Cytokinetic nodes in fission yeast arise from two distinct types of nodes that merge during interphase

Abstract

We investigated the assembly of cortical nodes that generate the cytokinetic contractile ring in fission yeast. Observations of cells expressing fluorescent fusion proteins revealed two types of interphase nodes. Type 1 nodes containing kinase Cdr1p, kinase Cdr2p, and anillin Mid1p form in the cortex around the nucleus early in G2. Type 2 nodes with protein Blt1p, guanosine triphosphate exchange factor Gef2p, and kinesin Klp8p emerge from contractile ring remnants. Quantitative measurements and computer simulations showed that these two types of nodes come together by a diffuse-and-capture mechanism: type 2 nodes diffuse to the equator and are captured by stationary type 1 nodes. During mitosis, cytokinetic nodes with Mid1p and all of the type 2 node markers incorporate into the contractile ring, whereas type 1 nodes with Cdr1p and Cdr2p follow the separating nuclei before dispersing into the cytoplasm, dependent on septation initiation network signaling. The two types of interphase nodes follow parallel branches of the pathway to prepare nodes for cytokinesis.

Figure 1.

Two types of interphase nodes observed by confocal fluorescence microscopy of asynchronous cells expressing fluorescent fusion proteins. (A and B) Localization of kinase Cdr2p-mEGFP, node protein Blt1p-mCherry, and SPB protein Sad1p-mRFP (black arrowheads in the reverse contrast image of the red channel). (A) Maximum intensity projections of five z sections in 400-nm steps including 2 µm (about half the thickness) of the cells. Cdr2p and Blt1p colocalize in nodes near the equator of some cells (arrow) but in separate nodes in other cells (paired white arrowheads). In cells with Blt1p concentrated in the contractile ring (single white arrowhead), Cdr2p is dispersed in the cytoplasm. (B) Cells such as in A were classified into interphase stages by length and other criteria (Materials and methods). (C and D) Localization of proteins in two types of nodes across the cell cycle. Reverse contrast, confocal fluorescence, and maximum intensity projections of 21 z sections taken at 260-nm steps and arranged in montages according to cell length. We adjusted the brightness and contrast of each image to account for differences in illumination intensity and exposure time. (bottom) Drawings of the distributions of the two types of nodes across the cell cycle. (C) Type 1 nodes located in broad bands around the equator (green boxes in one column) contain Cdr2p-mYFP and Cdr1p-3GFP. Mid1p-mEGFP is present in these nodes and transiently in the contractile ring (red box). (D) Type 2 nodes containing Blt1p-mYFP, mECitrine-Gef2p, and Klp8p-mYFP appear at the division site, redistribute to the equator during interphase (red boxes), and incorporate into the contractile ring. (E) Histogram of the distribution of nodes along the length of asynchronous interphase cells such as in A and B. Nodes were defined as colocalized if their intensity centers of mass in the red and green channels were less than one point spread function (3 pixels) apart. n = 188 nodes from 12 cells in two separate experiments. Dotted ovals outline cells. Bars, 2 µm.

Figure 2.

Localization of type 1 nodes across the cell cycle. Images are time series of maximum intensity projections of confocal fluorescence micrographs with time in minutes from SPB separation except in G. (A) Reverse contrast and merged images early in mitosis of a cell expressing Cdr2p-mEGFP, Blt1p-mCherry, and Sad1p-mRFP. Blt1p incorporated into the contractile ring as type 1 nodes moved from the equator and dispersed Cdr2p into the cytoplasm. Projections of four z sections taken at 400-nm steps. (B) Demonstration of the asymmetric movements of the nodes from the midline by a false-color kymograph of the time course of Cdr2p-mEGFP fluorescence along the length of the cell in Fig. 2 A. At each time point, the fluorescence intensity across the cell width was summed to obtain a fluorescence profile and colored according to intensity per area. (C) Images of a cell expressing Cdr2p-mEGFP and Cdc7p-mCherry showing that Cdr2p nodes moved in greater numbers and to a larger extent (arrowhead) in the direction of the active SPB marked with Cdc7p-mCherry. The same correlation was present in every cell we observed (n > 13). (green images) 10 z sections taken at 360-nm steps comprising half the cell. (red images) 20 z sections taken at 360-nm steps comprising the entire cell. (D) Time series of confocal maximum intensity projection images of a cell expressing Cdr2p-mEGFP, Blt1p-mCherry, and Sad1p-mRFP showing Cdr2p nodes reappearing in a cortical band of type 1 nodes around the daughter nuclei ∼10 min before the cell separated at time 80 min. Same conditions as A. (E) Outcome plots of the time course of the localization of Cdr2p-mEGFP to nodes in wild-type (n = 21; closed circles) and temperature-sensitive sid2-250 cells (n = 28; open circles). Cells were imaged at a permissive temperature (23–25°C). In wild-type cells, Cdr2p disappeared from nodes into the cytoplasm for the 40-min eclipse period. Most sid2-250 cells never dispersed Cdr2p from their nodes; in 18% of cells, Cdr2p disappeared from nodes into the cytoplasm for an abbreviated 1–3-min eclipse period. (F) Time-lapse images of temperature-sensitive sid2-250 cells expressing Cdr2p-mEGFP and Sad1p-mCherry at a permissive temperature (23–25°C) showing incomplete dispersal of type 1 nodes marked with Cdr2p-mEGFP. Projections of 20 z sections taken at 360-nm steps. (G) Movement of type 1 nodes from the equator of cells treated with 131 µM MBC to depolymerize microtubules. Time series of images at 3-min intervals of cells expressing Cdr2p-mEGFP and Sad1p-mCherry. The arrowhead points to Cdr2 nodes after movement from the equator. The arrow points to divided SPBs. Images from a representative experiment repeated twice. (H) Plot of mean squared displacement (MSD) versus time step for a sample of type 1 nodes imaged at 1-s intervals and marked with Cdr2p-mEGFP during interphase (n = 23; dark green) and mitosis (n = 6; light green). Bars, 2 µm.

Figure 3.

Redistribution of type 2 nodes in cells during the cell cycle. (A) This time series of maximum intensity projections of reverse contrast fluorescence micrographs shows type 2 nodes marked with Blt1p-mEGFP emerging from the disassembling contractile ring and moving to the equator over 2 h. (B) Outcome plots of the time course of the localization of type 2 interphase nodes with time 0 at SPB separation, calculated as time from cell separation −80 min. Symbols: fraction of cells with ≥1 Blt1p node at new cell equator (closed squares); fraction of cells with nodes distributed along the cortex (defined as the first time point when the mean fluorescence at the old division site is less than or equal to the mean fluorescence at any other point along the length; closed triangles); and fraction of cells with Blt1p nodes maximally relocalized to the new cell equator (closed circles). (C) Histogram of the proportion of interphase nodes containing Blt1p only, Cdr2p only, or colocalized Blt1p and Cdr2p binned by stage of the cell cycle; same data as Fig. 1 E. (D) Plots of mean squared displacement (MSD) versus time step for five type 2 (Blt1p-mEGFP) nodes from a single interphase cell imaged at 2-s intervals. (E) Temporal color projections of node movements at 2-s intervals to compare the behavior of type 2 nodes marked with Blt1p-mEGFP (top) and type 1 nodes marked with Cdr2p-mEGFP at different stages of interphase (bottom). Each pair of micrographs has a reversed contrast sum fluorescence micrograph (three 300-nm z slices) on the left. Streaks of color show movements according to the time scale. White spots are stationary. The leftmost images are of one of two daughter cells still connected by a septum. Dotted ovals outline cells. (F) Plot of node diffusion coefficients in interphase cells as a function of the normalized distance from the new end of the cell showing that type 2 nodes diffuse faster near the cell tips than near the equator. Early interphase: cells still connected by septum or ≤8.5 µm in length; late interphase: cells >8.5 µm long. Type 2 Blt1p-mEGFP nodes (n = 49); type 2 Klp8p-mEGFP nodes (n = 23); type 1 Cdr2p-mEGFP nodes (n = 23). Bars, 2 µm.

Figure 4.

Behavior of type 2 nodes marked with Blt1p-mEGFP in cdr2Δ cells lacking type 1 nodes. (A) MSDs versus time step of Blt1p-mEGFP nodes in cdr2Δ cells. Symbols: five type 2 nodes with intensity <12,000 A.U. (open triangles); three Blt1p-mEGFP particles with fluorescence intensity >12,000 A.U. (closed triangles), likely aggregates of nodes. (B) Plot of diffusion coefficients of Blt1p-mEGFP nodes in cdr2Δ cells during early interphase as a function of their fractional distance from the new end of the cell. Symbols: nodes with intensity <12,000 A.U. (open triangles); particles with intensity ≥12,000 A.U. (closed triangles). (C) Pairs of micrographs with a negative contrast sum intensity projection image of three 300-nm z slices (left) and a temporal color projection image at 2-s intervals for 200 s in early and late interphase cells expressing Blt1p-mEGFP in a cdr2Δ strain (right). Many more equatorial nodes were mobile in early G2 than G2/M (also see Fig. S4). Dotted ovals outline cells. (D) Time series of reverse contrast maximum intensity projections of seven z sections taken at 600-nm steps showing the assembly of a contractile ring in a cdr2Δ cell expressing Blt1p-mEGFP. See Fig. S4 for higher time resolution. (E) Outcome plots of the time course of contractile ring assembly and constriction in cells expressing type 2 node protein Blt1p-mEGFP ±Cdr2p. Symbols: wild-type cdr2⁺ cells (closed circles and squares); cdr2Δ cells (open circles and squares); fraction of cells with a compact contractile ring (circles); and fraction of cells with a constricting contractile ring (squares). Time is in minutes from SPB separation. Bars, 2 µm.

Figure 5.

Simulations of a mathematical model of a diffuse-and-capture mechanism to redistribute type 2 nodes from the old to the new division site. (A) Model of diffusing type 2 nodes being captured irreversibly (yellow) by stationary type 1 nodes (green) with a probability of binding p_bind if they reside within a defined capture radius R_c at any time point. (B) Drawing of the observations of type 2 nodes (red) redistributing from their origin at the new end of the cell (x = 0) until they colocalize with type 1 nodes in a broad band near the center of the cell. Y is the circumference of the cell. (C) Graphical output of a diffuse-and-capture simulation at the time points indicated. In this simulation, 65 type 2 nodes (red) started at x = 0 and diffused with D = 400 nm²/s until they were captured (yellow) by passing within 60 nm of one of the 65 type 1 nodes (green) located in a Gaussian distribution across a band centered 3 ± 0.8 µm from the new cell tip. The probability of capture was 50%. The simulation used steps of 2 s. Bar, 1 µm. Time expressed in minutes from the beginning of simulation, which corresponds to t = ∼40 min from SPB separation. Also see Fig. S5 J, in which some of the data are displayed for visual comparison with Fig. 1 B. Bar, 2 µm. (D) 10 simulations of the time course of binding of type 2 nodes to type 1 nodes with the parameters used in C. The gray region is the mean ± 1 SD of the time when type 2 nodes appeared maximally bound in live cells. (E) Dependence of simulated time courses of type 2 nodes binding to type 1 nodes on the diffusion coefficient of type 2 nodes. The horizontal dashed line indicates when 50% (bottom line) and 80% (top line) of the type 2 nodes were bound. Vertical dashed line is the mean time that type 2 nodes were maximally bound in live cells. The gray region is the mean ± 1 SD. (F) Number of captured type 2 nodes out of 65 at the end of the simulation as a function of diffusion coefficient D (n = 100). (G) Mean positions in simulations of captured type 2 nodes relative to the center of the band of type 1 nodes at 0, which is 3 µm from the new cell tip (n = 100). Error bars are ± 1 SD.

Figure 6.

Summary of the node cycle with cell cycle time on the y axis. Column 1 shows cartoons of cells with type 1 nodes in green, type 2 nodes in red, and colocalized type 1 and 2 nodes in red/green. The green cell shows Cdr1p and Cdr2p dispersed in the cytoplasm during the ∼40-min eclipse period (Fig. 2 E and Table S1). Columns 2, 3, and 4 indicate the times of events. The light green and red background shading shows when node markers are organized in structures. Horizontal lines on the side of each box mark the mean time of each event. Pairs of horizontal lines represent the beginning and end of events spread over time. The time when Mid1p is handed off from type 1 to type 2 nodes is not known.