Essential role for Schizosaccharomyces pombe pik1 in septation

Abstract

Background: Schizosaccharomyces pombe pik1 encodes a phosphatidylinositol 4-kinase, reported to bind Cdc4, but not Cdc4(G107S).

Principal findings: Gene deletion revealed that pik1 is essential. In cells with pik1 deleted, ectopic expression of a loss-of-function allele, created by fusion to a temperature-sensitive dihydrofolate reductase, allowed normal cell proliferation at 25 degrees C. At 36 degrees C, cells arrested with abnormally thick, misplaced or supernumerary septa, indicating a defect late in septation. In addition to being Golgi associated, ectopically expressed GFP-tagged Pik1 was observed at the medial cell plane late in cytokinesis. New alleles, created by site-directed mutagenesis, were expressed ectopically. Lipid kinase and Cdc4-binding activity assays were performed. Pik1(D709A) was kinase-dead, but bound Cdc4. Pik1(R838A) did not bind Cdc4, but was an active kinase. Genomic integration of these substitutions in S. pombe and complementation studies in Saccharomyces cerevisiae pik1-101 cells revealed that D709 is essential in both cases while R838 is dispensable. In S. pombe, ectopic expression of pik1 was dominantly lethal; while, pik1(D709A,R838A) was innocuous, pik1(R838A) was almost innocuous, and pik1(D709A) produced partial lethality and septation defects. The pik1 ectopic expression lethal phenotype was suppressed in cdc4(G107S). Thus, D709 is essential for kinase activity and septation.

Conclusions: Pik1 kinase activity is required for septation. The Pik1 R838 residue is required for important protein-protein interactions, possibly with Cdc4.

Figure 1. Tetrad analysis following S. pombe pik1 disruption and mutant allele integration.

(A) Diploid cells pik1/pik1 or pik1/Δpik1::ura4 (N1231) were incubated on ME plates to obtain azygotic asci. Spores were separated and incubated on YES plates. Results shown are representative of 8–10 asci. Each spore formed a colony when both chromosomal pik1 loci were intact (left panel). Only two of the four spores formed colonies when only one chromosomal locus was intact (right panel; i). In cases where colonies did not form, the site of spore deposition was examined microscopically and photographed (right panel; ii). Many spores failed to germinate. Some spores germinated but the resulting cell divided only once. (B) A diploid strain (N1550) homozygous for the wild-type pik1 coding region (WT), or heterozygous diploid strains carrying either the D709A substitution (N1565) or the R838A substitution (N1582), or both (N1596) were examined to evaluate the effects of the substitutions. For strains N1550 and N1582, each of four spores formed colonies. For strains N1565 and N1596, colonies were formed from only 2 of 4 spores. Results from one representative tetrad are shown for each strain.

Figure 2. Cell proliferation and morphology of pik1-td cells at permissive and restrictive temperatures.

pik1-td cells (N1366) carried Δpik1::ura4 and pREP41X-Ub-R-DHFR^ts-pik1. The plasmid expressed a thermolabile dihydrofolate reductase-Pik1 fusion protein. A shift from 25°C to 36°C is known to cause unfolding of the Ub-R-DHFR^ts fusion protein followed by ubiquitin dependent proteolysis (Dohmen et al., 1994). (A) At time 0, cultures at 25°C were shifted to 25°C (▴) or 36°C (▪) with thiamine present throughout. Cell proliferation ceased after a modest increase in cell number at 36°C. (B) Bright field microscopy of unfixed cells. Cells were incubated at 25°C for 24 h or incubated at 25°C for 12 h and then shifted to 36°C for 18 h. (C) Cells fixed with formaldehyde after 18 h at 25°C or 36°C were stained for F-actin (with FITC-phalloidin), DNA (DAPI) or septum (calcofluor white). Bars, 2 µm.

Figure 3. Septum morphology in pik1-td cells.

Wild-type or pik1-td cells (N1366) were cultured for 18 h at 25°C or 36°C. Cells were fixed with potassium permanganate and processed for transmission electron microscopy. At 36°C, there was one or more septum per cell. The secondary septum were thickened relative to septum in pik1-td cells at 25°C or in wild-type cells at 25°C or 36°C. Intracellular membranous and vacuolar structures also accumulate in pik1-td cells at 36°C (magnified square).

Figure 4. Localization of 2XeGFP-Pik1 in cells in asynchronous cultures, or in cultures synchronized by cdc25-22 block and release.

(A) N1369 cells carried Δpik1::ura4 and pREP41-2XeGFP-pik1. The plasmid expressed 2 eGFP moieties, fused tandemly to the N-terminus of Pik1. Cells were cultured for 24 h at 30°C in the absence of thiamine. Punctate fluorescence was observed throughout the cytoplasm and around the periphery. A medial fluorescent band was observed in 8% of the cells. (B) Cells that carried a GFP-tagged allele of Gma12p (a Golgi-associated galactosyltransferase) were cultured for 24 hours at 30°C, fixed with methanol and processed for indirect immunofluorescence staining with rabbit antiserum against Pik1p and Texas Red conjugated, goat anti-rabbit antibodies. Cells were examined for Gma12p-GFP fusion (green) and Pik1p (red) and the images merged for colocalization (yellow). (C) N1401 cells which carried cdc25-22 and pREP41-2XeGFP-pik1were synchronized by block and release. Cells were accumulated at G2/M by incubation at 36°C for 4 h and released to 25°C (time = 0 minute). Aliquots were taken every 20 minutes for microscopic examination for binucleate cells (nuclear DNA visualized with DAPI, ♦), F-actin ring index (visualized with FITC-phalloidin, ▪), septation index (visualized with calcofluor white, ▴) and 2XeGFP-Pik1 fluorescence at the medial region (•). Representative images are shown for cells collected 80 and 120 minutes after release from the temperature block. Results are representative of two independent experiments. Bars, 2 µm.

Figure 5. The C-terminal region of Pik1 is conserved.

(A) Schematic representation of the primary structure of S. pombe Pik1 is shown. The kinase domain was identified from sequence similarity comparison to known lipid kinases using Prosite (grey shaded regions), or using Pfam (dark grey region only). A pseudo IQ motif is in the light grey shaded region. The positions of site-directed mutations in the kinase domain and in the pseudo IQ motif are indicated. (B) Sequence of the C-terminal region of S.pombe Pik1 (Sp; Accession CAA93903) is compared to those of orthologs in Saccharomyces cerevisiae (Sc; CAA53658), Candida albicans (Ca; CAA09718), Caenorhabditis elegans (Ce; NP_508177), Drosophila melanogaster (Dm; NP_728519), Xenopus laevis (Xl; Q6GN16), bovine (Bt; 002810) and human (Hs; BAA21661). The lipid kinase DRH motif and the C-terminal pseudo IQ motif (IQxxxRGxxxR) are indicated by black bands.

Figure 6. Pik1 R838, but not D709, is required for binding to Cdc4.

(A) Yeast two-hybrid colony formation assay. S. cerevisiae cells were transformed with a two-hybrid bait vector carrying the S. pombe cdc4 cDNA sequence fused to the GAL4 DNA binding domain (Cdc4-GAL4 DB). This strain was co-transformed with the two-hybrid prey vector alone (-ve), or with the prey vector carrying the S.pombe pik1 C-terminal wild-type sequence fused to the GAL4 transcription activation domain (Pik1^507–851-GAL4 TA; WT), or with the latter carrying either the D709A or R838A substitution. Colony formation is indicative of a protein-protein interaction. Results shown are representative of 3 separate experiments with the same strains. (B) X-gal colony filter assays. Nitrocellulose filters were overlayed onto the S. cerevisiae colonies grown on SD –Leu −Trp plates, submerged in liquid nitrogen and incubated in X-gal solution to monitor β-galacatosidase activity. Cells carrying the pik1 wild-type and D709A alleles fused to the GAL4-TA domain in the pBI771 vector and cdc4 sequence fused to the GAL4-BD in the pBI880 vector turned blue and were thus positive for the Pik1p-Cdc4p interaction. (C) Immunoblot assay. Accumulation of Pik1 upon ectopic expression of pik1 alleles from an episome under the control of the nmt1 promoter in cells carrying the intact chromosomal pik1 locus was assessed by immunoblot analysis. S.pombe cells transformed with a pREP1 plasmid carrying either the full length pik1 wild-type or mutated cDNA sequences were cultured in the absence of thiamine. Cell lysates (5 µg protein) were subjected to SDS-PAGE and Pik1 was visualized with a primary anti-Pik1 rabbit antiserum and a secondary goat anti-rabbit IgG-HRP (horseradish peroxidase)-conjugated antibody. Bar = 87 kDa molecular weight marker. Results shown are representative of 3 separate experiments with the same strains. The upper arrow represents the accumulation of the 97 kDa Pik1 wild-type and mutant proteins. The lower arrow represents an unknown polypeptide of 93 kDa, which is present in all samples including lysates of cells carrying the vector alone and lysates of cells cultured in repressed and derepressed conditions. (D) Sandwich ELISA to assess interaction between full length Pik1 and Cdc4. ELISA plates were coated with purified Cdc4, and then washed and blocked. Wells were subsequently incubated with 2-fold serial dilutions of lysates prepared from S.pombe cells grown in the absence of thiamine (solid line) and expressing the full-length pik1 cDNA sequence (open triangle), or pik1 cDNA sequences carrying mutations R838A (circle), D709A (diamond), or both R838A and D709A (cross hatch). Negative controls included lysates prepared from cells transformed with the pREP1 plasmid alone (dotted line, open square) or cultured under repressed conditions (with thiamine, dotted line, closed triangle). Results shown are representative of 3 separate experiments with the same strains.

Figure 7. Pik1 D709, but not R838, is required for lipid kinase activity.

(A) Incorporation of ³²P into PtdInsP as visualized by autoradiography after chromatographic separation of phospholipids. S. pombe cells were transformed with pREP1 plasmids carrying either full-length wild-type (WT) or mutant (D709A or R838A) pik1 cDNA coding regions under the control of the nmt1 promoter. Cells were cultured in the presence or absence of thiamine as indicated. Cell lysates (0.05 µg total protein) were pulse labeled with [γ- ³²P] ATP. Phospholipids were extracted and subjected to thin layer chromatography, followed by autoradiography. PtdInsP, monophosphorylated inositol phosphate. (B) Protein concentration dependency of lipid kinase activity in cell lysates. Lysates were prepared and assayed for lipid kinase activity as in (A). The lipid kinase activity in lysates prepared from cells carrying the empty pREP1 plasmid (WT, dotted lines), or the plasmid with the wild-type pik1 sequence (WT, solid lines), incubated with thiamine (closed triangle) or without thiamine (open triangle) is shown in the top panel. Lipid kinase activity of lysates of cells after ectopic expression of pik1^D709A, or pik1^R838A or pik1^{D709A, R838A} cultured under identical conditions are shown in the lower panels. Results shown are representative of 2 independent experiments using the same strains.

Figure 8. Effects of ectopic expression of pik1 wild-type and mutant alleles on cell proliferation and morphology.

(A) Colony formation assay. S. pombe cells transformed with the pREP1 plasmid with no insert (-ve), or with the pREP1 plasmid carrying pik1 (WT), pik1^D709A (D709A), pik1^R838A (R838A), or pik1^{D709A, R838A} (D709A, R838A) were cultured in liquid medium in the presence of thiamine to saturation. Cells were washed free of thiamine and serial 10-fold dilutions (about 10⁵ to 10¹ cells per spot, from left to right) of each strain were spotted onto EMM –leucine plates with phloxin B and with (left panel) or without (right panel) thiamine. Plates were kept at 30°C for 6 days. (B) Cell proliferation in liquid culture. Cell cultures in EMM –leucine were started at a cell density of 1×10⁵ cells/mL and incubated at 30°C in the presence (closed triangle) or absence (open triangle) of thiamine. The dotted lines in the top left panel represent cells transformed with the pREP1 plasmid without the pik1 cDNA sequence. Strain designations were as in (A). (C) Cell morphology. After 24 hours of growth in liquid culture (B), cells were fixed with formaldehyde and examined by epifluorescence microscopy after staining with calcofluor (to visualize the septum), DAPI (nucleus) or FITC-Phalloidin (FITC-Ph; F-actin). Bar = 5 µm. The results shown are for cell cultures in the absence of thiamine. Cell cultures in the presence of thiamine gave results similar to those observed with cells transformed with the vector alone (-ve). Results shown are representative of 2 (C) or 3 (A, B) separate experiments with the same cell strains. The relative frequency of binucleate cells and cells with F-actin rings or septa in the cell population are given in Table 3.

Figure 9. Effects of ectopic expression of pik1 alleles in cdc4^G107S cells.

(A) Immunoblot assay. Accumulation of Pik1 in cdc4^G107S cells upon ectopic expression of pik1 alleles from an episome under the control of the nmt1 promoter in cells carrying the intact chromosomal pik1 locus was assessed by immunoblot analysis as described for Figure 6C. Results shown are representative of 3 separate experiments with the same strains. Colony formation assays (B) and proliferation in liquid medium (C) of S. pombe cdc4^G107S cells transformed with the pREP1 plasmid with no insert (-ve), or with the pREP1 plasmid carrying pik1 (WT), pik1^D709A (D709A), pik1^R838A (R838A), or pik1^{D709A, R838A} (D709A, R838A) were carried out as described in Figure 8A and B. (D) Lipid kinase activity assays. Cells carrying either cdc4 or cdc4^G107S were transformed with a pREP1 plasmid with no insert (-ve), or with wild-type pik1^wt (WT) or pik1^R838A (R838A) cDNA sequences. Cells were incubated at 25°C for 24 hours in the absence of thiamine, and lysates prepared and assayed immediately for lipid kinase activity as described in Figure 7. Results shown are mean±S.E. of at least 3 separate experiments with the same strains.

Figure 10. Complementation of S.cerevisiae pik1-101 by heterologous expression of S.pombe pik1 alleles.

Colony formation assays. (A) S.cerevisiae PIK1 cells were transformed with empty YEplac181 as negative control (-ve), or with YEplac181 recombinants that expressed eGFP fusions to wild-type Pik1 under the control of an attenuated nmt1 promoter (P_nmt41 -eGFP-pik1), or the same construct carrying either the D709A or R838A substitutions. To assay for colony formation at 25°C or 37°C, aliquots from serial dilutions of each culture, containing the number of cells indicated, were prepared and spotted onto SD-Leu plates lacking thiamine which were incubated for 5 days. (B) The experiment as described in (A) was performed with S.cerevisiae pik1-101 cells. The colony formation assays were replicated independently at least three times. The results shown are representative of each of the replicates.