Targeting APEX2 to the mRNA encoding fatty acid synthase β in yeast identifies interacting proteins that control its abundance in the cell cycle

Abstract

Profiling the repertoire of proteins associated with a given mRNA during the cell cycle is unstudied. Furthermore, it is easier to ask and answer what mRNAs a specific protein might bind to than the other way around. Here, we implemented an RNA-centric proximity labeling technology at different points in the cell cycle in highly synchronous yeast cultures. To understand how the abundance of FAS1, encoding fatty acid synthase, peaks late in the cell cycle, we identified proteins that interact with the FAS1 transcript in a cell cycle-dependent manner. We used dCas13d-APEX2 fusions to target FAS1 and label nearby proteins, which were then identified by mass spectrometry. The glycolytic enzyme Tdh3p, a known RNA-binding protein, interacted with the FAS1 mRNA, and it was necessary for the periodic abundance of Fas1p in the cell cycle. These results point to unexpected connections between major metabolic pathways. They also underscore the role of mRNA-protein interactions for gene expression during cell division.

FIGURE 1:

Engineered yeast cells express active dCas13d-APEX2 targeting the FAS1 mRNA. (A) Diagram of the engineered bicistronic locus introduced into yeast cells to express dCas13d-APEX2 and gRNAs targeting FAS1 (see Materials and Methods). (B) Schematic of the targeted positions on the FAS1 mRNA. (C) Cells of the indicated genotype, carrying different gRNAs targeting FAS1, express active APEX2 based on the conversion of Amplex Red to resorufin. The cells were processed as described in Materials and Methods. (D) Yeast cells expressing dCas13d-dsRBD-APEX2-V5, see A, target it preferentially to the FAS1 mRNA (ddPCR; see Materials and Methods) was used to measure the levels of FAS1 immunoprecipitated by dCas13d-dsRBD-APEX2-V5. The fold enrichment is on the y-axis, from the strains shown on the x-axis carrying the APEX construct and each of the FAS1 gRNAs depicted in (B), a strain expressing nontargeted APEX (NT_APEX; carrying the APEX-TU1 construct, see Materials and Methods), or the parental strain (BY4742). Transcript levels of FAS1 were normalized against the corresponding transcript levels of UBC6 (see Materials and Methods). The values used to generate the graphs are in Supplemental File S1/Sheet 1.

FIGURE 2:

Biotin labeling conditions in cells expressing dCas13d-APEX2 fusions. The immunoblot displays the signal from biotinylated proteins in cells treated in each condition shown on top. The first lane is from extracts prepared using strain BY4742, and the rest is from extracts using strain FAS1-1. The blot at the bottom is the one shown above before it was processed for immunodetection, stained with Ponceau S to reveal total protein loading.

FIGURE 3:

Proximity labeling of proteins targeting FAS1 in the cell cycle. (A) Schematic overview of our experimental approach. This panel was created with BioRender.com. (B) The cell size (y-axis) of the pools of cells we isolated from each strain is shown for the G1 and non-G1 cells (x-axis). The values used to generate the graphs are in Supplemental File S1/Sheet2. C) Volcano plots depicting the proteins identified by mass spectrometry in the indicated strain (shown above each panel) whose levels changed significantly in G1 versus non-G1 cells, based on the magnitude of the difference (x-axis; Log2-fold change) and statistical significance (y-axis), indicated by the red lines. The analytical and statistical approaches are described in Materials and Methods. The values used to generate the graphs are in Supplemental File S1/Sheet3. Note that the lowest calculated p values from the robust ANOVA were at the 0.0001 level. The input values used in the ANOVA analyses are in Supplemental File S1/Sheets 6, 7, and 8. D) Venn diagram of the proteins we identified to interact with FAS1 in a cell cycle–dependent manner (left set) against two reference sets (PMID_35618506, right; and PMID_26595419, middle). The values used to generate the graph are in Supplemental File S1/Sheet 9. (E) Schematic summary of the mRBPs that bind the FAS1 transcript in G1 or non-G1 phases.

FIGURE 4:

Tdh3p binds FAS1, and it is required for cell cycle–dependent changes in Fas1p levels. (A) Yeast cells expressing the corresponding TAP-tagged alleles were used to immunoprecipitate the indicated TAP-tagged proteins. The levels of the associated FAS1 mRNA in the immunoprecipitates (measured as in Figure 1D; see Materials and Methods) are shown on the y-axis in the strains shown on the x-axis. Transcript levels of FAS1 were normalized against the corresponding transcript levels of UBC6. The values used to generate the graphs are in Supplemental File S1/Sheet 10. B) The abundance of TAP-tagged proteins was monitored in strains of the indicated genotype, as described in Materials and Methods. Samples were collected by elutriation in a rich, undefined medium (YPD) and allowed to progress synchronously in the cell cycle. Experiment-matched loading controls (measuring Pgk1p abundance) were also quantified and shown in parallel. (Top), representative immunoblots, along with the percentage of budded cells (percentage budded) and the cell size (in fL) for each sample. (Bottom), from at least three independent experiments in each case, the TAP and Pgk1p signal intensities were quantified as described in Materials and Methods. The Log2 (expressed ratios) values are on the y-axis, and cell size values are on the x-axis. Loess curves and the standard errors at a 0.95 level are shown. All the immunoblots for this figure are in Supplemental File S2, while the values used to generate the graphs are in Supplemental File S1/Sheet 11.

FIGURE 5:

Altered cells size homeostasis and cell-cycle kinetics in cells lacking Tdh3p. (A) From the synchronous cultures shown in Figure 4B, the percentage of budded cells (y-axis) is shown against the mean cell size (in fL; x-axis). Loess curves and the standard errors at a 0.95 level are shown. (B) From the same experiments as above, the rate of size increase is indicated from the plots of the Ln-transformed cell size values (y-axis) against time (x-axis). The values used to generate the graphs in A and B are in Supplemental File S1/Sheet 11. (C) Box plots showing the mean (left panel) and birth (right panel) size (y-axis) for the indicated strains. Comparisons were made with the nonparametric Kruskal-Wallis rank sum test, and the indicated p values calculated from the pairwise comparisons using the Wilcoxon rank sum test with continuity correction, using R language functions. The values used to generate the graphs are in Supplemental File S1/Sheet 12.