Deficient sumoylation of yeast 2-micron plasmid proteins Rep1 and Rep2 associated with their loss from the plasmid-partitioning locus and impaired plasmid inheritance

Abstract

The 2-micron plasmid of the budding yeast Saccharomyces cerevisiae encodes copy-number amplification and partitioning systems that enable the plasmid to persist despite conferring no advantage to its host. Plasmid partitioning requires interaction of the plasmid Rep1 and Rep2 proteins with each other and with the plasmid-partitioning locus STB. Here we demonstrate that Rep1 stability is reduced in the absence of Rep2, and that both Rep proteins are sumoylated. Lysine-to-arginine substitutions in Rep1 and Rep2 that inhibited their sumoylation perturbed plasmid inheritance without affecting Rep protein stability or two-hybrid interaction between Rep1 and Rep2. One-hybrid and chromatin immunoprecipitation assays revealed that Rep1 was required for efficient retention of Rep2 at STB and that sumoylation-deficient mutants of Rep1 and Rep2 were impaired for association with STB. The normal co-localization of both Rep proteins with the punctate nuclear plasmid foci was also lost when Rep1 was sumoylation-deficient. The correlation of Rep protein sumoylation status with plasmid-partitioning locus association suggests a theme common to eukaryotic chromosome segregation proteins, sumoylated forms of which are found enriched at centromeres, and between the yeast 2-micron plasmid and viral episomes that depend on sumoylation of their maintenance proteins for persistence in their hosts.

Figure 1. Lysine mutations in Rep1 and Rep2 impair two-hybrid interaction with SUMO but do not affect Rep1-Rep2 interaction.

A cir⁰ two-hybrid reporter yeast strain was co-transformed with plasmids expressing the indicated Gal4_AD and LexA fusion proteins. Co-transformants were grown for 24 h on a nitrocellulose membrane and interaction of the fusion proteins was assessed by monitoring expression of the lacZ reporter gene using a β-galactosidase filter assay with the substrate X-gal, which produces a blue precipitate upon cleavage. Vector with no insert is indicated (--). Assays for interaction with conjugation-competent SUMO and conjugation-defective SUMOΔGG are shown on left and those for interaction between Rep1 and Rep2 on right.

Figure 2. Lysine-to-arginine mutations significantly reduce levels of SUMO-conjugated Rep1 and Rep2.

Yeast lacking the native 2 µm plasmid were co-transformed with a plasmid expressing untagged SUMO, HA-epitope-tagged SUMO (HA-SUMO), or no protein (-) from a galactose-inducible promoter, and an ADE2-tagged 2 µm plasmid encoding (A) Ubc9-tagged Rep2 and His₆-tagged Rep1 (wild-type (WT) (lanes 1, 3 and 5), or Rep1_3R (lanes 2, 4 and 6)) or (B) Ubc9-tagged Rep1 and His₆-tagged Rep2 (wild type (WT) (lanes 1, 3, and 5) or Rep2_13R (lanes 2, 4, and 6)). Yeast were also transformed solely with the plasmid expressing HA-epitope-tagged SUMO (A and B, lane 7). Yeast were cultured in medium containing galactose for 20 h to induce expression of SUMO proteins. Protein was extracted and His₆-tagged proteins were affinity purified with Co²⁺ resin and analyzed by western blotting with anti-Rep1, anti-Rep2, or anti-HA antibodies. Species consistent with SUMO-conjugated (asterisks) and HA-SUMO-conjugated (open circles) forms of Rep1 and Rep2, and an unknown Rep1 species (arrowhead) are indicated. Sizes of molecular weight standards resolved in the same gels are indicated.

Figure 3. Mutations that inhibit sumoylation of Rep1 and Rep2 also impair plasmid maintenance.

Yeast lacking the native 2 µm plasmid were transformed with KanMX6-tagged 2 µm plasmids encoding the indicated Rep1 and Rep2 alleles, grown in medium containing G418 and the proportion of plasmid-containing (G418-resistant) cells (mean ± SEM from six independent yeast transformants) was determined.

Figure 4. Effects of lysine-to-arginine mutations in Rep1 and Rep2 on post-translational stability.

Rep1 and Rep2 were (A) co-expressed or (B) expressed in the absence of one another and protein was extracted from yeast at the indicated time points following addition of cycloheximide. A closed circle indicates a phosphorylated species of Rep2 (unpublished data). (C) Yeast with the indicated alleles of REP1 and REP2 integrated in the genome were cultured and protein extracted. Rep protein levels were examined by western blot analysis.

Figure 5. Lysine mutations in Rep1_3R and Rep2_13R impair Rep-STB association but do not reduce levels of Rep fusion proteins.

(A) Five-fold serial dilutions of a cir ⁺ or cir ⁰ yeast one-hybrid reporter strain encoding STB upstream of a HIS3 reporter gene and transformed with plasmids encoding the indicated Rep1 and Rep2 proteins fused to B42_AD-HA were spotted onto galactose media to induce expression of fusion proteins. Recruitment of the Rep proteins to STB was monitored by growth on medium lacking histidine supplemented with 5 mM 3-aminotriazole. (B) Levels of the B42_AD-HA fusion proteins were monitored by western blot analysis of total protein extracted from the yeast transformants 24 h after galactose induction. (C) ChIP assays were performed with anti-Rep1, anti-Rep2, or anti-FLAG antibodies and the precipitated DNA amplified using primers specific for STB. ChIP efficiency is indicated by the percent of input DNA immunoprecipitated (avg ± sd from triplicate assays) (left) and ethidium-stained agarose gels of PCR products from a representative assay are shown (right). Template DNA amplified in “input” PCR reactions represented 40% of the DNA that was immunoprecipitated and used as template in “ChIP” PCR reactions.

Figure 6. Stable association of Rep2 with STB depends on Rep1.

(A) A cir ⁰ yeast one-hybrid reporter strain was co-transformed with single-copy plasmids allowing for galactose-inducible expression of untagged and B42_AD-HA-tagged Rep1 or Rep2. Expression of the STB-driven HIS3 reporter gene was monitored by growth on solid galactose media as described in the legend to Fig. 5. (B) ChIP assays were performed on extracts from yeast expressing Rep1, Rep2, or both Rep1 and Rep2 from an ADE2-tagged 2 µm plasmid as detailed in the legend to Fig. 5.

Figure 7. Localization of Rep1_3R and Rep2_13R.

(A) Spheroplasts were prepared from cir ⁰ yeast expressing wild-type or mutant Rep1 and Rep2 from an ADE2-tagged 2 µm plasmid and the Rep proteins were visualized by indirect immunofluorescence. Bulk chromatin was visualized by DAPI staining and cells by light microscopy (DIC). (B) Yeast were co-transformed with an ADE2-tagged 2 µm plasmid encoding the indicated Rep1 and Rep2 alleles and a plasmid containing the STB locus and 256 lacO repeats. Plasmid localization was visualized by fluorescence microscopy following induction of GFP-LacI repressor fusion protein expression.

Figure 8. Rep1_I202T does not associate with the plasmid STB locus but retains interaction with SUMO.

(A) Association of wild type and I202T mutant Rep1 (Rep1_I202T) with STB in cir⁰ yeast was monitored using a one-hybrid assay as described in the legend to Figure 5. (B) Interaction of Rep1 and Rep1_I202T with SUMO and with Rep2 was monitored in cir⁰ yeast using a two-hybrid assay as described in the legend to Figure 1. Levels of the Rep1 and Rep1_I202T fusion proteins in the (C) one-hybrid and (D) two-hybrid reporter yeast strains, in A and B respectively, were monitored by western blotting analysis with antibodies specific for Pgk1 (C and D, bottom), anti-HA (C, top) and anti-LexA (D, top).