Evolutionarily conserved sites in yeast tropomyosin function in cell polarity, transport and contractile ring formation

ABSTRACT Tropomyosin is a coiled-coil protein that binds and regulates actin filaments. The tropomyosin gene in Schizosaccharomyces pombe, cdc8, is required for formation of actin cables, contractile rings, and polar localization of actin patches. The roles of conserved residues were investigated in gene replacement mutants. The work validates an evolution-based approach to identify tropomyosin functions in living cells and sites of potential interactions with other proteins. A cdc8 mutant with near-normal actin affinity affects patch polarization and vacuole fusion, possibly by affecting Myo52p, a class V myosin, function. The presence of labile residual cell attachments suggests a delay in completion of cell division and redistribution of cell patches following cytokinesis. Another mutant with a mild phenotype is synthetic negative with GFP-fimbrin, inferring involvement of the mutated tropomyosin sites in interaction between the two proteins. Proteins that assemble in the contractile ring region before actin do so in a mutant cdc8 strain that cannot assemble condensed actin rings, yet some cells can divide. Of general significance, LifeAct-GFP negatively affects the actin cytoskeleton, indicating caution in its use as a biomarker for actin filaments.


Fig. S1. Cell length and nuclear number of wildtype and cdc8 mutants.
Wildtype and mutant strains were grown to mid-log phase in YEA at 30°C. Nuclear number and cell length were determined in Calcofluor/DAPI stained micrographs using Image J 1.43.
Left: Fraction of mononuclear and binuclear cells. The cell density was calculated using a hemocytometer.
Right: Length distribution of the cells. Cell length was measured in pixels and converted to μm using a conversion factor of 1 pixel = 0.0535 µm obtained using an AO micrometer (2 mm divisions subdivided into units of 10 μm). The distributions were plotted by rounding the lengths to nearest μm and sorting them into groups.
The results suggest that the cdc8 mutations do not affect length or nuclear number.
Nuclear number was determined from two (three for cdc8 WT ) independent measurements, each >200 cells.
Length measurements were obtained from two (three for cdc8 WT ) independent measurements, each comprising >50 binuclear, septated cells.  The strains cdc8 Antibody. Rabbit anti-Cdc8p was prepared by Covance (Princeton, NJ) and affinity purified by Genscript (Piscataway, NJ). Recombinant AlaSerCdc8p was expressed and purified from E. coli (Materials and Methods). Crude serum crossreacts with a single band in immunoblots of purified AS-Cdc8p or crude extracts.
Indirect immunofluorescence. For immunofluorescence imaging, cells were grown in YEA at 30°C overnight to mid-log phase. 0.6×10 7 cells were fixed in 4% paraformaldehyde (EM grade) at 30°C for 5 min, inverting once after 2.5 min. The cells from each strain were washed with PBS and resuspended in 140 μl 1.2 M sorbitol. 60 μl protoplasting solution was added (8.5 mg/ml Zymolase, 3 mg/ml final; 12 mg/ml lysing enzyme, final 5 mg/ml in 1.2M sorbitol). The cells were incubated at room temperature for 10-20 min and visually checked for protoplasting by mixing an aliquot 1:1 with 10% SDS. Protoplasted cells burst and disintegrate. To inactivate the protoplasting enzymes, 1 ml of 1% Triton was added and incubated for 2 min. Following centrifugation, cells were blocked by resuspension in 0.5 ml PBAL (10% BSA, 100 mM lysine HCl, 50 ng/ml Carbencillin, 1 mM NaN3 in PBS) and incubated for 1 hr at room temperature, gently rocking. Following centrifugation, primary, affinity-purified anti-Cdc8p was added (100 μl, 1:10 in PBAL) and incubated overnight at 4°C. Cells were then spun down and washed thrice in 500 μl PBAL. Secondary antibody (Jackson ImmunoResearch Laboratories, West Grove, PA, 111-165-003, Cy3-goat anti-rabbit IgG (H+L)) was added (1:100 in PBAL) for 90 min. at room temperature. After six washes in PBAL the cells were ready for imaging. Cells were stored at 4°C and imaged within two days (details for fluorescence microscopy in Materials and Methods for fluorescence microscopy). The background with secondary antibody alone was negligible in the exposure times used (100 msec).  A. cdc8 WT -crn1-GFP cells (FC661) with monopolar, asymmetric, and bipolar patch distribution. The ratio of fluorescence intensity at the two ends of the cell was used to determine the polarity of the patch distribution: monopolar (≤0.33), bipolar (≥0.66), asymmetric bipolar (  cdc8 WT (SH107) and cdc8 E104A cells (SH98) expressing Myo52p-3x-YFP were grown in YEA. Myo52p is initially present at cell tips and in the contractile ring. Samples were spun down, washed, and resuspended in 1 ml H2O and incubated at 30°C for 10 min. Myo52p-3x-YFP is visible in patches throughout the cell since it has redistributed to facilitate vacuole fusion. Samples were spun down, resuspended in YEA and incubated at 30°C for 10 min. Myo52p-3x-YFP returned to its normal distribution.  Strains designated as grcdc8 + are strains used for marker reconstitution mutagenesis as described in methods. In most cases the fluorescent protein (FP) strains were crossed with SH30 as well as the mutant strains so that the all fluorescent protein strains are ura4 + . We did not note any morphological differences between the grcdc8 + FP strains and the original FP strains obtained from colleagues. Consequently we did not create FP grcdc8 + for myo1, crn1, or LifeAct.