Measuring affinities of fission yeast spindle pole body proteins in live cells across the cell cycle

Abstract

Characterizing protein-protein interactions is essential for understanding molecular mechanisms, although reproducing cellular conditions in vitro is challenging and some proteins are difficult to purify. We developed a method to measure binding to cellular structures using fission yeast cells as reaction vessels. We varied the concentrations of Sid2p and Mob1p (proteins of the septation initiation network) and measured their binding to spindle pole bodies (SPBs), the centrosome equivalent of yeast. From our measurements we infer that Sid2p and Mob1p both exist as monomeric, heterodimeric, and homodimeric species throughout the cell cycle. During interphase these species have widely different affinities for their common receptor Cdc11p on the SPB. The data support a model with a subset of Cdc11p binding the heterodimeric species with a Kd < 0.1 μM when Sid2p binds Mob1p-Cdc11p and Kd in the micromolar range when Mob1p binds Sid2p-Cdc11p. During mitosis an additional species presumed to be the phosphorylated Sid2p-Mob1p heterodimer binds SPBs with a lower affinity. Homodimers of Sid2p or Mob1p bind to the rest of Cdc11p at SPBs with lower affinity: Kds > 10 μM during interphase and somewhat stronger during mitosis. These measurements allowed us to account for the fluctuations in Sid2p binding to SPBs throughout the cell cycle.

Figure 1

Contrast-independent segmentation of images of fission yeast cells expressing a range of concentrations of the SIN kinase Sid2p. (A–D) Cells expressed mYFP-Sid2p at a range of levels determined by the promoter and inducing conditions. These cells also expressed the nucleoporin Seh1p-mCFP to mark the nuclear envelope and Sad1p-2x(mCFP) to mark the SPB. The medium contained membrane-impermeant dye Alexa 647 to outline cells. The seven columns correspond (L–R) to 1), counts of mean numbers (±SD) of the total fluorescent mYFP-Sid2p molecules present in each cell of the selected field; (2), a DIC micrograph; 3), a sum projection fluorescent micrograph of the signal from mYFP-Sid2p; 4), a sum projection fluorescent micrograph of the signal from Seh1p-mCFP and Sad1p-2x(mCFP); 5), sum projection fluorescent micrograph of signal from Alexa 647 to outline cells; 6), a merged fluorescent micrograph showing (green) mYFP-Sid2p, (red) Sad1p-2x(mCFP), and Seh1-mCFP and (blue) Alexa 647; and 7), automatically segmented results showing individual cells in green and SPBs of any fluorescence intensity in yellow. Each row is one field of cells with the expression of Sid2p controlled by different promoters and inducing conditions: (A) the 41nmt1 promoter mYFP-Sid2p repressed by thiamine for 18 h; (B) WT mYFP-Sid2p under the control of the native promoter; (C) the 41nmt1 promoter induced by removing thiamine from growth media for 18 h; and (D) overexpression from the 3nmt1 promoter induced by removing thiamine from growth media for 18 h. Scale bar 5 μm.

Figure 2

Measurements by quantitative fluorescence microscopy of total and local (SPB) numbers of cytokinesis proteins across the cell cycle. (A) Cartoon of S. pombe life cycle. Bin designations are based on cell volumes (average volume is 92 μm³), the positions of SPBs and the presence of a complete septum from DIC images: (I1) interphase cells with apparent volume <80 μm³; (I2) cells with volumes between 80 and 100 μm³; (I3) interphase cells with volumes >100 μm³ before SPB separation; (M1) anaphase A; (M2) anaphase B to maximum SPB separation; (M3) after maximum separation but before complete septum formation; and (I0) complete septum formation without cells splitting. Scale bar 5 μm. (B) Total numbers of molecules per cell (●) of SIN proteins (top to bottom) Sid2p, Mob1p, and Cdc11p throughout the cell cycle. Horizontal solid lines are average counts of total molecules across the cell cycle with ±30% error as dashed lines. (C) Absolute counts of molecules localized to the SPB with the same format as in (B). In cells with two SPBs (●) represent active dim SPBs and (○) represent inactive bright SPBs.

Figure 3

Affinity of (A–F) Sid2p and (G–L) Mob1p for the SPB in live interphase cells in volume bin I3 representing Late G2/M. (A) Dependence of the numbers of mYFP-Sid2p molecules associated with the SPB on the cytoplasmic concentration of mYFP-Sid2p in bin I3. A hyperbola representing a simple bimolecular reaction did not fit the data well. (B) Substructure in the data was revealed by smoothing with a robust local weighted regression (rlowess) using a span of the square root of number of cells measured. A nodal point at 3 μM Sid2p separates the reactions. (Inset) Cartoon of late G2/M cell. Scale bar 5 μm. (C) Hyperbolic fit to low concentrations of [Sid2p]_free isolates an initial plateau for Reaction 1 in which Receptor 1 binds Ligand 1. (D) After subtracting the contribution of the reaction from the measurements cells expressing all concentrations of [Sid2p]_free, the residual data have a sigmoidal shape that can be fit with a Hill equation assuming cooperativity or multimer Sid2p binding. We define this as Reaction 2 with Ligand 2 binding to Receptor 2. (E) The fit of the sum of a hyperbola and sigmoid, done exclusively with raw data. (F) Same fit sumperimposed on smoothed data to evaluate the goodness of fit. R² is 0.95. (G–L) Affinity of mYFP-Mob1p for the SPB in live cells during interphase. The data for cells expressing a range of mYFP-Mob1p concentrations were analyzed using the same procedures used for Sid2p (panels A–F).

Figure 4

Changes in the affinity of Sid2p for the SPB across the cell cycle reveals a cycle of high- and low-affinity states that can be used to calculate the number of species present on SPB throughout the cell cycle. (A) Cartoon of portions of S. pombe life cycle from which affinity measurements were derived. Scale bar 5 μm. (B) Log plot of the EC50 (μM) measured from (□, ■) the low concentrations of Sid2p that bind receptor type 1, and (○, ●) the high concentrations of Sid2p that bind receptor type 2 throughout the cell cycle. The cells in bins I0 and I1 were mixtures of cells at different cell cycle stages, so they could not be analyzed. (C) Plot of distribution of Sid2p under WT conditions at the SPB throughout the cell cycle. Calculated species of (□, ■) total heterodimer, (○, ●) Sid2p homodimer, and (▵, ▴) monomer bound to the SPB in each cell cycle bin given the WT concentrations of Sid2p and Mob1p in the cell, the total Mob1p and Cdc11p bound to the SPB, and determined affinities from (B) and Table S2. Evaluating the affinity of Sid2p for type 1 receptors given the total number of Mob1p bound to the SPB in WT cells provides the Mob1p monomer bound and heterodimer bound, evaluating the affinity of Mob1p for type 1 receptors given the total number of heterodimer bound provides the Sid2p monomer bound, and evaluating the affinity of Sid2p for type 2 receptors given the total number of Cdc11p receptors in WT cells and the calculated occupied receptors from monomeric and heterodimeric bound species gives the number of Sid2p homodimers bound. Bound Sid2p was left out of the calculations so that it could be used to determine efficacy of our method. (D) Comparison of deriving bound Sid2p from free concentrations and affinities with measuring Sid2p bound to SPB throughout the cell cycle in WT cells. (Striped bars) Measured WT Sid2p bound to SPB as in Fig. 2C top panel. (Solid bars) Summation of bound heterodimer, twice the bound homodimers, and bound Sid2p monomer from part (C). Solid symbols and bars represent dim active daughter SPBs. Open symbols and bars represent bright inactive mother SPBs. Asterisks indicate low estimation of bound Sid2p from calculations.