SATAY can identify loss- and gain-of-function mutations conferring antifungal sensitivity or resistance

A-C) Scatter plots comparing the number of sequencing reads mapping to each of the 6603 yeast coding DNA sequences (CDSs) in transposon libraries treated with DMSO versus A) Amphotericin B (0.375 μg/mL) B) Caspofungin (0.015 μg/mL) C) Myriocin (0.2 μg/mL). Each dot represents a single gene. Gene names are colour-coded by biological process. A) Orange - ergosterol homeostasis, pink - TORC1 complex subunits, blue – EGO complex subunits. B) Orange – sphingolipid biosynthesis, pink – β-1,3-glucan biosynthesis, blue – membrane lipid homeostasis. C) Orange – vesicular-mediated transport, pink – GARP complex subunits, blue – phospholipid translocation, teal – retromer subunits, maroon – chromatin remodelling.

D) Transposon insertion maps for five antifungal drug targets. The height of the bars corresponds to the number of sequencing reads, shown on a linear scale that is capped at 500. Horizontal blue bars represent CDSs. Red arrowheads indicate positively selected insertions in the promoter regions of genes encoding drug targets.

E) Transposon insertion maps for four drug efflux pumps (PDR5, SNQ2, YOR1, FLR1) and two transcriptional regulators of the PDR network (PDR3, YAP1). The height of the bars corresponds to the number of sequencing reads, shown on a linear scale that is capped at 500. Pink arrowheads indicate positively selected insertions in promoter regions, or in the YAP1 inhibitory domain. Blue arrowheads indicate regions of FLR1 and YAP1 where insertions were negatively selected.

SATAY can be performed in drug-sensitive strains of S. cerevisiae

A) Scatter plot comparing the number of sequencing reads mapping to each of the 6603 yeast CDSs in a PDR-proficient transposon library treated with DMSO versus Prochloraz (10 μg/mL). Each dot represents a single gene. Gene names are colour-coded by biological process: orange – mitochondrial gene expression, blue – mitochondrial translation, pink – mitochondrial protein processing, teal – mitochondrial genome maintenance, maroon – mitochondrial phospholipid homeostasis.

B) As in (A), for a PDR-attenuated (12geneΔ0HSR) transposon library treated with DMSO versus Prochloraz (0.4 μg/mL). Genes involved with vesicle-mediated transport are coloured blue.

C) Tetrad dissection of SLN1/sln1Δ BdDRK1+/- heterozygote on YPD medium. The genotypes of the non-growing spores have been inferred from the genotypes of the growing ones, assuming a Mendelian meiotic segregation pattern.

D) Growth curves of wild-type, sln1Δ BdDRK1 and SLN1 BdDRK1 strains in SC 2% glucose medium containing either 1% DMSO (v/v), Fludioxonil (0.1 μg/mL) or Iprodione (3 μg/mL). OD600 measurements were recorded every 20 minutes using a Bioscreen C™ instrument. Each data point is the mean OD600 measurement for three independent biological replicates. Error bars show standard error of mean (SEM).

E-F) Scatter plots comparing the number of sequencing reads mapping to each of the 6603 yeast CDSs in a sln1Δ BdDRK1 transposon library treated with (E) Fludioxonil (0.1 μg/mL) or (F) Iprodione (3 μg/mL). Gene names are colour-coded by biological process: blue – HOG signalling pathway, pink – RNA polymerase holoenzyme or mediator complex subunit, orange – chromatin remodelling. Note that YDR010C is a dubious open reading frame (ORF) located in the promoter of the drug-efflux pump SNQ2 (Figure 1E).

Chitosan electrostatically interacts with cell wall mannosylphosphate

A) Chemical structure of Chitosan. Protonated amino groups are shown in red.

B) Minimal Inhibitory Concentrations (MICs) of Chitosan for five strains of S. cerevisiae. All strains are PDR-attenuated except for BY4741. MICs were determined after 24 hours for SC 2% glucose cultures incubated at 30°C, 200 rpm. Note in this figure, BY4741 refers to ByK352 (Supplementary Table 4).

C) Scatter plot comparing the number of sequencing reads mapping to each of the 6603 yeast CDSs for an AD1-8 transposon library treated with acetic acid (0.0004%) or Chitosan (2 μg/mL). Gene names are colour-coded by biological process: blue – mannan chain elaboration on N-glycans of cell wall proteins; orange - CO G complex subunits; pink – intra-Golgi retrograde transport; purple – nucleoside phosphatases required for GDP-mannose transport into the Golgi; teal - manganese transport.

D) Transposon insertion map for MNN4, MNN6 and MNN1 in the AD1-8 transposon library following treatments with acetic acid (0.0004% (v/v)) and Chitosan (2 μg/mL). The height of the bars corresponds to the number of sequencing reads, shown on a linear scale that is capped at 2,000. Horizontal blue bars represents CDSs. Red arrowhead indicates positively selected insertions in the promoter region of MNN1.

E) Growth inhibition curves of a BY4741 strain, AD1-8, and two independent AD1-8 mnn4Δ strains treated with Chitosan in SC 2% glucose medium. OD600 measurements were recorded using a Bioscreen C™ instrument set at 30°C, continuous shaking. Each data point is the mean OD600 measurement for three technical replicates. Error bars show the standard deviation.

F) Alcian blue staining of a BY4741 strain, AD1-8, and two independent AD1-8 mnn4Δ strains. Saturated cultures were treated with 10 μg/mL alcian blue solution in SC 2% glucose medium for 3 minutes. Image is displayed in the red channel, since alcian blue strongly absorbs red light.

G) Quantification of background-subtracted red channel intensity of alcian blue-stained pellets of BY4741 strain, AD1-8, and two independent AD1-8 mnn4Δ strains. Three technical replicates for each strain. Error bars show the standard deviation.

H) Representative images of AD1-8 and AD1-8 mnn4Δ strains following treatment with 25 μg/mL fluorescein-labelled Chitosan (green) for 2 hours in SC 2% glucose medium. Cells were treated in stationary phase. MitoTracker (magenta) was applied at 0.1 μM. Yellow arrowheads indicate representative AD1-8 mnn4Δ cells. Images are Z-stacks. Scale bars represent 5 μm.

I) Model for the mechanism by which Chitosan binds to the yeast cell wall. Model depicts electrostatic interactions (black dashes) between Chitosan’s positively charged amino groups (red) and negatively charged mannosylphosphate groups (blue) on an N-linked glycan chain. Mannosylphosphate groups are added onto α1,2-linked mannose side branches by the mannosylphosphate transferases Mnn4 and Mnn6. Meanwhile, Mnn1 adds terminal α1,3-linked mannose residues (black circles) to α1,2-linked mannose side chains. Terminal α1,3-linked mannose residues compete with mannosylphosphate groups for available α1,2-linked mannose acceptor sites. MNN1 overexpression in AD1-8 likely conferred Chitosan resistance by tipping this competition towards non-phosphorylated α1,3-linked mannose residues.

Hol1 imports ATI-2307, Pentamidine and Iminoctadine into S. cerevisiae

A) Scatter plot comparing the number of sequencing reads mapping to each of the 6603 yeast CDSs in a transposon library after a control treatment versus ATI-2307 treatment (0.000175 μg/mL) in SC -TRP 2% ethanol medium. HOL1 is coloured red. Genes involved in the regulation of the proton-motive force are coloured blue. RMD1 and RMD8 are coloured green.

B) Transposon insertion map for HOL1 for the control and ATI-2307 treatments. The height of the bars corresponds to the number of sequencing reads, shown on a linear scale that is capped at 500. Red arrowhead indicates region upstream of HOL1 where transposon insertions were negatively selected by ATI-2307. This region negatively regulates HOL1 expression in a post-transcriptional manner (Vindu et al., 2021).

C) Growth inhibition curves for wild-type, hol1Δ and hol1Δ CaHOL1 strains treated with ATI-2307. WT and hol1Δ strains were transformed with pRS316 empty vector. OD600 measurements were recorded after 48 hours of growth in SC -URA 2% ethanol medium at 30°C, 200rpm. Each data point is the mean OD600 measurement for three independent biological replicates. Error bars show the standard deviation. Error bars for some measurements were too small to be displayed.

D) Transposon insertion maps for TRK1, HAL4 and HAL5 after the control and ATI-2307 treatments. The height of the bars corresponds to the number of sequencing reads, shown on a linear scale that is capped at 500. Blue arrowhead indicates positively selected insertions in the 3’-end of TRK1. Red arrowheads indicate positively selected insertions in the promoter regions of HAL4 and HAL5.

E) Growth inhibition curves for WT and hol1Δ strains treated with lysate derived from ATI-2307-treated-WT or hol1Δ cells. OD600 measurements were recorded after 40 hours of growth in SC 2% ethanol medium at 30°C, 200rpm. Each data point is the mean of three independent biological replicates. Each culture was inoculated with a total of 64 μL lysate derived from WT or hol1Δ cells. The volume of lysate from ATI-2307-treated versus mock-treated cells is indicated on the x-axis. Error bars show the standard deviation.

F) LC-MS analysis of ATI-2307 uptake in WT, hol1Δ and hol1Δ CaHOL1 cells. Cells were cultured in 200 mL SC 2% ethanol at 30°C, 200rpm from a starting OD600 0.5. Upon reaching OD600 1, cells were treated for 2 hrs with ATI-2307 (0.003 μg/mL). The hol1Δ strains expressing CaHOL1 (pBK994) were grown SC -URA 2% ethanol medium. Quantifications of ATI-2307 were performed using matrix-matched calibration curves. Each dot represents an independent biological replicate (n = 3). Error bars show the standard deviation.

G) Growth inhibition curves for wild-type, hol1Δ and hol1Δ CaHOL1 strains treated with Pentamidine, as in C).

H) LC-MS analysis of Pentamidine uptake in WT or hol1Δ cells. Cells were treated with 1 μg/mL Pentamidine. Other experiment details are identical to those described for ATI-2307 in F).

I) Growth inhibition curves for wild-type, hol1Δ and hol1Δ CaHOL1 strains treated with Iminoctadine, as in C).

J) LC-MS analysis of Iminoctadine uptake in WT or hol1Δ pellets. Cells were treated with 1 μg/mL Iminoctadine. Other experiment details are identical to those described for ATI-2307 in F).

K) Mutations identified in Hol1 from sixteen spe1Δ hol1Δ p[HOL1] strains exhibiting a separation-of-function phenotype. For each strain, all the mutations identified in Hol1 are listed, including synonymous mutations. Non-synonymous mutations shown in red were considered causal of the separation-of-function phenotype. For plasmids bearing multiple non-synonymous mutations in Hol1, the causal mutation was assigned to mutations identified in several independent plasmids e.g. N57 and E327.

L) AlphaFold2 model of the S. cerevisiae Hol1 protein (UniProt Identifier P53389) showing the positions of the amino acid residues that were considered causal of the separation-of-function phenotype when mutated.

Scatter plots comparing the number of transposons mapping in the coding DNA sequence (CDS) of every gene in transposon libraries treated with each compound.

Scatter plots comparing the number of sequencing reads mapping in the coding DNA sequence (CDS) of every gene in transposon libraries treated with each compound.

Scatter plots comparing the number of sequencing reads mapping to the promoters in transposon libraries treated with each compound.

Promoters are defined as the 200 bp upstream of the transcription start sites, as assessed by Xu et al., (2009).

Scatter plots comparing the number of sequencing reads mapping to the promoters in transposon libraries treated with each compound.

Promoters are defined as the 200 bp upstream of the translation start site.

Scatter plots comparing the number of sequencing reads mapping to the promoters in transposon libraries treated with each compound.

Promoters are defined as the 500 bp upstream of the translation start site.

Growth inhibition curves of a BY4741 strain (ByK352), AD1-8, and two independent mnn4Δ AD1-8 strains treated with acetic acid in SC 2% glucose medium. Acetic acid concentrations (% v/v) correspond to those in each Chitosan treatment (i.e. the 200 μg/mL Chitosan treatment contained 0.04% acetic acid (v/v)). End-point OD600 measurements were recorded after 24 hours using a Bioscreen C™ instrument set at 30°C, continuous shaking. Each data point is the mean OD600 measurement for three technical replicates. Error bars show standard deviation (SD). Error bars for some measurements were too small to be displayed.

Representative images of AD1-8 and a BY4741 strain (ByK352) following treatment with 25 μg/mL Chitosan Fluorescein (green) for 2 hours in stationary phase in SC 2% glucose medium. MitoTracker (magenta) was applied at 0.1 μM. Yellow arrowheads indicate representative BY4741 cells. Images are Z-stacks. Scale bar represents 5 μm.

Chemical structures of A) ATI-2307 B) Pentamidine C) Spermidine D) Iminoctadine.

Chromatograms and mass spectra for A) yeast lysate spiked with 1.25

μg/mL ATI-2307 B) lysate obtained from ATI-2307-treated WT sample C) lysate obtained from ATI-2307-treated hol1Δ sample. ATI-2307 was identified with a retention time of 1.30-minutes. For mass spectrometry analysis, ATI-2307 was detected with a +2 charge. Red asterisks denote molecular ion peak ([M + 2H]2+) at m/z = 219.6472. Orange and dark blue asterisks denote isotope peaks. Note that ATI-2307 was not detected in the lysate from the hol1Δ sample.

Chromatograms and mass spectra for A) yeast lysate spiked with 1.25

μg/mL Pentamidine B) lysate obtained from Pentamidine-treated WT sample (diluted 1/10 in yeast lysate) C) lysate obtained from Pentamidine-treated hol1Δ sample. Pentamidine was identified with a retention time of 1.43-minutes. For mass spectrometry analysis, Pentamidine was detected with a +2 charge. Red asterisks denote molecular ion peak ([M + 2H]2+) at m/z = 171.1027. Orange and dark blue asterisks denote isotope peaks.

Chromatograms and mass spectra for A) yeast lysate spiked with 1.25

μg/mL Iminoctadine B) lysate obtained from Iminoctadine-treated WT sample (diluted 1/10 in yeast lysate) C) lysate obtained from Iminoctadine-treated hol1Δ sample. Iminoctadine was identified with a retention time of 1.32-minutes. For mass spectrometry analysis, Iminoctadine was detected with a +2 charge. Red asterisks denote molecular ion peak ([M + 2H]2+) at m/z = 178.6790. Orange and dark blue asterisks denote isotope peaks.

Streaks of spe1Δ and spe1Δ hol1Δ strains on SC 2% ethanol medium supplemented with 1 µM spermidine.

Image taken after 6 days of incubation at 30 °C.

Disc diffusion assays testing ATI-2307, Pentamidine and Iminoctadine susceptibility of separation-of-function Hol1 mutants.

A) hol1Δ transformed with pBK958 (HOL1 WT) (top-left). spe1Δ hol1Δ transformed with pBK958 (HOL1 WT) or one of seven plasmids (pBK995-pBK1001) encoding HOL1 alleles bearing separation-of-function mutations. Non-synonymous mutations in HOL1 for each plasmid are shown above the corresponding plate. For each transformant, 175 μL of saturated culture (OD600 ∼4) was plated on SC 2% ethanol medium supplemented with 1 μM spermidine. On each plate, discs were treated with 17 μL of 0.4 μg/mL ATI-2307 (top-left), 80 μg/mL Pentamidine (bottom) or 30 μg/mL Iminoctadine (top-right). Images taken after 5 days of incubation at 30°C. The SPE1 hol1Δ strain (top-left) did not require the plasmid-borne HOL1 gene for viability on SC 2% ethanol + 1 μM spermidine medium. Hence, this strain can lose the plasmid and grow in the presence of the three drugs. In contrast, the spe1Δ strains required the plasmid-borne HOL1 gene for growth on this medium. As such, the spe1Δ strain transformed with the WT HOL1 gene was susceptible to the three drugs.

B) Table of plasmids used in disc diffusion assay. Non-synonymous mutations shown in red were considered to be causal of the separation-of-function phenotype.

Pairwise sequence alignments of Hol1 protein sequences from Candida species and S. cerevisiae.

Pairwise sequence alignments were performed using EMBOSS Needle. Hol1 protein sequences for each species were obtained from UniProt with the following identifiers: C. albicans Q5AP74; C. auris A0A890CZK8; C. parapsilosis G8BA41; C. tropicalis C5MB76; S. cerevisiae P53389.