Novel E3 ubiquitin ligases that regulate histone protein levels in the budding yeast Saccharomyces cerevisiae

Abstract

Core histone proteins are essential for packaging the genomic DNA into chromatin in all eukaryotes. Since multiple genes encode these histone proteins, there is potential for generating more histones than what is required for chromatin assembly. The positively charged histones have a very high affinity for negatively charged molecules such as DNA, and any excess of histone proteins results in deleterious effects on genomic stability and cell viability. Hence, histone levels are known to be tightly regulated via transcriptional, posttranscriptional and posttranslational mechanisms. We have previously elucidated the posttranslational regulation of histone protein levels by the ubiquitin-proteasome pathway involving the E2 ubiquitin conjugating enzymes Ubc4/5 and the HECT (Homologous to E6-AP C-Terminus) domain containing E3 ligase Tom1 in the budding yeast. Here we report the identification of four additional E3 ligases containing the RING (Really Interesting New Gene) finger domains that are involved in the ubiquitylation and subsequent degradation of excess histones in yeast. These E3 ligases are Pep5, Snt2 as well as two previously uncharacterized Open Reading Frames (ORFs) YKR017C and YDR266C that we have named Hel1 and Hel2 (for Histone E3 Ligases) respectively. Mutants lacking these E3 ligases are sensitive to histone overexpression as they fail to degrade excess histones and accumulate high levels of endogenous histones on histone chaperones. Co-immunoprecipitation assays showed that these E3 ligases interact with the major E2 enzyme Ubc4 that is involved in the degradation related ubiquitylation of histones. Using mutagenesis we further demonstrate that the RING domains of Hel1, Hel2 and Snt2 are required for histone regulation. Lastly, mutants corresponding to Hel1, Hel2 and Pep5 are sensitive to replication inhibitors. Overall, our results highlight the importance of posttranslational histone regulatory mechanisms that employ multiple E3 ubiquitin ligases to ensure excess histone degradation and thus contribute to the maintenance of genomic stability.

Figure 1. Identification of strains lacking specific E3 ligases that confer sensitivity to histone overexpression.

(A) The tom1 deletion strain can degrade exogenous histones with much slower kinetics than the wild type strain. Histone degradation assay was carried out exactly as described previously whereby a galactose inducible, HIS10-TEV-HA (HTH) tagged histone H3 was expressed in G1 arrested cells for 90 minutes, following which cells were switched to glucose media and harvested at the indicated time points. Whole cell extracts were prepared and resolved on 18% polyacrylamide gels and subjected to Western blotting with the H3-C antibody described previously that recognizes both the endogenous chromatin associated histone H3 as well as the overexpressed H3. The entire experiment was carried out in G1 arrested cells to prevent incorporation of exogenous histones into chromatin that occurs in replicating cells. (B.) Strains lacking Hel1, Hel2, Pep5 and Snt2 E3 ligases are sensitive to histone overexpression. The histone overexpression sensitivity screen was carried out exactly as described previously . Briefly, wild type (WT) BY4741 and deletion strains corresponding to the indicated E3 ligases were transformed with either an empty vector (pYES2-HTH) or a galactose inducible HTH tagged histone H3 expressing plasmid (pYES2-HTT-HHT2). Then 10-fold serial dilutions of the resulting transformants were plated on glucose (no histone overexpression) or galactose (allows histone overexpression) containing media without uracil to select for the plasmid. Plates were photographed after 3 days of incubation at 30°C.

Figure 2. Strains lacking putative E3 ligases Hel1, Hel2, Pep5 and Snt2 are defective in degrading excess histones.

(A) Strains carrying deletions of hel1, hel2, pep5 and snt2 are deficient in degrading exogenously expressed histones. The exogenous histone degradation assay was carried out in the indicated strains carrying the pYES2-HTH-HHT2 plasmid as described in Figure 1A. The rad53 deletion strain serves as a positive control in this assay as we have previously shown that it is defective in the degradation of exogenously expressed histones . The endogenous chromatin bound histone H3 is not degraded in this assay and serves as a loading control. Glu = glucose. (B) The RING domains of Hel1, Hel2 and Snt2 are required for efficient degradation of excess histones. Strains carrying mutations in the RING (r) domains of Hel1, Hel2 and Snt2 were generated as described in the Materials and Methods section. These mutant strains were then transformed with the pYES6/CT-HA-HHT2 plasmid carrying a Blasticidin resistance marker and a galactose inducible, HA-tagged histone H3 gene (HA-H3). The excess histone degradation assay was carried out as described in (A) except that the duration of the experiment was limited to 90 minutes and the exogenous HA-H3 was detected using HA.11 antibodies, while the endogenous H3 was detected using the H3-C antibody described previously , .

Figure 3. hel1, hel2, pep5 and snt2 deletion strains accumulate excessive amounts of endogenous histones on the histone chaperone Asf1.

The chromosomal copy of the gene encoding Asf1 was tagged with a 3×FLAG epitope at the C-terminus in the indicated strains. Whole cell extracts were prepared from 0.5 L cell cultures treated with or without DNA damaging agent Methylmethane Sulfonate (MMS; 0.033%) as described previously . Then, Asf1-FLAG was immunoprecipitated (IPed) using FLAG-M2 antibodies and the co-immunoprecipitated (Co-IPed) histones were analyzed by Western blotting using the H3-C and H4 polyclonal antibodies described previously .

Figure 4. Hel1, Hel2, Pep5 and Snt2 interact with the E2 enzyme Ubc4 as well as histones.

(A) Hel1, Hel2, Pep5 and Snt2 interact with Ubc4. Whole cell extracts prepared from 0.5 liter cultures of the indicated strains were used to IP Ubc4-MYC essentially as described for Asf1-FLAG IPs in Figure 3, but using anti-MYC EZ view beads (Sigma). Co-IPed HA-tagged E3 enzymes were detected using HA.11 antibodies. Apart from the predominant bands corresponding to the full length proteins (indicated by the asterisks), additional bands of weaker intensity are detected and these are likely to be either degradation products (faster migrating) or ubiquitylated (slower migrating) forms of these proteins. No co-IPed proteins were detected in the parental untagged strain carrying Ubc4-MYC but lacking HA-tagged proteins. 1% of the whole cell extracts were loaded on a separate gel to determine that roughly equal amounts of proteins were present in the input fraction by measuring the amount of histone H4. (B) Hel1, Hel2, Pep5 and Snt2 interact with histones. The HA-tagged E3 enzymes were IPed from 1 liter cultures of the indicated strains using HA.11 antibodies. Co-IPed endogenous histones H3 and H4 was detected using the polyclonal antibodies described previously . No co-IPed histones were detected in the “no tag” control. The amount of Ubc4-MYC was measured in 1% of the input fraction to demonstrate that roughly equal amounts of cell extracts were used for the IP reactions. (C) Hel1 and Hel2 interact with Rad53. MYC tagged Hel1 or Hel2 were IPed using MYC antibody beads from 3 liter cultures of strains carrying FLAG-tagged Rad53. IPed and Co-IPed proteins were detected by Western Blotting using appropriate antibodies. The relative amount of histone H3 present in each of the “input” extracts used for the IP is shown to demonstrate that same amount of material was used for each IP. (D) Snt2 interacts with Rad53 whereas Pep5 does not. Tandem Affinity Purification (TAP) tagged Pep5 or Snt2 was IPed from 3 liter cultures of strains carrying FLAG-tagged Rad53. Any co-IPed Rad53 was detected using FLAG antibodies. The similar levels of Rad53-FLAG present in the extracts confirm that the same amount of extracts was used for each IP. No evidence for any interaction between Rad53 and Pep5 was observed even upon scaling up the Pep5-TAP IP by 3-fold (data not shown).

Figure 5. Hel1, Hel2, Pep5 and Snt2 can efficiently ubiquitylate histone H4 in vitro in a reaction that is stimulated by Rad53.

Recombinant and purified components were used to reconstitute the ubiquitylation of histone H4 in vitro exactly as described previously . The reaction products were resolved on an 18% polyacrylamide gel and processed for Western blotting with H4 antibodies. Addition of commercially available recombinant yeast Uba1 (E1), human UbcH5A (E2; homolog of yeast Ubc4) and HIS6-tagged ubiquitin to recombinant human histone H4 with or without Rad53 did not result in appreciable histone modifications, apart from some monoubiquitylation (lanes 2 and 1 respectively). Only the addition of the E3 ligases Hel1 (H1), Hel2 (H2), Pep5 (P5) and Snt2 (S2) purified from yeast extracts via TAP epitope tags resulted in considerable high molecular weight histone modifications, which were further stimulated to varying degrees by the addition of recombinant Rad53 to the reaction mixture.

Figure 6. The E3 ligases Hel1, Hel2, Pep5, Snt2 and Tom1 do not regulate the protein levels of Ubc4, the major E2 enyme involved in excess histone degradation.

Whole cell lysates were prepared as described previously from wild type and the indicated E3 ligase deletion strains carrying MYC-tagged Ubc4. The lysates were resolved on a 12% polyacrylamide gel and processed for Western blotting using MYC antibodies. No significant differences are observed in the levels of full-length Ubc4-MYC, or in its slower migrating modified forms and faster migrating degradation products.

Figure 7. Histone E3 ligase mutants exhibit varying degrees of sensitivity to the replication inhibitor hydroxyurea.

(A) hel1, hel2 and pep5 mutants are sensitive to hydroxyurea. 10-fold serial dilutions of the indicated strains were plated on media containing glucose with or without the indicated concentrations of hydroxyurea (HU) and incubated at 30°C for 3 days before being photographed. (B) Mutations in the RING finger domains of Hel1 and Hel2 render them sensitive to HU. 10-fold serial dilutions of the indicated strains were plated on media containing galactose with or without 200 mM HU and incubated at 30°C for 3 days before being photographed. The hel1-r₁r₂ mutant carries inactivating mutations in both r finger domains of Hel1, while the hel2-r mutant carries inactivating mutations in the single r finger domain of Hel2. The rad53 deletion strain was included as a positive control.