A) Zinc limitation in LZM+1 µM ZnCl₂ (LZM1) results in decreased growth of wild-type cells relative to replete LZM+100 µM ZnCl₂ (LZM100). A tsa1Δ mutant shows increased sensitivity to limiting zinc relative to wild type. B) Numbers of low zinc sensitive and resistant deletion strains identified by differential strain sensitivity analysis (DSSA). The number of significantly affected strains identified was greater after more generations of growth. C) Venn diagrams showing the number of genes whose mutants showed growth effects after five and fifteen generations. The degree of overlap is also indicated.

Untagged wild-type BY4743 (panels A, B), gal2Δ (panels C, D), and tsa1Δ (panels E, F) cells were mixed with approximately equal numbers of GFP-expressing BY4743 cells and inoculated into zinc-replete (LZM+100 µM ZnCl₂, panels A, C, E) or zinc-limiting (LZM+1 µM ZnCl₂, panels B, D, F) media and grown for fifteen generations prior to analysis by flow cytometry. Approximately 20,000 total cells per culture were assessed for GFP fluorescence (x-axis) and autofluorescence (y-axis). The red line in each panel marks the boundary between the sub-populations of tagged and untagged cells. Quantitation of these data is provided in Table 1. The elongated distribution of fluorescence in zinc-limited cells is likely due to alterations in cell size and cell wall composition relative to zinc-replete cells and was observed for both GFP fluorescence and autofluorescence.

The fitness scores (the difference in the mean of the log₂ hybridization signal between LZM+1 µM ZnCl₂ and LZM+100 µM ZnCl₂) of these sensitive strains were then used to identify and create a smaller sub-network (283 genes) containing the sensitive genes and the non-sensitive and essential genes that link them through known genetic and physical interactions. The sub-network was then assessed for significant overrepresentation of Gene Ontology (GO) Cellular Component categories. These categories were visualized as a linked network. Node color of categories indicates the significance of representation (white = not identified as significant) and node size indicates the number of genes identified present in each category. Edge arrows indicate hierarchy of GO terms. For clarity, only GO Cellular Component categories with a p-value<0.0005 are shown. A separate GO enrichment assessment identified overrepresentation of all GO categories in the sub-network. This analysis was used to generate visual representations of the GO processes and cellular components identified showing the genes involved in these processes. In these cases, node color indicates the sensitivity of each deletion strain in our study (fitness score). The edge color defines the interaction type between nodes (from the BioGRID database).

Wild-type (BY4743) cells were inoculated into zinc-replete (LZM+100 µM ZnCl₂, +Zn) or zinc-limiting (LZM+1 µM ZnCl₂, −Zn) media containing a range of tunicamycin concentrations and grown overnight prior to measuring the culture optical densities at 600 nm (OD₆₀₀). Hypersensitivity to tunicamycin was observed for zinc-limited cells at 0.1, 0.25, and 0.5 mM tunicamcyin. Data presented are the averages of triplicate cultures for each condition and the error bars indicate ±1 S.D. For most points, the symbols obscure the error bars.

A) Confirmation of the ice2Δ low zinc growth defect. Wild type (BY4743, filled columns) and ice2Δ (BY4743 ice2Δ, open columns) cells were inoculated into LZM supplemented with either 1 or 3 µM ZnCl₂ and grown overnight prior to measuring the culture optical densities at 600 nm (OD₆₀₀). B) Loss of Ice2 causes a zinc-suppressible hyper-induction of the unfolded protein response (UPR). Wild type (BY4743) and homozygous ice2Δ mutant (BY4743 ice2Δ) cells were transformed with the UPRE-lacZ reporter pMCZ-Y and inoculated into low zinc medium (LZM) supplemented with 0.3, 1, 3 or 10 µM ZnCl₂. These cells were then grown overnight prior to measuring β–galactosidase activity. C) Loss of Ice2 exacerbates the zinc-suppressible hyper-induction of the UPR in msc2Δ zrg17Δ zrc1Δ cot1Δ quadruple mutants. JSY5 (msc2Δ zrg17Δ zrc1Δ cot1Δ) and JSY5 ice2Δ (msc2Δ zrg17Δ zrc1Δ cot1Δ ice2Δ) cells were transformed with the UPRE-lacZ reporter pMCZ-Y and inoculated into low zinc medium (LZM) supplemented with 1, 3, 10 or 100 µM ZnCl₂. These cells were then grown overnight prior to measuring β–galactosidase activity. The wild-type strain used was the isogenic CM100 strain. D) Zinc treatment does not inhibit the UPR induction in response to tunicamycin. Wild-type BY4743 cells bearing the UPRE-lacZ reporter were grown to exponential phase in LZM supplemented with the indicated concentration of zinc, then treated for 2 hours with 2 µg/ml tunicamycin prior to β–galactosidase activity assay. Data presented are the averages of triplicate cultures for each condition and the error bars indicate ±1 S.D.