SWI/SNF complex containing ∆Qsnf5 | |
| Other | LH3652 | HtoAsnf5-HIS3 ADH2::PADH2-mCherry-URA3 snf2::SNF2-TAP-kanMX6 | Yeast strain used to purify SWI/SNF complex containing HtoAsnf5 | |
| Recombinant DNA reagent | Plasmid (pRS316) | GenBank: U03442 | Used to complement SNF5 gene in snf5∆ strains prior to removal using 5FOA | |
| Recombinant DNA reagent | Plasmid(pRS306) | GenBank: U03438 | SNF5 and snf5 mutant alleles were all cloned into pRS306 and pRS303 | |
| Recombinant DNA reagent | Plasmid(pRS303) | GenBank: U03435 | SNF5 and snf5 mutant alleles were all cloned into pRS306 and pRS303 | |
| Antibody | Rabbit polyclonal IgG | Sigma | Cat# 12-370 | |
| Antibody | Fluorescently labeled goat anti-rabbit polyclonal | LI-COR Biosciences | Cat# 926-68071 | Western blot (1:15,000 dilution) |
| Antibody | Rabbit polyclonalanti-glucokinase | US Biological | Cat# H2035-01 | Western blot (1:3000 dilution) |
| Antibody | Fluorescently labeled goat anti-rabbit polyclonal | LI-COR Biosciences | Cat# 926-32211 | Western blot (1:15,000 dilution) |
Additional files
-
Supplementary file 1
Sequences of glutamine-rich low-complexity sequences (QLCs) in the Saccharomyces cerevisiae genome.
All S. cerevisiae QLCs identified using the parameters optimized in Figure 1—figure supplement 2 are included in this summary table.
- https://cdn.elifesciences.org/articles/70344/elife-70344-supp1-v2.xlsx
-
Supplementary file 2
Comparison of sequence properties of SNF5 N-terminal intrinsically disordered regions (IDRs).
Comparison of the IDRs of SNF5 orthologues from Ascomycete fungi, with the number of glutamines and histidines indicated.
- https://cdn.elifesciences.org/articles/70344/elife-70344-supp2-v2.xlsx
-
Supplementary file 3
Transcription factors enriched in each gene group from RNA-seq analysis.
The YEASTRACT server used to find transcription factors enriched within the promoters of each of four gene sets defined by hierarchical clustering of genes significantly regulated upon carbon starvation (see Figure 4E). YEASTRACT search settings were DNA binding plus expression evidence; TF acting as either activator or inhibitor.
- https://cdn.elifesciences.org/articles/70344/elife-70344-supp3-v2.xlsx
-
Supplementary file 4
SNF5 subregions examined by all-atom Monte Carlo simulations.
- https://cdn.elifesciences.org/articles/70344/elife-70344-supp4-v2.xlsx
-
Supplementary file 5
Parameters used for all-atom Monte Carlo simulations.
- https://cdn.elifesciences.org/articles/70344/elife-70344-supp5-v2.xlsx
-
Supplementary file 6
Yeast strains used in this study.
- https://cdn.elifesciences.org/articles/70344/elife-70344-supp6-v2.xlsx
-
Supplementary file 7
Plasmids used in this study.
- https://cdn.elifesciences.org/articles/70344/elife-70344-supp7-v2.xlsx
-
Transparent reporting form
- https://cdn.elifesciences.org/articles/70344/elife-70344-transrepform1-v2.pdf
