Cbk1p, a protein similar to the human myotonic dystrophy kinase, is essential for normal morphogenesis in Saccharomyces cerevisiae

Abstract

We have studied the CBK1 gene of Saccharomyces cerevisiae, which encodes a conserved protein kinase similar to the human myotonic dystrophy kinase. We have shown that the subcellular localization of the protein, Cbk1p, varies in a cell cycle-dependent manner. Three phenotypes are associated with the inactivation of the CBK1 gene: large aggregates of cells, round rather than ellipsoidal cells and a change from a bipolar to a random budding pattern. Two-hybrid and extragenic suppressor studies have linked Cbk1p with the transcription factor Ace2p, which is responsible for the transcription of chitinase. Cbk1p is necessary for the activation of Ace2p and we have shown that the aggregation phenotype is due to a lack of chitinase expression. The random budding pattern and the round cell phenotype of the CBK1 deletion strain show that in addition to its role in regulating chitinase expression via Ace2p, Cbk1p is essential for a wild-type morphological development of the cell.

Fig. 1. Inactivation phenotypes of CBK1 and ACE2. Fresh overnight cultures of wild-type, Δcbk1 and Δace2 haploid strains in rich complete medium were stained with Calcofluor white (see Materials and methods). Inactivation of CBK1 and ACE2 leads to formation of large aggregates, which cannot be separated using a micro-manipulator. (A–C) The wild-type strain AW303-2b; (D–F) the Δcbk1 strain WR03-5d; (G–I) the Δace2 strain WR27-1a. (A), (D) and (G) Normaski, (B), (C), (E), (F), (H) and (I) Calcofluor white staining.

Fig. 2. Schematic representation of Ace2p and the dominant suppressor mutations. (A) Schematic representation of Ace2p showing the position of the Zn finger DNA-binding domain, the minimal two-hybrid domain and the dominant suppressor mutations. (B) Sequence of the dominant suppressor mutations.

Fig. 3. Phenotype of the Δcbk1 suppressor strains. Fresh overnight cultures in rich complete medium of haploid and diploid wild-type (AW303-2b; AW303), Δcbk1 recessive suppressor (5d1; WR175) and Δcbk1 dominant suppressor (5d2; WR176) strains were observed by Nomarski interference microscopy. The suppressor strains no longer form large aggregates of cells, but are still round rather then ellipsoidal. This is most clearly seen in the diploid strains.

Fig. 4. Expression of the CTS gene. (A) Total RNA was extracted from wild-type (AW303-2b), Δcbk1 (WR03-5d), Δace2 (WR27-1a) Δcbk1 dominant suppressor (5d2) and Δcbk1 recessive suppressor (5d1) strains. Aliquots of 10 µg were used in a northern analysis, probed with CTS1, the ACT1 gene was used as an internal standard. CTS1 transcription is almost undetectable in the Δcbk1 and Δace2 strains but is re-established in the suppressor strains. (B) Haploid strains with a deleted CBK1 gene derived from W303 (WR05-5a) and S288c (WR208-1a) were transformed by a plasmid where the expression of the CTS1 gene is under the control of the GAL promotor (YEpWJR064). Transient expression of chitinase on galactose medium causes the large aggregates associated with the deletion of the CBK1 gene to separate.

Fig. 5. Cell cycle-dependent expression of ACE2 and CTS1. MATa haploid cultures of wild-type (AW303-2a) and Δcbk1 dominant suppressor (5a3) were synchronized with α factor, samples were taken at 10 min intervals and stopped by adding 2 vols of ethanol at –60°C. Total RNA was extracted and used in a northern analysis probed with ACE2 and CTS1, the ACT1 gene was used as an internal standard. No difference in the cell cycle-coordinated expression of ACE2 and CTS1 could be detected between the wild type and Δcbk1 dominant suppressor.

Fig. 6. Budding pattern of wild-type and Δcbk1 recessive and dominant suppressors. Old cultures of haploid and diploid: wild-type [AW303-2b (A); AW303 (D)], Δcbk1 recessive suppressor [5d1 (B); WR175 (E)] and Δcbk1 dominant suppressor [5d2 (C); WR17 (F)] all derived from W303 and the wild-type haploid YPH499 derived from S288c (G) were stained with Calcofluor white to reveal the bud scars and observed by fluorescence microscopy. The W303-derived haploid and diploid wild-type cells show a bipolar budding pattern, whereas the S288c-derived wild-type haploid shows an axial budding pattern. All suppressor strains show a random budding pattern. The remaining panels show cells liberated from aggregates of haploid strains with a deleted CBK1 gene derived from S288c [WR208-1a (H)] and W303 [WR05-5a (I)] during transient expression of chitinase. In the S288c-derived cells the axial budding pattern is maintained, while in the W303-derived cells the budding pattern is random.

Fig. 7. Subcellular localization of CBK1p. Fresh overnight cultures of control diploid cells (AW303) and diploid and haploid cells (WR154; WR152-1a) expressing the Cbk1p–GFP fusion protein were grown in complete medium supplemented with a 4-fold excess of adenine and observed by phase contrast and fluorescence microscopy. No GFP fluorescence could be detected in the control cells. In both diploid and haploid cells some cytoplasmic staining can be seen and specific staining of different parts of the cell depending on their position in the cell cycle. The staining is brighter in the diploid strain.

Fig. 8. Cell cycle-dependent localization of Cbk1p. Representative individual haploid cells expressing the Cbk1p–GFP fusion, taken from the experiment presented in Figure 7 have been selected to show the localization of Cbk1p at different stages of the cell cycle. During G₁ the protein is seen in a punctate pattern around plasma membrane as well as in the cytoplasm. The protein concentrates at the emerging bud site, then at the tip of the bud, dispersing around the bud as it grows. In cells with a large bud the majority of the protein forms a ring around the bud neck and just before cell separation, a plate or ring is present in both the mother and daughter cell, either side of the cell junction.