Summary Summary (text) Abstract Abstract (text) MEDLINE XML PMID List

#### Send to
jQuery(document).ready( function () {
jQuery("#send_to_menu input[type='radio']").click( function () {
var selectedValue = jQuery(this).val().toLowerCase();
var selectedDiv = jQuery("#send_to_menu div." + selectedValue);
if(selectedDiv.is(":hidden")){
jQuery("#send_to_menu div.submenu:visible").slideUp();
selectedDiv.slideDown();
}
});
});
jQuery("#sendto").bind("ncbipopperclose", function(){
jQuery("#send_to_menu div.submenu:visible").css("display","none");
jQuery("#send_to_menu input[type='radio']:checked").attr("checked",false);
});

File Clipboard Collections E-mail Order My Bibliography Citation manager

Format Summary (text) Abstract (text) MEDLINE XML PMID List CSV

- 1 selected item: 31050341
Format Summary Summary (text) Abstract Abstract (text) MEDLINE XML PMID List MeSH and Other Data E-mail Subject Additional text

Generate a file for use with external citation management software.

# Stable Pom1 clusters form a glucose-modulated concentration gradient that regulates mitotic entry.

### Author information

- 1
- Department of Biochemistry and Cell Biology, The Geisel School of Medicine at Dartmouth, Hanover, United States.

### Abstract

Control of cell size requires molecular size sensors that are coupled to the cell cycle. Rod-shaped fission yeast cells divide at a threshold size partly due to Cdr2 kinase, which forms nodes at the medial cell cortex where it inhibits the Cdk1-inhibitor Wee1. Pom1 kinase phosphorylates and inhibits Cdr2, and forms cortical concentration gradients from cell poles. Pom1 inhibits Cdr2 signaling to Wee1 specifically in small cells, but the time and place of their regulatory interactions were unclear. We show that Pom1 forms stable oligomeric clusters that dynamically sample the cell cortex. Binding frequency is patterned into a concentration gradient by the polarity landmarks Tea1 and Tea4. Pom1 clusters colocalize with Cdr2 nodes, forming a glucose-modulated inhibitory threshold against node activation. Our work reveals how Pom1-Cdr2-Wee1 operates in multiprotein clusters at the cortex to promote mitotic entry at a cell size that can be modified by nutrient availability.

© 2019, Allard et al.

#### KEYWORDS:

S. pombe; cdr2; cell biology; cell cycle; kinase; pombe; size; yeast

**A**) Schematic of the Cdr2-Pom1-Wee1 signal transduction pathway and its coarse cellular localization. (

**B**) Individual frame (top panel) and sum projection (bottom panel) of a high-speed TIRF microscopy movie. Movie was continuous 15 s time-lapse acquisition of 200 ms exposures. Scale bar 1 µm. Blue arrows mark position of line scans performed for data in panel (

**D**). (

**C**) Pom1-mNG also forms clusters at the cell tip. Image is a sum projection of three consecutive 200 ms exposures from continuous time-lapse TIRF movie of the cell tip as depicted in the cartoon diagram. Scale bar 1 µm. (

**D**) Line scans of fluorescence intensity along the long axis of the snap shot (blue line, left Y axis) and projection images (red line, right Y axis) in panel (

**B**). Note that time-averaging of Pom1 clusters smoothens the concentration gradient. (

**E**) Comparison of Pom1 cluster binding frequency at the cell tip or side (****p=<0.0001, n = 10 cells, 42–170 traces/cell). (

**F**) Comparison of cortical dwell time of individual Pom1-mNG clusters (n.s., p=0.6747,

*n*= 10 cells, 42–170 traces/cell) at the cell tip or side. For (

**E–F**), each data point represents a single cell mean, and line and error bars represent mean and standard deviation of all cells. Statistical significance was tested using a Student’s T-test.

**A**) Pom1-mNG clusters are apparent at the lateral cell cortex of cells in individual frames (200 ms) of spinning-disc microscopy movies (top panel) but become averaged out in a projection of the entire movie (bottom panel). Sum projection is of a continuous 15 s time-lapse acquisition of 200 ms exposures at the top cell focal plane. Scale bar 1 µm. (

**B**) Line scans of fluorescence intensity along the long axis of the cell in the snap shot (blue line, left Y axis) and projection image (red line, right Y axis) at positions marked by blue arrow heads in panel (

**A**). Note that time-averaging of Pom1 clusters smoothens the concentration gradient. (

**C**) Representative images of Airyscan super-resolution micrographs of Pom1-mNG at cell tip or side. Scale bar, 1 µm. (

**D**) Fluorescence intensities of all single clusters resolved in Airyscan images as in panel (

**C**). Data points represent individual cluster values, error bars are the mean and standard deviation (

*n*= 87 tip, 158 side). (

**E**) Histogram of the fluorescence intensity of individual Pom1 clusters at the cell tip and cell side. Clusters were binned by fluorescence intensity and displayed as a population frequency (

*n*= 48 tip, 158 side).

**A**) Pom1-mNG clusters positioned at the cell middle, in a 2 × 2 µm ROI as used for particle tracking. (

**B**) Single-particle diffusion traces of Pom1-mNG clusters show limited lateral diffusion in cells. (

**C**) Mean squared displacement (MSD) curves calculated from traces of individual Pom1-mNG cluster diffusion as in panel (

**B**). (

**D**) Mean MSD curves generated from averaging of 439 single-particle traces, coming from six different cells. Linear regression of the initial linear segment (red line) of the curve provides the diffusion coefficient D = 0.134 ± 0.017 µm

^{2}/s. Black line represents the weighted mean of all traces, and gray lines indicate standard deviation. (

**E**) Average life-time displacement of individual clusters (***p=0.0002,

*n*= 10 cells, 42–170 traces/cell). Line and error represent mean and standard deviation. Statistical significance was tested using a Student’s T-test.

**A**) TIRF microscopy image of Pom1-mNG clusters in the extruded cytoplasm (white dotted line) of a lysed cell (yellow dotted line). Scale bar 5 µm. (

**B**) TIRF microscopy images of cell extracts prepared from wild-type (no tag) or Pom1-mNG cells. Images are 50 frame sum projections of continuous 200 ms time-lapse exposures. The two images were contrasted equally. Scale bars 1 µm. (

**C**) Cytoplasmic extracts of

*pom1-3HA*cells were subjected to velocity sucrose gradient sedimentation, and fractions were probed against the HA tag (upper blot). Fraction one corresponds to the top of the gradient and contains smaller structures; fraction nine corresponds to bottom of the gradient and contains larger structures. Fractions 6–8 were pooled, sucrose was removed by dialysis, and then the sample was subjected to a second identical round of sucrose gradient sedimentation and western blotting of the resulting fractions (lower blot). (

**D**) TIRF microscopy of Pom1-mNG clusters from cytoplasmic extracts on supported lipid bilayers. Scale bar 1 µm. Left panel is single time point image. Right panel is kymograph taken from a line scan of time-lapse TIRF experiment. (

**E**) Quantification of binding duration of Pom1-mNG clusters on supported lipid bilayers imaged by TIRF microscopy as in panel (

**D**). Values are compared to cellular measurements of Pom1 clusters on cell sides (n.s., p=0.05954, n = 713 in vivo, 421 in vitro). (

**F**) Quantification of life-time displacement of Pom1-mNG clusters diffusing on supported lipid bilayers imaged by TIRF microscopy as in panel (

**D**). Values are compared to cellular measurements of Pom1 clusters on cell sides (****p<0.0001, n = 713 in vivo, 421 in vitro). For (

**E**) and (

**F**), statistical significance was tested using a Student’s T-test. (

**G**) Purified GST-Pom1 was subjected to sucrose gradient sedimentation and the fractions were probed against the GST tag (upper blot). Purified GST-Pom1 was also added to wild-type or

*pom1∆*cell extracts and incubated for 1 hr at 4 ˚C in the presence of ATP before velocity sucrose sedimentation and western blotting (bottom blots).

**A**) Fluorescence intensity trace of a single Pom1-mNG cluster binding to the cell cortex, measured by high speed TIRF microscopy (~50 frames per second with continuous acquisition). Rapid intensity changes indicating binding and unbinding are marked by green upward and magenta downward pointing arrows respectively. (

**B**) FRAP analysis of supported lipid bilayers. Upper panel: A circular ROI (teal dashed line) was used to bleach fluorescent PE in the bilayers. Lower panel: Kymograph at white dashed line in upper panel of fluorescence recovery as monitored by TIRF microscopy, indicating fluidity. (

**C**) Quantification and comparison of binding duration of either wild-type or kinase-dead Pom1-mNG clusters to supported lipid bilayers in TIRF microscopy movies (***p=<0.0001,

*n*= 1526 wild-type, 1993 kinase-dead). Statistical significance was tested with a Student’s T-test. (

**D**) TIRF microscopy images of cell extracts prepared from cells co-expressing Pom1-mNG and Cdr2-tagRFP-t. Pom1 clusters are shown in green and Cdr2 nodes are shown in magenta. Images are three frame sum projections of continuous 200 ms time-lapse exposures. Scale bar 1 µm.

**A**) Quantification of Pom1-3HA band intensities from western blot of yeast extracts in main (). (

**B**) Size standards for velocity sucrose gradients, as in main (). Data are quantified from single Coomassie-stained gels or using spectrophotometry data. (

**C**) Plot of S-value versus sedimentation peak. Line is the linear regression of the three size standards, with the S-value for Pom1 peaks interpolated.

**A**) Pom1-mNG clusters diffusing on supported lipid bilayers imaged by high-speed (100 ms acquisition) TIRF microscopy. (

**B**) Single-particle diffusion traces of Pom1-mNG clusters on supported lipid bilayers. (

**C**) Mean squared displacement (MSD) curves calculated from traces of individual Pom1-mNG cluster diffusion as in panel (

**B**). (

**D**) Mean MSD curves generated from averaging of 421 single-particle traces. Weighted linear regression of the curve (red line) provides the diffusion coefficient D = 0.168 ± 0.018 µm

^{2}/s. Note that the weighted regression best fits the initial region of the curve because few clusters bind longer than 10 s. Black line represents weighted mean of all traces, gray lines indicate standard deviation, and error bars represent weighted standard deviation.

**A**) Western blots of sucrose gradient fractions after sedimentation analysis as in of extracts prepared from

*tea1∆*or

*tea4∆*cells expressing Pom1-3HA. (

**B**) TIRF micrographs of Pom1-mNG localization at the tip or side cortex in

*tea1∆*or

*tea4∆*cells. Scale bar 1 µm. (

**C**) The mean binding frequency of Pom1-mNG clusters at the tips and sides of wild-type,

*tea1∆*or

*tea4∆*cells (

*n*= 10 cells, 7–170 traces/cell). (

**D**) The mean dwell time of Pom1-mNG clusters at the tips and sides of wild-type,

*tea1∆*or

*tea4∆*cells. For panels C-D, wild-type data are replotted from main (). Each data point represents a single cell mean, and line and error bars represent mean and standard deviation of all cells (

*n*= 10 cells, 7–170 traces/cell). Comparisons are 1-way ANOVA with Tukey’s multiple comparisons tests. (****) indicates p<0.0001, (n.s.) indicates p>0.05. See for all results of statistical analysis.

**A**) Confocal micrographs of Pom1-mNG in

*tea1∆*or

*tea4∆*cells. Images are from the middle focal plane; scale bar 5 µm. (

**B**) Quantification of cortical binding dwell time of individual Pom1-mNG clusters at the tips or sides of representative wild-type,

*tea1∆*or

*tea4∆*cells (

*n*= 44–129 traces). (

**C**) Quantification of total lifetime displacement of individual Pom1-mNG clusters at the tips or sides of representative wild-type,

*tea1∆*or

*tea4∆*cells. For B-C, line and error represent single cell mean and standard deviation (

*n*= 44–129 traces). (

**D**) Results of ANOVA and Tukey’s multiple comparison tests performed in main ().

**A**) A subset of Pom1 clusters colocalize with Cdr2 nodes. Panels are TIRF microscopy images of cells expressing Pom1-mNG and Cdr2-tagRFP-t. Yellow dashed brackets outline the ROI of the lower zoomed panels. Orange arrows point to Pom1 clusters colocalized with Cdr2 nodes. Scale bar is 1 µm. (

**B**) Kymographs generated using the ROI indicated in panel A. Cyan arrows indicate prominent Pom1 clusters, magenta arrows indicate prominent Cdr2 nodes, and orange arrows indicate colocalization. (

**C**) Localization of Cdr2-mEGFP (upper panels) or Pom1-mNG (lower panels) in representative cells of increasing size, imaged by TIRF microscopy. Scale bar is 1 µm. (

**D**) Total number of Cdr2-mEGFP nodes (blue circles) or Pom1-mNG clusters (red squares) measured per cell. Quantification is limited to clusters detected and resolvable in the TIRF illumination field. The slopes of the corresponding linear regressions are not significantly different (p=0.3757,

*n*> 40 cells). (

**E**) Density of Cdr2-mEGFP nodes (blue circles) or Pom1-mNG clusters (red squares) in 2 × 2 µm square ROIs at the cell middle, counted using TIRF microscopy. The slopes of the corresponding linear regressions are significantly different (p<0.0001,

*n*> 40 cells). (

**F**) Example of colocalization analysis for Pom1 clusters (magenta ROI) and Cdr2 nodes (cyan ROI) (left panels). Colocalized structures are marked with overlapping yellow ROIs, whereas non-colocalizing structures are marked with gray ROIs (right panel, gray circles). (

**G**) Ratio of free Cdr2 nodes to free Pom1 clusters, plotted as a function of cell size. The slope of the linear regression is positive and significantly non-zero (p=0.0002, R

^{2}= 0.33,

*n*= 36 cells).

**A**) Schematic showing the ROI locations used to measure the concentration of Pom1-mNG or Cdr2-mNG in the cytoplasm, cortex, tip cortex, or side cortex. Images are spinning disc confocal micrographs of middle cell focal planes. Scale bars 1 µm. (

**B**) Quantification of Cdr2 concentration (average fluorescence per pixel) at different cellular ROIs as in panel (

**A**). Each dot represents a measurement from a different cell, numbers and the line and error represent the mean and standard deviation. (

**C**) Quantification of Pom1 concentration (average fluorescence per pixel) at different cellular ROIs as in panel A. Each dot represents a measurement from a different cell, numbers and the line and error represent the mean and standard deviation. (

**D**) Quantification of the cytoplasmic concentration (average fluorescence per pixel) of Cdr2-mNG as a function of cell length, using a cytoplasmic ROI as in panel (

**A**). The slope of the linear regression does not significantly deviate from zero (b

_{1}≠0, (n.s.) p=0.6449). Each point indicates a measurement from a single cell. (

**E**) Quantification of the cytoplasmic concentration (average fluorescence) of Pom1-mNG as a function of cell length, using a cytoplasmic ROI as in panel (

**A**). The slope of the linear regression does not significantly deviate from zero (b

_{1}≠0, (n.s.) p=0.4926). Each point indicates a measurement from a single cell.

**A**) A subset of Pom1 clusters colocalize with Cdr2 nodes. Images are dual-channel simultaneously-acquired TIRF microscopy images of cells expressing Pom1-tdTomato and Cdr2-mNeonGreen. Yellow dashed brackets outline the ROI of the lower zoomed panels. White arrows point to Pom1 clusters colocalized with Cdr2 nodes. Scale bar is 1 µm. (

**B**) Kymographs generated using the ROI drawn on the merged image in panel A. Cyan arrows indicate prominent Cdr2 nodes, magenta arrows indicate prominent Pom1 clusters, and orange arrows adjacent to the merged kymograph indicate colocalization.

**A**) Comparison of Pom1 cluster binding frequency on the cell sides of wild-type and

*cdr2∆*cells (n.s., p=0.1076,

*n*> 10 cells, 25–42 traces/cell). (

**B**) Comparison of Pom1 cluster binding duration on the cell sides of wild-type and

*cdr2∆*cells (n.s., p=0.371,

*n*> 10 cells, 25–42 traces/cell). (

**C**) Comparison of Pom1 cluster displacement on the cell sides of wild-type and

*cdr2∆*cells (n.s., p=0.8191,

*n*> 10 cells, 25–42 traces/cell). In A-C, significance was tested using a Student’s T-test. (

**D**) Schematic and quantification of Pom1 concentration at the lateral cell cortex measured by confocal microscopy, using 2 µm line scans at the cell middle from images such as . The slope of the linear regression is negative, but not significantly non-zero (p=0.2603, R

^{2}= 0.03, n = 45 cells). (

**E**) Plot of the number of colocalized structures as a function of cell size. The slope of the linear regression is significantly non-zero (p=0.0006, R

^{2}= 0.3,

*n*= 36 cells). (

**F**) Plot of free, non-colocalized Cdr2 nodes (blue circles) and free Pom1 clusters (red squares). The slope of the linear regression of #Pom1 clusters versus cell length is negative, and significantly non-zero (p=0.001, R

^{2}= 0.27,

*n*= 36 cells). The slope of the linear regression of #Cdr2 nodes versus cell length is positive, and significantly non-zero (p=0.021, R

^{2}= 0.15,

*n*= 36 cells).

**A**) Localization of Pom1-mNG in cells grown in either high (2%) or low (0.03%) glucose media. Images were collected using TIRF microscopy. Scale bar, 5 µm. (

**B**) Comparison of the binding frequency of Pom1 clusters at the lateral cell cortex in cells grown under high and low glucose (***p=0.0003,

*n*= 10 cells, 32–113 traces/cell). Statistical significance was tested using a Student’s T-test. (

**C**) Comparison of the binding dwell time of Pom1 clusters at the lateral cell cortex in cells grown under high and low glucose (n.s., p=0.6833,

*n*= 10 cells, 32–113 traces/cell). Statistical significance was tested using a Student’s T-test. (

**D**) Localization of Cdr2-mNG in wild-type or

*pom1∆*cells grown in either normal (2%) or low (0.03%) glucose media. Images were collected using TIRF microscopy. Scale bar, 1 µm. (

**E**) Comparison of the fluorescence intensity of individual Cdr2-mNG nodes in wild-type or

*pom1∆*cells grown in either normal (2%) or low (0.03%) glucose media. Low glucose induces partial node disassembly in wild-type cells (****p<0.0001,

*n*= 92–169 nodes from >5 cells) but not in

*pom1∆*cells (n.s., p=0.4015,

*n*= 113–156 nodes from >5 cells). Measurements were taken from Airyscan Super-Resolution confocal micrographs. (

**F**) Quantification and comparison of the total number of Cdr2-mNG nodes visible in TIRF micrographs of wild-type or

*pom1∆*cells grown in either normal (2%) or low (0.03%) glucose media. There is no significant difference in any condition (p>0.05,

*n*= 28–33 cells). Statistical significance was tested using a one-way ANOVA.

**A**) Analysis of Pom1 and Cdr2 redistribution by spinning-disc confocal microscopy, as in . Images are confocal micrographs of cells co-expressing Pom1-tdTomato and Cdr2-mNG in high (2%) or low (0.03%) glucose media. Scale bar, 5 µm. (

**B**) Cortical fluorescence intensity of Cdr2 and Pom1 was measured using line scans drawn around the circumference of the representative cell marked with an arrowhead in the 2% glucose portion of panel A. (

**C**) Cortical fluorescence intensity of Cdr2 and Pom1 was measured using line scans drawn around the circumference of the representative cell marked with an arrowhead in the 0.03% glucose portion of panel A.

**A–B**) Localization of Pom1-mNG (panel A) and Cdr2-mNG (panel B) in confocal micrographs at middle cell focal planes. Regions where line-scans were used to measure fluorescence intensity are marked at the tip and side (orange lines). Scale bar, 1 µm. (

**C**) Concentration of Pom1 (red squares) or Cdr2 (blue circles) at cell sides in a range of glucose concentrations, measured using ‘Side ROI’ as in panels A and B (

*n*> 10 cells/concentration). (

**D**) Concentration of Pom1 (red squares) or Cdr2 (blue circles) at cell tips in a range of glucose concentrations, measured using ‘Tip ROI’ as in panels A and B (

*n*> 10 cells/concentration). (

**E**) Correlation of Cdr2 vs Pom1 concentrations at cell sides in each glucose concentration from panel C. Concentrations are anticorrelated across all tested media glucose concentrations (p=0.0006, Pearson r = −0.9352,

*n*= 8 concentrations). (

**F**) Correlation of Cdr2 vs Pom1 concentrations at cell tips in each glucose concentration from panel D. Concentrations show weak anticorrelation across the tested media glucose concentrations, but the correlation is not statistically significant (p=0.09, Pearson r = −0.65,

*n*= 8 concentrations).

**A**) Localization of Wee1-mNG in wild-type or

*pom1∆*cells grown in either high (2%) or low (0.03%) glucose media. Images were collected using TIRF microscopy. Scale bar, 5 µm. (

**B**) Quantification of Wee1 bursting kinetics in wild-type cells grown under high (2%, blue circles) and low (0.03%, red squares) glucose conditions. The top panel is a plot of the total number of Wee1 bursts as a function of cell length, counted in single time point TIRF micrographs. The middle panel is a plot of the frequency of Wee1 bursts as a function of cell length. The bottom panel is a plot of Wee1 burst duration as a function of cell length. (

**C**) Quantification of Wee1 bursting kinetics in

*pom1∆*cells grown under normal (2%, green circles) and low (0.03%, magenta squares) glucose conditions, as in panel B. For (

**B–C**), insets represent whether the best fit values of the slope (

**b**) and/or Y-intercept (

_{1}**b**) of the two linear regressions are different at the 95% confidence level. See and for parameters of linear regressions and full statistical comparisons. (

_{0}**D**) Western blot of Wee1 in wild-type

*, pom1∆*, and

*cdr2∆*cells grown in high (2%) or low (0.03%) glucose media showing phosphorylation dependent band-shifts.

**A**) Plot of the frequency of Wee1 bursts as a function of cell length in wild-type cells grown in high glucose (blue circles) and low glucose (red squares), or kinase-dead

*cdr2(E177A)*mutant cells grown in high glucose (green triangles). (

**B**) Plot of Wee1 burst duration as a function of cell length in wild-type cells grown in low glucose (red squares) and kinase-dead

*cdr2(E177A)*mutant cells grown in high glucose (green triangles). For (

**A–B**), insets represent whether the best fit values of the slope (

**b**) and/or Y-intercept (

_{1}**b**) of the linear regressions are different at the 95% confidence level. See and for parameters of linear regressions and full statistical comparisons.

_{0}### Conflict of interest statement

CA, HO, JM No competing interests declared