Fig 1.
Protein expression control during the mating response and the analysis workflow.
(A) Signal transduction and regulation of protein expression in the yeast mating response. The control of protein concentration is subject to two levels of regulation: genome-wide regulation relevant to the physiological response and gene-specific regulation, which was identified by our work. (B) Analysis workflow to uncover the regulatory circuits underlying distinct cell fates.
Fig 2.
High-resolution profiling of protein expression and synthesis rates.
(A) Protein expression, growth rate and protein synthesis rates for the investigated genes in yeast cells in response to a high concentration of pheromone. The order of the genes in the heatmap is determined via hierarchical clustering of the gene expression profiles. Values are normalized and transformed to a z-score. (B) The normalized values of protein expression, growth rate and protein synthesis rate for genes in yeast cells exposed to the intermediate concentration of pheromone.
Fig 3.
Mechanism of PSDA and identified regulation events in shmooing cells.
(A) PSDA deconvolutes the gene-specific regulation by modeling the effect of the physiological response of the yeast cells. (B) Examples for PSDA results of six different strains. The dashed line indicates the protein dynamic changes caused by global regulation and the dots represent the observed protein abundance. Time window of gene-specific regulation are gray shaded. Fold change of regulation is indicated; green arrow (activation), red arrow (inhibition). (C) Identified gene regulation events in shmooing cells. Values are the log-transformed fold change of the regulation level. Distinct temporal modes were distinguished by k-means clustering, and the 50% activation/inhibition time for each cluster is represented by the dashed line. C1-C6 denote clusters 1–6. Right column, proteins responsible for the cell cycle (dark brown), conjunction (brown) and chemical response (light brown).
Fig 4.
Reconstruction of the regulatory circuit responsible for shmoo formation.
(A) Workflow of the reverse engineering process. A Boolean network model was used to deduce the constraints on the network interactions from the discrete time trajectory. After generating all possible networks, the minimal network constraint and prior knowledge about signal transduction were incorporated to select for minimal circuits responsible for the shmoo formation. (B) The resulting regulatory circuit. Solid and dashed edges are used to denote the canonical pheromone response pathway and novel regulations, respectively. Edges with bar-end, regulation of inhibition; edges with arrow-end, activation. The input of the network is the TFs activated by pheromone that can directly regulate gene expression, such as Ste12 and Tec1.
Fig 5.
Dose-dependent regulation patterns and mechanism for cell fate determination.
(A) Regulation modes of the 6 gene groups in elongated cells (colored in blue), and in shmooing cells (colored in red). The activity of each cluster was averaged from all of the gene-specific regulation information for the corresponding genes, with regulation level normalized to 1/-1 for activation/inhibition. Several clusters show divergent behaviors in the two phenotypes, as in observed for C1 and C3, which contributes to the differential regulation that determines the cell fate switching. Schematic illustration of active gene groups and regulatory circuits responsible for different cell fates, including vegetative growth (B), elongated growth (C) and shmooing (D). Edges that result in the activation/ inhibition of gene groups are illustrated.
Fig 6.
Pheromone-dependent translocation of Sfp1 suggests a dose-dependent crosstalk between the pheromone response pathway and the TOR-regulated ribosomal biogenesis pathway.
Time course of Sfp1-GFP localization in cells committed to different cell fates, as indicated.