ago1 and dcr1, two core components of the RNA interference pathway, functionally diverge from rdp1 in regulating cell cycle events in Schizosaccharomyces pombe

Abstract

In the fission yeast Schizosaccharomyces pombe, three genes that function in the RNA interference (RNAi) pathway, ago1+, dcr1+, and rdp1+, have recently been shown to be important for timely formation of heterochromatin and accurate chromosome segregation. In the present study, we present evidence that null mutants for ago1+ and dcr1+ but not rdp1+, exhibit abnormal cytokinesis, cell cycle arrest deficiencies, and mating defects. Subsequent analyses showed that ago1+ and dcr1+ are required for regulated hyperphosphorylation of Cdc2 when encountering genotoxic insults. Because rdp1+ is dispensable for this process, the functions of ago1+ and dcr1+ in this pathway are presumably independent of their roles in RNAi-mediated heterochromatin formation and chromosome segregation. This was further supported by the finding that ago1+ is a multicopy suppressor of the S-M checkpoint deficiency and cytokinesis defects associated with loss of Dcr1 function, but not for the chromosome segregation defects of this mutant. Accordingly, we conclude that Dcr1-dependent production of small interfering RNAs is not required for enactment and/or maintenance of certain cell cycle checkpoints and that Ago1 and Dcr1 functionally diverge from Rdp1 to control cell cycle events in fission yeast. Finally, exogenous expression of hGERp95/EIF2C2/hAgo2, a human Ago1 homolog implicated in posttranscriptional gene silencing, compensated for the loss of ago1+ function in S. pombe. This suggests that PPD proteins may also be important for regulation of cell cycle events in higher eukaryotes.

Figure 3.

ago1 null mutants fail to activate the S-M DNA replication checkpoint. (A) FY254 and Δago1 strains transformed with vector (pREP3X), pREP3X +ago1⁺, or pREP3X +hAgo2 were streaked onto leucine-deficient EMM plates containing 10 mM HU and incubated for 3 d at 30°C. Viability of the Δago1 strain containing pREP3X was confirmed by growth on EMM lacking both leucine and HU (our unpublished data). (B) Immunoblot analyses of total and phosphorylated Cdc2 (Cdc2-P) from FY254 and Δago1 strains in unsynchronized cultures (US), during HU treatment (+10 mM HU) and after removal of HU (Wash Out). Mean septation indices (determined from four independent experiments) are shown for each time point (bar graph).

Figure 1.

ago1⁺ and dcr1⁺ are both required for normal cytokinesis. All strains were cultured at 30°C in YE to mid-log phase (OD₅₉₅ = 0.6-0.9), stained with 4,6-diamidino-2-phenylindole and placed onto microscope slides coated with poly l-lysine (5 mg/ml). Samples were examined by fluorescence and differential interference contrast microscopy. Representative fields were photographed and scored for nuclei or septation. For each sample, at least 500 cells were scored. (A) Quantitation of binucleated cells in parental strain TV294, Δago1 (TV292), Δdcr1 (TV293), and Δrdp1 (TV296) strains. (B) Quantitation of septation indices of TV294, Δago1 (TV292), Δdcr1 (TV293), and Δrdp1 (TV296). (C) Quantitation of septation indices in FY254 and Δago1 (JC254K) strains transformed with vector (pREP3X), pREP3X +ago1⁺, or pREP3X +hAgo2.

Figure 2.

ago1 and dcr1 mutants are defective for G₁ arrest and mating after nitrogen limitation. (A) FACS analyses of log phase (+Nitrogen) and nitrogen starved (-Nitrogen) TV294, Δago1 (TV292), Δdcr1 (TV293), and Δrdp1 (TV296) cultures. Insets at the right show the morphologies of yeast cells under nitrogen-deficient conditions. (B and C) Mating mixtures were incubated at 30°C for 18-30 h on malt extract plates and scored for zygote production microscopically. (B) Mating frequencies of TV294, Δago1 (TV292), Δdcr1 (TV293), and Δrdp1 (TV296) crossed with FY261 (parental type) or JC261K (Δago1). The low level of self-mating in strains FY261 and JC261K presumably occurs due to mating type switching. (C) Mating frequencies for homothallic wild-type (h⁹⁰) or homothallic ago1 null (h⁹⁰Δago1) strains under nitrogen-deficient or low glucose/low nitrogen condition. (D) 4,6-diamidino-2-phenylindole-stained and corresponding differential interference contrast (DIC) images of TV294, Δago1, Δdcr1, and Δrdp1 strains cultured under normal growth (unstarved), nitrogen starvation (-Nit.) and low glucose/low nitrogen (LowGluc./LowNit.) conditions. Average cell lengths for each strain (unstarved condition) are shown at the bottoms of the DIC panels.

Figure 2.

ago1 and dcr1 mutants are defective for G₁ arrest and mating after nitrogen limitation. (A) FACS analyses of log phase (+Nitrogen) and nitrogen starved (-Nitrogen) TV294, Δago1 (TV292), Δdcr1 (TV293), and Δrdp1 (TV296) cultures. Insets at the right show the morphologies of yeast cells under nitrogen-deficient conditions. (B and C) Mating mixtures were incubated at 30°C for 18-30 h on malt extract plates and scored for zygote production microscopically. (B) Mating frequencies of TV294, Δago1 (TV292), Δdcr1 (TV293), and Δrdp1 (TV296) crossed with FY261 (parental type) or JC261K (Δago1). The low level of self-mating in strains FY261 and JC261K presumably occurs due to mating type switching. (C) Mating frequencies for homothallic wild-type (h⁹⁰) or homothallic ago1 null (h⁹⁰Δago1) strains under nitrogen-deficient or low glucose/low nitrogen condition. (D) 4,6-diamidino-2-phenylindole-stained and corresponding differential interference contrast (DIC) images of TV294, Δago1, Δdcr1, and Δrdp1 strains cultured under normal growth (unstarved), nitrogen starvation (-Nit.) and low glucose/low nitrogen (LowGluc./LowNit.) conditions. Average cell lengths for each strain (unstarved condition) are shown at the bottoms of the DIC panels.

Figure 4.

ago1⁺ and dcr1⁺ are both required for hyperphosphorylation of Cdc2 and enactment of the DNA damage and replication checkpoints. (A) Serial dilutions of TV294 and corresponding Δago1, Δdcr1, and Δrdp1 strains were spotted on YE agar or YE agar containing 3.5 mM HU and were cultured for 3 and 5 d at 30°C, respectively. (B) TV294, Δago1, Δdcr1, and Δrdp1 liquid cultures were grown at 30°C to an OD₅₉₅ = 0.5, and aliquots were removed before (-) or after (+) a 4-h HU treatment. Whole cell lysates were prepared and separated by SDS-PAGE before immunoblot analyses of total and phosphorylated Cdc2. The normalized levels of HU-induced Cdc2 phosphorylation are shown below the immunoblots in B and C. (C) Levels of total and phosphorylated Cdc2 were analyzed in strains transformed with vector alone or vector encoding Ago1 or Dcr1. (D) TV294, Δago1, Δdcr1 Δrdp1, and Δrad3 liquid cultures were grown at 30°C to an OD₅₉₅ = 0.8, spread at a density of 100-300 cells per plate and exposed to UV radiation (0, 75, 150, 225, or 300 J/M²). Cell survival is shown as a percentage of the nonirradiated control cell survival. (E) TV294, Δago1, Δdcr1, and Δrdp1 were subjected to mock (0 J/m²) treatment and UV irradiation (75 J/m²), and the septation index for each strain was determined every hour postirradiation (n = 3, 300 cells counted/time point). Similar results were obtained when these strains were subjected to a 100 J/m² dose (our unpublished data).

Figure 5.

ago1⁺ functions downstream of dcr1⁺ to regulate cell cycle events. Yeast strains were transformed with plasmids encoding Ago1, HA-Dcr1, or vector alone. (A) Yeast strains grown in EMM-leu to log phase were examined by phase contrast microscopy, and septated cells were counted and expressed as a percentage of total cells. (B) Yeast strains were streaked onto EMM lacking leucine (EMM) in the presence or absence of 7.5 mM HU (+HU) and were then cultured at 30°C for 3 to 5 d.

Figure 6.

Ectopic overexpression of Ago1 does not correct the chromosome segregation defect of dcr1 nulls. Yeast strains were transformed with plasmids encoding Ago1, HA-Dcr1 or vector alone. Strains were streaked onto EMM lacking leucine in the presence (+TBZ) or absence (EMM) of 10 μg/ml TBZ and were then cultured at 30°C for 3 to 5 d.

Figure 7.

Model for RNAi-independent functions of Ago1 and Dcr1. The right side of the model depicts the classical RNAi pathway in which Rdp1 is required for production/amplification of centromere-derived double-strand transcripts (cenH) that serve as substrates for Dcr1. The resulting siRNAs are incorporated into Ago1-containing RISCs that facilitate heterochromatin formation, an event that is required for attachment of mitotic spindles to kinetochores and subsequent orderly chromosome segregation. The left side of the model depicts a pathway in which Dcr1 and Ago1 in response to genotoxic insults such as DNA damage or HU, mediate downstream events that lead to inhibitory phosphorylation of Cdc2. This process is proposed to be independent of siRNAs and Rdp1.