Septin-dependent assembly of a cell cycle-regulatory module in Saccharomyces cerevisiae

Abstract

Saccharomyces cerevisiae septin mutants have pleiotropic defects, which include the formation of abnormally elongated buds. This bud morphology results at least in part from a cell cycle delay imposed by the Cdc28p-inhibitory kinase Swe1p. Mutations in three other genes (GIN4, encoding a kinase related to the Schizosaccharomyces pombe mitotic inducer Nim1p; CLA4, encoding a p21-activated kinase; and NAP1, encoding a Clb2p-interacting protein) also produce perturbations of septin organization associated with an Swe1p-dependent cell cycle delay. The effects of gin4, cla4, and nap1 mutations are additive, indicating that these proteins promote normal septin organization through pathways that are at least partially independent. In contrast, mutations affecting the other two Nim1p-related kinases in S. cerevisiae, Hsl1p and Kcc4p, produce no detectable effect on septin organization. However, deletion of HSL1, but not of KCC4, did produce a cell cycle delay under some conditions; this delay appears to reflect a direct role of Hsl1p in the regulation of Swe1p. As shown previously, Swe1p plays a central role in the morphogenesis checkpoint that delays the cell cycle in response to defects in bud formation. Swe1p is localized to the nucleus and to the daughter side of the mother bud neck prior to its degradation in G(2)/M phase. Both the neck localization of Swe1p and its degradation require Hsl1p and its binding partner Hsl7p, both of which colocalize with Swe1p at the daughter side of the neck. This localization is lost in mutants with perturbed septin organization, suggesting that the release of Hsl1p and Hsl7p from the neck may reduce their ability to inactivate Swe1p and thus contribute to the G(2) delay observed in such mutants. In contrast, treatments that perturb actin organization have little effect on Hsl1p and Hsl7p localization, suggesting that such treatments must stabilize Swe1p by another mechanism. The apparent dependence of Swe1p degradation on localization of the Hsl1p-Hsl7p-Swe1p module to a site that exists only in budded cells may constitute a mechanism for deactivating the morphogenesis checkpoint when it is no longer needed (i.e., after a bud has formed).

FIG. 1

Role of Swe1p-dependent inhibition of Cdc28p in the elongated-bud morphology of septin mutants. Homozygous diploid strains M-905 (cdc11-6), M-1207 (cdc11-6 swe1Δ), and M-1208 (cdc11-6 mih1Δ) were grown to exponential phase at 23°C in YM-P medium and examined by DIC microscopy before (panels 1, 3, and 5) and after (panels 2, 4, and 6) a shift to 37°C for 6 h. Arrows indicate multibudded cells (panel 4) or elongated buds (panel 5), as discussed in the text.

FIG. 2

Role of Swe1p-dependent inhibition of Cdc28p in the elongated-bud morphology of gin4, cla4, and nap1 mutants. The indicated strains were grown to exponential phase in YM-P medium at 30°C and examined by DIC microscopy. (A) YEF473; (B) M-1077; (C) M-600; (D) M-272; (E) M-825; (F) M-829; (G) M-515; (H) M-522; (I) M-1025; (J) M-546; (K) M-544; (L) M-976.

FIG. 3

Effects of gin4, cla4, and nap1 mutations, but not of hsl1 or kcc4 mutations, on septin organization. Cells of the indicated strains were grown to exponential phase in YM-P medium at 30°C and examined by immunofluorescence microscopy to localize Cdc11p. (A) YEF473; (B) M-272; (C) M-515; (D) M-546; (E) M-522; (F) M-544; (G) M-601; (H) M-286. The arrow and arrowhead in panel D indicate cells displaying septin misorganization, classified as “fuzzy” and “bars,” respectively, as discussed in Table 3.

FIG. 4

Effects of gin4, hsl1, and kcc4 mutations on bud morphology. The indicated strains were grown to exponential phase or (for the hsl1 strain) to a high cell density in YM-P medium at 30°C and examined by DIC microscopy. (A) YEF473; (B) M-601; (C) M-1031; (D) M-272; (E) M-603; (F) M-604.

FIG. 5

Localization of Swe1p in wild-type, hsl1 mutant, and hsl7 mutant cells. The indicated strains were grown to exponential phase in YPD medium at 30°C and stained for Swe1p (left-hand panels) or DNA (right-hand panels) as described in Materials and Methods. (A) JMY1441; (B) JMY1477; (C) JMY1475; (D) JMY1479; (E) DLY1. Strains JMY1441, JMY1477, JMY1475, and JMY1479 all contain two integrated copies of SWE1myc, whereas strain DLY1 lacks SWE1myc as a negative control for the antibody sandwich protocol.

FIG. 6

Septin dependence of Hsl1p and Hsl7p localization. Cells of the indicated strains were grown to exponential phase in YM-P medium at 23°C and fixed either before (left-hand panels) or 30 min after (right-hand panels) a shift to 37°C. (A) Cells of wild-type strains JMY1441 (SWE1-myc), M-1427 (HSL1-13myc), and M-1423 (HSL7-3HA) were stained with antibodies to the tagged proteins. (B and C) Cells of cdc12-6 strains M-1552 (HSL1-13myc) (B) and M-1554 (HSL7-3HA) (C) were double stained with antibodies to Cdc11p and to the tagged protein.

FIG. 7

Localization of Hsl1p and Hsl7p in hsl7, hsl1, swe1, and mih1 mutant cells. Cells of the indicated strains were grown to exponential phase in YM-P medium at 30°C and examined by fluorescence microscopy to localize Hsl1p-GFP (A) or fixed and examined by immunofluorescence microscopy to detect Hsl7p-HA (B). (A) HSL1-GFP strains M-1272 (hsl7Δ), M-1426 (swe1Δ), and M-1158 (mih1Δ). (B) HSL7-3HA strains M-1445 (hsl1Δ), M-1440 (swe1Δ), and M-1421 (mih1Δ).

FIG. 8

Loss of Hsl1p and Hsl7p localization in gin4 and cla4 mutants. Strains M-1443 (gin4Δ HSL1-13myc) (A), M-1451 (cla4Δ HSL1-13myc) (B), M-1442 (gin4Δ HSL7-3HA) (C), and M-1454 (cla4Δ HSL7-3HA) (D) were grown to exponential phase in YM-P medium at 30°C and examined by immunofluorescence microscopy to localize Hsl1p-myc or Hsl7p-HA. Arrows indicate the minority of cells that show localization of the tagged proteins to the neck.

FIG. 9

Localization of Hsl1p, Hsl7p, and Swe1p following actin depolymerization. (A) HSL1-13myc (M-1427) and HSL7-3HA (M-1423) cells growing exponentially in YM-P medium at 30°C were treated with 100 μM latrunculin-A for 30 min or 3 h, as indicated, and Hsl1p-myc and Hsl7p-HA were localized by immunofluorescence. Equal exposure times were used for all four panels. (B) SWE1myc cells (JMY1441) were grown to exponential phase in YPD medium at 30°C and treated with 100 μM latrunculin-A. The proportions of preanaphase budded cells displaying Swe1p staining at the neck (●) or in the nucleus (○) were determined at the indicated times. (Cells with Swe1p at both locations were counted in both categories.) More than 200 cells were counted per sample.

FIG. 10

Model summarizing the results presented in this paper. See text for details.