et al.

Abstract
Full Text
Signaling and Circuitry of Multiple MAPK Pathways Revealed by a Matrix of Global Gene Expression Profiles
Christopher J. Roberts, Bryce Nelson, Matthew J. Marton, Roland Stoughton, Michael R. Meyer, Holly A. Bennett, Yudong D. He, Hongyue Dai, Wynn L. Walker, Timothy R. Hughes, Mike Tyers, Charles Boone, and Stephen H. Friend

Supplementary Material

To download the data for the gene responses in the 56 microarray experiments included in this study, as well as spreadsheets describing the 16 correlation plots (text Fig. 2, B to E and G to I; text Fig. 3, A and C to D; web figs. 1, 3, 4, and 7 to 9), please see www.rii.com/TECH/pubs.htm.

Figure 1. Correlation plot of response profiles induced by 30-min treatments with α-factor versus a-factor. BY4741 [wt, MATa; C. B. Brachmann et al., Yeast14, 115 (1998)] 50 nM α-factor (αF), for 30 min (Figure 2A), compared with BY4742 [wt, MATα) 500nM a-factor for 30 min. See Figure 2 legend and (11) for further details. Correlation coefficient ρ = 0.92 (colored stars), 0.81 (all genes).


Medium version | Full size version


Figure 2. Vegetative profile of R276 (wt) versus R4044 (kss1Δ), plotted against the mean intensity of each spot. Genes induced in the kss1Δ strain relative to wt are plotted as positive values. The KSS1 gene is down nearly 10-fold; several genes regulated by TEC1 (e.g., NDJ1 and YLR343W; not shown, YLR334C) are down 1.5- to 3-fold; several pheromone-induced genes (e.g., FIG1 and YML047C; not shown, ASG7, FUS1) are down 1.3- to 1.6-fold.


Medium version | Full size version


Figure 3. Correlation plot of pheromone response expression profiles of wild-type (Figure 2A; Figure 5, experiments 2 and 5) and R4044 (kss1Δ) cells (Figure 5, experiment 8). kss1Δ cells were mock treated or treated with 50 nM α-factor (αF) for 30 min, and transcript profiles were determined by hybridizatin to a pair of microarrays. Specific genes of interest are labeled; response profiles are correlated at ρ = 0.96 (colored stars), 0.90 (all genes).


Medium version | Full size version


Figure 4. Correlation plot of pheromone response profiles of wild-type (Figure 2A; experiments 2 and 5) and R404 (fus3Δ) cells (Figure 5, experiment 19). fus3Δ cells were mock treated or treated with 50 nM α-factor (αF) for 30 min, and transcript profiles were determined by hybridization to a pair of microarrays. Specific genes of interest are labeled; response profiles are correlated at ρ = 0.83 (colored stars), 0.70 (all genes).


Medium version | Full size version


Figure 5. Vegetative profile of Y1543 (tec1Δ) cells. Data are displayed as mean intensity [log10(intensity)] for each probe versus the red/green ratio [log10(expression ratio)] for ech probe. Genes induced in the tec1Δ strain relative to wt are plotted as positive values. The TEC1 transcript is down 40-fold in the tec1Δ mutant. Several genes are reduced 1.7- to 2.9-fold, including YLR042C, which has several Tec1p consensus binding sites in its promoter, and many open reading frames (ORFs) with Δ elements in their promoters, including NDJ1, YLR334C, and numerous other genes labeled "Ty element ORFs." These include YBL005W-B, YJR029W, YOL106W, YHR214C-B, YMR045C, YCR018C-A, YDR034C-A, YER138W-A, YMR046W-A, YMR158C-B, and YPR002C-A. The YLR343W ORF does not have a Δ element in its promoter, but a Δ element is located immediately 3' to the gene.


Medium version | Full size version


Figure 6. Vegetative profile of Y1787 (fus3Δtec1Δ). Data are displayed as mean intensity [log10(intensity)] for each probe versus the red/green ratio [log10(expression ratio)] for each probe. Genes induced in the fus3Δtec1Δ strain relative to wild type are plotted as positive values. The FUS3 and TEC1 transcripts are both down greater than 10-fold in the fus3ΔteclΔ mutant. The genes repressed in tec1Δ cells (web fig. 4; e.g., NDJ1, FIG1, and the Ty element ORFs) are also repressed in the fus3Δtec1Δ mutant. The genes that were induced in fus3Δ cells (Figure 3B; PGU1, KSS1, and YLR042C) are not unduced in the fus3Δtec1Δ mutant, indicating that these genes are regulated by Tec1p.


Medium version | Full size version


Figure 7. Correlation plot of pheromone response profiles of wild-type (Figure 2A; experiments 2 and 5) and Y1787 (fus3Δtec1Δ) cells. fus3Δtec1Δ cells were mock treated or treated with 50 nM α-factor (αF) for 30 min, and the transcript profiles were determined by hybridization to a pair of microarrays. Specific genes of interest are labeled; response profiles are correlated at ρ = 0.74 (colored stars), 0.59 (all genes).


Medium version | Full size version


Figure 8. Comparison of the pheromone response profile of wild-type cells (Fig. 2A; Fig. 5, experiments 2 and 5) and the vegetative profile of Y2121 (rst1Δrst2Δtec1Δ). Note that genes defined as Kss1p regulated (PGU1, YLR042C, SVS1 and KSS1), which were induced in the rst1Δrst2Δ vegetative profile (see Fig. 3A) are not induced in the rst1Δrst2Δtec1Δ mutant. Specific genes of interest are labeled; response profiles are correlated at ρ = 0.62 (colored stars), 0.48 (all genes).


Medium version | Full size version


Figure 9. Correlation plot of the pheromone response profiles of wild-type (Figure 2A; Figure 5, experiments 2 and 5) and Y1612 (rst1Δrst2Δ) cells (Figure 5, experiment 30). rst1Δrst2Δ cells were mock treated or treated with 50 nM α-factor (αF) for 30 min, and the transcript profiles were determined by hybridization to a pair of microarrays. Specific genes of interest are labeled; the response profiles are correlated at ρ = 0.72 (colored stars), 0.51 (all genes).


Medium version | Full size version


Figure 10. lacZ reporter assay of Kss1p-regulated genes. The relevant genotypes of host cells are indicated on the side. p3019 (CEN LEU2) carries FG::TY1-lacZ [H. D. Madhani and G. R. Fink, Science275, 1314 (1997)]. V85, a YCplac111-based plasmid containing the lacZ gene, was used for reporter construction. Promoter sequences were amplified by PCR from W303-1A (MATaura3-1 leu2-3,112 his3-11,15 trp1-1 ade2-1 can1-100) genomic DNA. The 3' downstream primer included the start codon of the gene, designed for an in-frame fusion with lacZ. The 5' upstream primer determined the size of the promoter-containing PCR product: YLR042C, 600 bp; PGU1, 743 bp; FUS3, 690 bp; KSS1, 791 bp; FUS1, 834 bp; FIG1, 800 bp; SVS1, 793 bp; YPL192C, 800 bp. Haploid MATa cells containing reporter plasmids were grown to midlogarithmic phase in SC medium lacking leucine and assayed for β-galactosidase activity as described [D. C. Hagen, G. McCaffrey, G. F. Sprague Jr., Mol. Cell. Biol. 11, 2952 (1991)]. For pheromone assays, cells were grown similarly and treated with synthetic α-factor (500 nM) for 2 hours. β-Galactosidase activity was calculated in Miller units (averages and standard deviations of three independent assays). Each value represents the average and stndard deviation for three cultures. For confirmation, experiments were repeated as an independent culture.


Medium version | Full size version


Figure 11. Expanded view of the cluster diagram in Fig. 5, showing all 400 gene names.


Medium version | Full size version