A novel role of the N terminus of budding yeast histone H3 variant Cse4 in ubiquitin-mediated proteolysis

Abstract

Regulating levels of centromeric histone H3 (CenH3) variant is crucial for genome stability. Interaction of Psh1, an E3 ligase, with the C terminus of Cse4 has been shown to contribute to its proteolysis. Here, we demonstrate a role for ubiquitination of the N terminus of Cse4 in regulating Cse4 proteolysis for faithful chromosome segregation and a role for Doa1 in ubiquitination of Cse4.

Keywords: Cse4; Doa1; Psh1; chromosome segregation; ubiquitin-mediated protein degradation.

Figure 1

N-terminal tail of Cse4 regulates protein stability and faithful chromosome segregation. (A) The N terminus of Cse4 regulates its protein stability. We constructed plasmids expressing CSE4 with lysines (K) in the N terminus (1–135 amino acids) or the C terminus (136–229 amino acids) mutated to arginine (R) (shaded diagonals and asterisks denote the position of K) (see Table S1). KC denotes a fusion between the mutant N-terminal tail with all Ks changed to Rs and the wild-type C terminus of Cse4, whereas CK has the reverse configuration. Essential N-terminal domain (END) and CATD are as shown. (B) Western blot analysis of proteins prepared from a wild-type (YPH1018) strain with Cse4 (pSB816), 16KR (pSB817), KC (pMB1257), or CK (pMB1301) grown in raffinose/galactose (2%) for 6 hr followed by growth in glucose (2%)-containing media with cycloheximide (CHX, 10 µg/ml) for the indicated time, as previously described (Au et al. 2008). Blots were probed with anti-Myc (A-14, Santa Cruz, Inc.) and anti-Tub2 for a loading control. Tub2-normalized Cse4 protein levels after 120 min of CHX treatment is presented as percentage of remaining relative to the sample (t = 0; lane 1 = 100%). Protein half-lives were determined by least squares regression of log (percentage remaining) vs. time and reported as the mean ± SD of three biological replicates. (C) The graph shows kinetics of turnover from B plotted on a semilog scale. (D) The N terminus of Cse4 regulates the stability of histone H3. We constructed plasmids expressing fusions of histone H3 (HHT1) with CSE4 or 16KR. Shaded diagonals correspond to that in A; dotted shading represents regions homologous to HHT1. (E) Western blot analysis of proteins prepared from a wild-type (YPH1018) strain with CH (pMB1263), KH (pMB1273), full-length H3 (pMB1581), or the H3 histone fold domain (H3 HFD) (pMB1582) as described in B. (F) The graph shows kinetics of turnover from E plotted on a semilog scale. (G) Mutations in the N terminus of Cse4 lead to increased chromosome loss. Strains expressing constructs from A were assayed for the loss of a reporter chromosome that results in half-red/half-white colonies due to reporter chromosome loss in the first division as described previously (Hieter et al. 1985). Mean of three biological repeats with standard deviation is shown.

Figure 2

Role of Doa1 in proteolysis of Cse4 in a Psh1-independent manner. (A) The Cse4 N-terminus-mediated proteolysis is not abolished in psh1Δ strains. Western blot analysis of proteins prepared from psh1Δ strains (BY4741, Open Biosystems) with Cse4 (pSB816), KC (pMB1257) and CK (pMB1301), and KH (pMB1273) and CH (pMB1263) grown in raffinose/galactose (2%) for 6 hr followed by treatment with cycloheximide (CHX, 10 µg/ml) for the indicated times in glucose (2%)-containing media as described (Au et al. 2008). Blots were probed with anti-Myc and anti-Tub2 for a loading control. Tub2-normalized Cse4 protein level after 120 min of CHX treatment is presented as the percentage remaining relative to the sample (t = 0; lane 1 = 100%). The mean half-life (t_1/2) of two biological repeats is shown and is derived as in Figure 1B. (B) The graph shows kinetics of turnover from wild-type and psh1Δ strains plotted on a semilog scale. (C) Synthetic dosage lethality of GALCSE4 in doa1Δ and psh1Δ strains. Growth of serial dilutions of wild-type (WT) and doa1Δ and psh1Δ strains with vector or pMB1458 (GALCSE4) on glucose (2%)- or galactose (2%)-containing media at 30° for 4 days. (D) Cse4 and its chimeras are highly stable in doa1Δ strains. Protein stability assays are as described in A. The half-life (t_1/2) represents the mean of two biological repeats and is derived as in Figure 1B. (E) The graph shows kinetics of turnover plotted on a semilog scale. (F) Cse4 expressed from its own promoter is highly stable in doa1Δ strains. Western blot analysis of proteins prepared from wild-type (WT, YMB6398)- and doa1Δ (YMB8582)-expressing Myc-Cse4 integrated at the LEU2 locus as described in A except strains were grown in glucose (2%)-containing media. Half-life (t_1/2) represents the mean of three biological repeats with standard deviation. (G) Graph shows kinetics of turnover from F plotted on a semi log scale.

Figure 3

DOA1 contributes to the ubiquitination of the N-terminal tail of Cse4. (A) Overexpression of UBI4 suppresses the synthetic lethality of doa1Δ strains overexpressing CSE4. Growth of serial dilutions of wild-type (WT), doa1Δ, and psh1Δ strains with 2µ UBI4 (kindly provided by Charlie Boone), GALCSE4 (pMB1597), or both plasmids were assayed on glucose- or galactose-containing media at 30° for 4 days. (B) Overexpression of UBI4 decreases the stability of Cse4 in doa1Δ strains. Protein stability assays using strains from A grown in raffinose/galactose (2%)-containing media for 2 hr, followed by treatment with cycloheximide (CHX, 10 µg/ml) for indicated times in glucose (2%)-containing media as described (Au et al. 2008) using anti-HA (12CA5, Roche Inc.). The half-life (t_1/2) represents the mean of two biological repeats is shown and is derived as in Figure 1B. The graph shows kinetics of turnover plotted on a semilog scale. (C) The N terminus of Cse4 is ubiquitinated in a DOA1-dependent pathway. Wild-type and doa1Δ strains expressing CSE4 or cse4 alleles were used for pull-down with control (Ub⁻) or tandem ubiquitin-binding entities, Agarose-TUBE1 (Ub⁺) (LifeSensors, Inc.; catalog nos. UM401 and UM400, respectively) followed by Western blot analysis with anti-Myc antibody (A-14, Santa Cruz Inc.) as described (File S1). Control agarose and non-HA-tagged (vector) strains were used as negative controls. Western blot analysis from doa1Δ strains was exposed for a longer time to rule out any residual laddering (lanes 20–23). (D) The N terminus of Cse4 mediates its ubiquitination. Wild-type (WT, YPH1015) and doa1Δ strains expressing HA-tagged Cse4 (pMB1458) or HA-tagged Δ129 (pMB1459) were assayed as described in C. CHX (10 µg/ml) treatment was carried out for 30 min (lanes 6 and 12). (E) doa1Δ strains show defects in ubiquitination of Cse4. Ratio of ubiquitinated HA-Cse4 (white box labeled “A” in panel D) to non-ubiquitinated HA-Cse4 (white box labeled “B” in panel D) in wild-type (WT) and doa1Δ strains from D were calculated. The graph represents the mean of three biological repeats with standard deviation.