The anaphase promoting complex regulates yeast lifespan and rDNA stability by targeting Fob1 for degradation

Abstract

Genomic stability, stress response, and nutrient signaling all play critical, evolutionarily conserved roles in lifespan determination. However, the molecular mechanisms coordinating these processes with longevity remain unresolved. Here we investigate the involvement of the yeast anaphase promoting complex (APC) in longevity. The APC governs passage through M and G1 via ubiquitin-dependent targeting of substrate proteins and is associated with cancer and premature aging when defective. Our two-hybrid screen utilizing Apc5 as bait recovered the lifespan determinant Fob1 as prey. Fob1 is unstable specifically in G1, cycles throughout the cell cycle in a manner similar to Clb2 (an APC target), and is stabilized in APC (apc5(CA)) and proteasome (rpn10) mutants. Deletion of FOB1 increased replicative lifespan (RLS) in wild type (WT), apc5(CA), and apc10 cells, and suppressed apc5(CA) cell cycle progression and rDNA recombination defects. Alternatively, increased FOB1 expression decreased RLS in WT cells, but did not reduce the already short apc5(CA) RLS, suggesting an epistatic interaction between apc5(CA) and fob1. Mutation to a putative L-Box (Fob1(E420V)), a Destruction Box-like motif, abolished Fob1 modifications, stabilized the protein, and increased rDNA recombination. Our work provides a mechanistic role played by the APC to promote replicative longevity and genomic stability in yeast.

Keywords: Fob1; Saccharomyces cerevisiae; anaphase promoting complex; replicative lifespan; two-hybrid screen.

Figure 1

Deletion of FOB1 in APC mutants has little impact on growth, but a positive influence of APC mutant RLS. (A) WT, apc5^CA, fob1Δ, and apc5^CA fob1Δ cells were grown to an OD₆₀₀ of 0.5, then 10-fold serially diluted and spotted onto YPD plates. The plates were incubated at 30° and 37° for 3 days. (B) WT, apc10Δ, fob1Δ, and apc10Δ fob1Δ cells were treated as above. (C) RLS for WT, apc5^CA, fob1Δ, and apc5^CA fob1Δ cells was determined. The numbers of mother cells used for RLS determination were 39, 37, 36, and 42, respectively. (D) RLS for WT, apc10Δ, fob1Δ, and apc10Δ fob1Δ cells was determined. The numbers of mother cells used for RLS determination were 39, 34, 36, and 34, respectively. The mean RLS for panels C and D are shown in parentheses. Mean RLS was determined using the formula =FORECAST in Excel, as described in Materials and Methods. Statistical significance of differences for each RLS relative to WT was determined using the nonparametric Mann–Whitney U-test. Differences are considered significant, as each P-value is <0.05.

Figure 2

Increased FOB1 expression reduces RLS while overexpression is toxic. (A) RLS determination for WT and apc5^CA cells harboring 2μ-FOB1 or YCp50 empty vector. The number of mother cells used for WT + YCp50, WT + 2μ-FOB1, apc5^CA + YCp50, and apc5^CA + 2μ-FOB1 were 31, 32, 32, and 31, respectively. Mean RLS is shown in parentheses. Differences from WT are statistically significant with P-values <0.05. (B) WT and apc5^CA cells harboring YCp50 empty vector, CEN-APC5, 2μ-FOB1 or GAL_pro-FOB1-HA were serially diluted and spotted onto 2% glucose or 2% galactose-supplemented SD media. Plates were incubated at 30°, 3 days for glucose and 5 days for galactose plates. (C) WT and apc5^CA cells expressing GAL_pro-FOB1-HA were grown to early log phase, then induced with 2% galactose. Samples were taken every hour during the induction period and analyzed by Western blotting with antibodies against HA. The Ponceau S stained blot is shown for load control. (D) The cells used above were released into fresh 2% glucose media after the 6-hr galactose induction and allowed to reenter the cell cycle. Every hour for 6 hr, samples were removed for flow cytometry and Western analyses. NS, nonspecific band in WT cells.

Figure 3

Deletion of FOB1 alleviates cell cycle and rDNA recombination defects in apc5^CA cells. (A) Flow cytometry studies of steady state WT, fob1Δ, apc5^CA, and apc5^CA fob1Δ cells grown in YPD at 30°. (B) Western analyses of the cells used above with antibodies against Clb2 and GAPDH as a loading control. (C) Schematic of the CAN1-ADE2 construct cloned into one of the 25S rDNA repeats and the corresponding recombinant WT form. The ratio between red colony forming units (CFUs) and total CFUs is a comparative measure of the frequency of recombination events within the rDNA locus. The strain was a generous gift from K. Runge. (D) The average frequency of recombination in the rDNA for five individual colonies of WT, fob1Δ, apc5^CA, and apc5^CA fob1Δ, as judged by the loss of the ADE2 gene. Recombination frequency is expressed as the number of recombinant red CFUs per 10⁴ cells. Standard error of the mean is shown. (E) The average frequency of plasmid loss for five individual colonies of WT, fob1Δ, apc5^CA, and apc5^CA fob1Δ, as judged by the loss of the CEN-ADE2 plasmid. The frequency is expressed as the number of red CFUs per 10⁴ cells. Standard error of the mean is shown.

Figure 4

Fob1 oscillates throughout the cell cycle, is predominantly unstable during G1, and requires APC^Cdh1 for turnover. (A) WT cells expressing endogenously tagged FOB1-TAP were arrested with α-factor, then washed and released into fresh media lacking α-factor. Samples were collected for flow cytometry and Western analyses every 20 min for 100 min. Westerns were performed using antibodies against TAP, Clb2, and GAPDH as a loading control. (B) WT cells expressing endogenous Fob1-TAP were arrested in G1 with α-factor, S phase with hydroxyurea, or mitosis using nocodazole. The cells were washed and resuspended in fresh media containing CHX to prevent further protein synthesis. Samples were removed every 30 min and analyzed by Western blotting with antibodies against TAP and GAPDH. (C) apc5^CA FOB1-TAP cells were arrested in G1 using α-factor. Flow cytometry was used to confirm arrest. Following arrest, the cells were resuspended in fresh media containing CHX. Samples were removed every 30 min for TAP and Clb2 Westerns. NS, nonspecific band. (D) cdh1Δ and cdc20-1 cells expressing GAL_pro-FOB1-HA were arrested in G1 with α-factor, followed by induction of the FOB1 construct with galactose. Cells were then washed and released into fresh media containing 2% glucose and CHX, with samples collected every 20 min for protein stability studies. Samples were analyzed using Westerns with antibodies against HA and GAPDH as a loading control.

Figure 5

Apc5 interacts with Fob1 in a two-hybrid assay. (A) Schematic of the FOB1 fragments cloned into pGAD⋅C2 to identify a Fob1 binding motif responsible for Apc5 interactions. +++ denotes best growth, + denotes weak growth, and − indicates no growth in a yeast two-hybrid assay. The hatched area denotes a putative Apc5 binding domain. (B) Growth analysis of FOB1 derivatives and full-length APC5 in a two-hybrid assay. Top panels show the FOB1 derivatives (cloned into pGAD⋅C2) paired with pGBT9 empty vector. The bottom panel shows the FOB1 derivatives paired with full-length APC5 cloned into pGBT9. The cells were grown on SD −trp −leu to select for plasmids and on SD −trp −leu −ade to select for interactions. The positive control plasmids encode the SV40 large T-antigen (pTD1) and p53 (pVD3).

Figure 6

Fob1 modification requires the L box, a D box-like motif. (A) Creation of point mutations within FOB1 that alter the putative APC interaction motifs: the L box, E₄₂₀V; DB1, R₃₀₉A, L₃₁₂A; and DB2, R₅₄₅A, L₅₄₈A. (B) Modified Fob1 bands (Fob1^mod) accumulate in rpn10Δ cells defective in proteosome polyubiquitin binding. WT and rpn10Δ cells expressing GAL_pro-FOB1-HA grown in glucose or galactose, at 30° or 37°C, to induce expression for 6 hr. (C) GAL_pro-FOB1-HA mutant plasmid constructs DB1, DB2, DB1DB2, and E₄₂₀V in WT or rpn10Δ cells grown at 30° in glucose or galactose for 6 hr. Protein samples were analyzed by Westerns with antibodies against HA and GAPDH as a loading control.

Figure 7

Fob1 is an unstable protein that requires the APC and the L box for turnover. (A) WT and apc5^CA cells expressing either GAL_pro-FOB1-HA or GAL_pro-FOB1^E420V-HA were arrested in G1 with α-factor, followed by induction of the FOB1 constructs with galactose. Cells were then washed and released into fresh media containing CHX, with samples collected periodically for assessment of protein stability. Samples were analyzed by Westerns with antibodies against HA and GAPDH as a loading control. Films were scanned for analysis with ImageJ. (B) Band intensities from A were normalized to their corresponding GAPDH signal and then compared to band intensities after the 4-hr galactose induction. (C) The band intensity of WT Fob1 in WT cells after the 4-hr induction was set to 1. The band intensities from each sample after the 4-hr induction were compared to WT Fob1 in WT cells and plotted.

Figure 8

Fob1^E420V increases rDNA recombination frequency. The average frequency of rDNA recombination for five individual colonies of WT, fob1^E420V, and fob1∆, as judged by the loss of the ADE2 gene. Recombination frequency is expressed as the number of recombinant red colony forming units (CFUs) per 10⁴ cells. Standard error of the mean is shown.

Figure 9

Model depicting how APC targeted degradation of Fob1 may regulate genomic stability and longevity. The APC impinges on genomic stability through a variety of mechanisms. Fob1 promotes genomic instability by increasing recruitment of cohesin to the rDNA, which when not removed, can result in incomplete sister chromatid segregation during mitosis, by stalling DNA replication machinery during S phase. Both of these mechanisms can result in DSBs and DNA recombination in this already highly unstable rDNA region.