BBAP monoubiquitylates histone H4 at lysine 91 and selectively modulates the DNA damage response

Abstract

Although the BBAP E3 ligase and its binding partner BAL are overexpressed in chemotherapy-resistant lymphomas, the role of these proteins in DNA damage responses remains undefined. Because BAL proteins modulate promoter-coupled transcription and contain structural motifs associated with chromatin remodeling and DNA repair, we reasoned that the BBAP E3 ligase might target nucleosomal proteins. Herein, we demonstrate that BBAP selectively monoubiquitylates histone H4 lysine 91 and protects cells exposed to DNA-damaging agents. Disruption of BBAP-mediated monoubiquitylation of histone H4K91 is associated with the loss of chromatin-associated H4K20 methylase, mono- and dimethyl H4K20, and a delay in the kinetics of 53BP1 foci formation at sites of DNA damage. Because 53BP1 localizes to DNA damage sites, in part, via an interaction with dimethyl H4K20, these data directly implicate BBAP in the monoubiquitylation and additional posttranslational modification of histone H4 and an associated DNA damage response.

Figure 1. BBAP ubiquitylates core histones in vitro

Recombinant wild-type BBAP (BBAP_WT) or BBAP with a deleted ring domain (BBAP_RDdel) was incubated with individual core histone proteins (H3, H2A, H2B or H4), E1, E2 and His ubiquitin in in vitro ubiquitylation assays. Samples were size-fractionated, immunoblotted and analyzed with anti-multihistidine (ubiquitin) (A) or antibodies directed against the individual histones (H2A, H2B, H3, H4) (B). Monoubiquitylated histone proteins are indicated with arrows in B.

Figure 2. BBAP monoubiquitylates histone H4 in vivo

(A) Histone H4 monoubiquitylation in HEK293 cells cotransfected with FLAG-tagged histone H4 and either HA-tagged BBAP_WT, BBAP_RDdel or no BBAP construct. Samples were size-fractionated, immunoblotted and analyzed with anti-FLAG (H4). (B) Coimmunoprecipitation of BBAP and histone H4. HEK293 cells cotransfected with HA-tagged BBAP_WT and FLAG-tagged histone H4 were lysed and immunoprecipitated (IP) with anti-HA or control Ig and immunoblotted with anti-HA (BBAP) (top panel) or anti-FLAG (histone H4) (bottom panel). LC indicates light chain (bottom panel). Cotransfected HEK293 cell lysate (input, left lane) was also directly immunoblotted. (C) Coimmunoprecipitation of nuclear chromatin-associated BBAP and histone H4. DNA-bound nuclear protein complexes were isolated from Hela cells, immunoprecipitated with BBAP antiserum or control IgG and immunoblotted with monoclonal anti-BBAP antibody (top panel) or anti-H4 (bottom panel). The nuclear protein complexes (Nuc. Ext., left lane) were also directly immunoblotted. (D). BBAP depletion by siRNA. Untreated Hela cells and cells treated with a scrambled control siRNA (ctl) or one of two independent BBAP siRNAs (#1, #2) were lysed, size-fractionated and immunoblotted with anti-BBAP (top panel). Blots were subsequently reprobed with anti-actin to confirm equal loading (bottom panel). (E) Monoubiquitylated histone H4 is decreased in BBAP-depleted cells. After Hela cells were transfected with either BBAP siRNA (#1 or #2) or the control siRNA (ctl), histones were purified, size-fractionated and immunoblotted with anti-H4. Histone samples were also immunoprecipitated with anti-ubiquitin and immunoblotted with anti-histone H4 to confirm that the 17kd H4 protein is monoubiquitylated (Supplemental Data, Fig. 2).

Figure 3. BBAP monoubiquitylates histone H4 lysine 91 (H4K91) in vivo

(A) Covalent modifications of histone H4. Serine phosphorylation (P) and acetylation (Ac) and/or methylation (Me) of lysine and arginine residues and ubiquitylation of H4K91 are shown. (B) Histone H4K91 is targeted by the BBAP E3 ligase. Site-directed mutagenesis of each histone H4 lysine residue (5, 8, 12, 16, 20, 31, 44, 59, 77, 79 and 91) was performed and the resulting histone H4 sequences were cloned into a FLAG-tagged expression vector. Thereafter, the individual FLAG-tagged wild type or K → A histone H4 mutants were cotransfected with HA-tagged BBAP and His ubiquitin and the resulting transfectants were lysed, size-fractionated and immunoblotted with anti-FLAG (histone H4). (C) BBAP ubiquitylates H4 in histone octamers (core histones) in vitro. Recombinant wild-type BBAP_WT or BBAP_RDdel was incubated with either free H4 or intact histone octamers (core histones) in the presence with E1, E2 and His ubiquitin in in vitro ubiquitylation assays. Samples were size-fractionated, immunoblotted and analyzed with anti-multihistidine (ubiquitin). Monoubiquitylated histone H4 and free His-ubiquitin are indicated with arrows. In Supplemental Data, Fig. 3, the same samples were immunoblotted with anti-H4.

Figure 4. BBAP modulates the cellular response to DNA damage

(A) BBAP overexpression protects cells against Hu or Dox treatment. HEK293 cells transfected with BBAP_WT, BBAP_RDdel or vector alone were subsequently treated with Hu or Dox for 24 hrs and counted. Although BBAP_WT did not alter the proliferation of untreated cells, BBAP_WT protected cells against Hu and Dox toxicity (p < .01 for both [ANOVA]). (B) BBAP depletion augments the cellular response to DNA damaging agents. Hela cells were transfected with control or BBAP siRNAs (siRNA#1, #2), treated with Dox at 50ng/ml, 200ng/ml, and 400ng/ml or left untreated for 1 – 96 hrs. and subsequently evaluated by MTS assay. Although BBAP RNAi reduced the growth of untreated cells, the consequences of BBAP depletion were most striking in cells treated with low-dose Dox (50 ng) (p < .001, two-way ANOVA). After 48 hr. of treatment with low-dose Dox (50 ng), cellular proliferation (as assessed by MTS assay) was ≈ 70 – 80% lower in BBAP-depleted cells than in control RNAi or parental cells. (C) Cellular apoptosis following BBAP depletion and Dox treatment. Parental, control and BBAP siRNA-transfected Hela cells were untreated or treated with Dox at 50ng/ml and 200ng/ml for 24 hrs and analyzed for apoptosis with Annexin V/PI staining. Error bars in A and B represent the standard deviation (SD) of the mean for three replicates in a representative experiment.

Figure 5. BBAP selectively modulates 53BP1 foci formation

Hela cells were transfected with BBAP or control siRNAs and either left untreated or treated with doxorubicin (Dox) for 1 – 24 hr. Thereafter, DDR foci formation - p1981-ATM, γ-H2AX and MDC1 (A) and 53BP1 and BRCA1 (B) – was evaluated at serial timepoints. In both A) and B), representative photographs (left) and accompanying plots of the percentage of cells with greater than 10 DDR-foci/nuclear capture (right) are shown. A minimum of 100 BBAP siRNA#1, siRNA#2 and control siRNA treated cells were analyzed at each timepoint. At 1, 2 and 4 hrs following induction of DNA damage with Dox, there were significantly fewer 53BP1 foci in BBAP-depleted cells than in controls (with a maximal reduction at 1 hour, p<0.01 for siRNA1 or siRNA2 versus control, B). At the earliest timepoint following DNA damage, there was also a minor reduction in BRCA1 repair foci in BBAP-depleted cells (at 1 hr, p<0.01 for siRNA1 or siRNA2 versus control, B). Error bars on the right represent the SD of the mean for three independently stained slides for each timepoint and condition.

Figure 6. Histone H4 modifications following DNA damage and BBAP depletion

Parental, BBAP-depleted and control siRNA Hela cells were treated with Dox (50ng/ml) for 1 to 24 hrs. Thereafter, total cell lysates were prepared from one set of samples and immunoblotted to assess BBAP abundance and re-blotted for actin (to confirm equal loading) (immunoblots, A, and scanning densitometric tracings, B, top panels). Histones were purified from a second set of identically-treated samples and immunoblotted with anti-histone H4, antibodies directed against mono- or dimethyl H4K20 or H4K91Ac or anti-H2A (to confirm equal loading) (immunoblots, A, and scanning densitometric tracings, B, additional panels).

Figure 7. BBAP depletion decreases the abundance of nuclear chromatin-associated Set8

After confirming the specificity of the Pr-Set7/Set8 antibody (Supplemental Data, Fig. 6), chromatin-associated nuclear proteins were prepared from parental control siRNA or BBAP-depleted Hela cells that were untreated (−) or treated with Dox (50ng/ml) for 1 hr (+). Thereafter, the samples were immunoblotted for BBAP, PR-Set7/Set8 and H2A (to confirm the enrichment of chromatin-associated proteins and equal loading).