Skip to main page content
U.S. flag

An official website of the United States government

Dot gov

The .gov means it’s official.
Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

Https

The site is secure.
The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2010 Jun 14;11(9):1291-301.
doi: 10.1002/cbic.200900681.

A high-throughput screen for chemical inhibitors of exocytic transport in yeast

Lisha Zhang  1 , N Miranda Nebane, Krister Wennerberg, Yujie Li, Valerie Neubauer, Judith V Hobrath, Sara McKellip, Lynn Rasmussen, Nice Shindo, Melinda Sosa, Joseph A Maddry, Subramaniam Ananthan, Gary A Piazza, E Lucile White, Edina Harsay
Affiliations

A high-throughput screen for chemical inhibitors of exocytic transport in yeast

Lisha Zhang et al. Chembiochem. .

Abstract

Most of the components of the membrane and protein traffic machinery were discovered by perturbing their functions, either with bioactive compounds or by mutations. However, the mechanisms responsible for exocytic transport vesicle formation at the Golgi and endosomes are still largely unknown. Both the exocytic traffic routes and the signaling pathways that regulate these routes are highly complex and robust, so that defects can be overcome by alternate pathways or mechanisms. A classical yeast genetic screen designed to account for the robustness of the exocytic pathway identified a novel conserved gene, AVL9, which functions in late exocytic transport. We now describe a chemical-genetic version of the mutant screen, in which we performed a high-throughput phenotypic screen of a large compound library and identified novel small-molecule secretory inhibitors. To maximize the number and diversity of our hits, the screen was performed in a pdr5Delta snq2Delta mutant background, which lacks two transporters responsible for pleiotropic drug resistance. However, we found that deletion of both transporters reduced the fitness of our screen strain, whereas the pdr5Delta mutation had a relatively small effect on growth and was also the more important transporter mutation for conferring sensitivity to our hits. In this and similar chemical-genetic yeast screens, using just a single pump mutation might be sufficient for increasing hit diversity while minimizing the physiological effects of transporter mutations.

PubMed Disclaimer

Figures

Figure 1
Figure 1
Summary of a high-throughput screen for identifying compounds that preferentially inhibit the growth of a mutant strain having apl2Δ and vps1Δ mutations. PubChem Assay ID’s (AID) are indicated for each screen step.
Figure 2
Figure 2
Z’ factors for the high-throughput primary screen. The screen for the background strain (JTY2953, snq2Δ pdr5Δ) was done in two batches. The screen with the mutant strain (LZY35, vps1Δ apl2Δ snq2Δ pdr5Δ) was done in three batches, and a fourth batch was a repeat of plates with ambiguous results.
Figure 3
Figure 3
Summary of dose-response end-point growth assays for identifying 93 hit compounds. A) Structural clustering of compounds from the 93 hits. PubChem SID’s for our designated SR numbers are shown in Table 2. The SR number corresponds to the rank from the dose-response assay, with SR2 being the top-ranked compound. B) Data from the dose-response assays allowed grouping of compounds according to specificity for mutants (PubChem AID 788, 789, 790). The majority of our hit compounds were specific for the apl2Δ vps1Δ double mutant. Among the top 56 confirmed compounds (ranked according to dose response growth assay), 13 compounds caused inhibition due to the vps1Δ mutation and were similarly toxic for the apl2Δ vps1Δ mutant; 12 compounds inhibited primarily the apl2Δ vps1Δ mutant but also had significant effect on the apl2Δ mutant; and 31 compounds inhibited the growth of primarily the the apl2Δ vps1Δ double mutant with little or no effect on the apl2Δ or vps1Δ mutants. None of the top 56 compounds inhibited growth the apl2Δ vps1Δ mutant due only to the apl2Δ mutation.
Figure 4
Figure 4
Exponential-growth-rate assays of mutant and background strains. A) Shaking-culture growth assays showed that the pdr5Δ mutation is sufficient for sensitizing yeast strains to most of the compounds (at 20µM) identified as hits in our dose-response assay. Rates were determined from the equation of an exponential curve fitted to the exponential growth curve of each strain (correlation coefficient >0.9) and shown as a percent of the rate without compound added. The sequence of SR numbers corresponds with ranking in the dose-response growth assay, with SR2 being the highest-ranked hit. SR numbers are color-coded according to mutant specificity, as determined by dose-response assays (PubChem AID 788, 789, 790). Corresponding PubChem SID’s are shown in Table 2. B) The pdr5Δ and snq2Δ mutations confer a growth defect on a vps1Δ apl2Δ strain, which is most pronounced if both pdr5Δ and snq2Δ mutations are present and least-pronounced with pdr5Δ. Rates were determined as in (A). C) Microwell-plate exponential growth curves of strains having a pdr5Δ mutation. Strains for A-C are as follows: LZY109 (snq2Δ); LZY108 (pdr5Δ); LZY96 (apl2Δ vps1Δ snq2Δ); LZY104 (apl2Δ vps1Δ pdr5Δ); NY10 (wild-type = “wt”); LZY53 (snq2Δ pdr5Δ); EHY644 (apl2Δ vps1Δ); LZY82 (apl2Δ vps1Δ snq2Δ pdr5Δ = “quad”); LZY119 (vps1Δ pdr5Δ) and LZY120 (apl2Δ pdr5Δ).
Figure 5
Figure 5
The largest structural group of hit compounds had greatest specificity for the apl2Δ vps1Δ mutant for inhibiting growth and secretion. A) Compound structures with SR# and PubChem SID. B) 3-dimensional overlay of the five compounds shown in (A), with carbons in each compound colored as the SID #. C) Dose-response end-point growth assay for SR5. The yeast strains are LZY53 (background); LZY81 (vps1Δ); LZY80 (apl2Δ); and LZY82 (vps1Δ apl2Δ); all four strains have snq2Δ pdr5Δ mutations. D) Microwell plate growth curves for SR5 (compare to Figure 4C). E) Invertase secretion assay for SR5 shows a secretion defect specific for vps1Δ apl2Δ. Strains are as in Figure 4C. The means of three experiments (from three independent cultures) are shown. Error bars, SEM. F) Strain LZY104 (apl2Δ vps1Δ pdr5Δ) grown in microwell plates, in either 2.5 μM SR5 or DMSO control. There is no growth defect until ~4h after adding compound.
Figure 6
Figure 6
SR28 was the most potent secretory inhibitor obtained from the screen. A) Structure of SR28. B) Dose-response end-point growth assay for SR28. C) Microwell plate growth curves for strains grown in 5 μM SR28. D) Invertase secretion assay for strains grown in 2 μM or 5 μM SR28. All strains had a secretory defect in this compound, and the defect was most severe for the vps1Δ and vps1Δ apl2Δ mutants. The means of three experiments (from three independent cultures) are shown. Error bars, SEM. E) Western blots showing the accumulation of internal Bgl2 in 5 μM SR28 at the indicated times. Bgl2 is almost entirely in the cell wall at steady state, so the internal accumulation of this protein can be detected by removing the cell wall. PGK (a cytoplasmic protein) is shown as a loading control. Strains are as in Figure 4C.
Figure 7
Figure 7
SR9 is a vps1Δ-specific growth and secretory inhibitor. A) Structure of SR9 (left) and an analog (right) with similar activity. B) Dose-response end-point growth assay for SR9 (strains as in Figure 5C). C) Microwell plate growth curves for strains grown in 5 μM SR9 (strains as in Figure 4C). D) Invertase secretion assay for strains grown in 5 μM SR9. The secretory defect is dependent entirely on the vps1Δ mutation. The means of three experiments (from three independent cultures) are shown. Error bars, SEM. E) Western blots showing accumulation of internal Bgl2 after growing strains in 5 μM SR9 for the indicated times. PGK is a loading control.
Figure 8
Figure 8
Cluster 2 and similar compounds were highly selective for the vps1Δ apl2Δ double mutant. A) Structures for two Cluster 2 and one closely related compound. B) Additional compounds that resemble Cluster 2 compounds. C) The Cluster 2 compound, SR2, ranked highest in our dose-response screen. It is highly selective for inhibiting the growth of the apl2Δ vps1Δ double mutant in microwell plates (shown) and agar plates (not shown). D) Compounds shown in (B) were also specific for the vps1Δ apl2Δ mutant. Yeast strains are as in Figure 4C.
See this image and copyright information in PMC

Similar articles

See all similar articles

Cited by

References

    1. Zaman S, Lippman SI, Zhao X, Broach JR. Annu. Rev. Genetics. 2008;42:27–81. - PubMed
    1. Chen RE, Thorner J. Biochimica et biophysica acta. 2007;1773(8):1311–1340. - PMC - PubMed
    1. Novick P, Field C, Schekman R. Cell. 1980;21:205–215. - PubMed
    1. Schekman R. Nat. Cell Biol. 2004;6(6):483–486. - PubMed
    1. Harsay E, Bretscher A. J. Cell Biol. 1995;131(2):297–310. - PMC - PubMed

Publication types

MeSH terms

Substances

LinkOut - more resources