Cyclin F-Mediated Degradation of SLBP Limits H2A.X Accumulation and Apoptosis upon Genotoxic Stress in G2

Abstract

SLBP (stem-loop binding protein) is a highly conserved factor necessary for the processing, translation, and degradation of H2AFX and canonical histone mRNAs. We identified the F-box protein cyclin F, a substrate recognition subunit of an SCF (Skp1-Cul1-F-box protein) complex, as the G2 ubiquitin ligase for SLBP. SLBP interacts with cyclin F via an atypical CY motif, and mutation of this motif prevents SLBP degradation in G2. Expression of an SLBP stable mutant results in increased loading of H2AFX mRNA onto polyribosomes, resulting in increased expression of H2A.X (encoded by H2AFX). Upon genotoxic stress in G2, high levels of H2A.X lead to persistent γH2A.X signaling, high levels of H2A.X phosphorylated on Tyr142, high levels of p53, and induction of apoptosis. We propose that cyclin F co-evolved with the appearance of stem-loops in vertebrate H2AFX mRNA to mediate SLBP degradation, thereby limiting H2A.X synthesis and cell death upon genotoxic stress.

Keywords: DNA damage response; H2A.X; SLBP; apoptosis; canonical histone mRNA metabolism; cyclin F; genotoxic stress; proteasome; ubiquitin.

Figure 1. SLBP specifically interacts with cyclin F

(a) Mass spectrometry analysis of the sequential affinity- and immuno-purifications of FLAG-STREP-tagged SLBP listing number of unique peptides and peptide spectrum matches (total number of identified peptides, including those redundantly identified) for the indicated proteins. (b) HEK293T cells were transfected with either an empty vector (EV) or FLAG-STREP-tagged SLBP constructs. MLN4924 was added to the cells for four hours prior to collection. Whole cell extracts (WCE) were affinity purified (AP) with anti-STREP resin and immunoblotted as indicated. (c) Endogenous SLBP or endogenous cyclin F was immunoprecipitated from HEK293T cell extracts using SLBP or cyclin F antibody, respectively (IgG was used as negative control). Whole cell extracts and immunoprecipitations were immunoblotted as indicated. See also was used at 200 ng/mLFigure S1

Figure 2. SLBP is degraded in G2 in a cyclin F-dependent manner

(a) HEK293T cells were transfected with the indicated constructs, lysed, and immunoblotted as indicated. Where indicated, cells were treated with MG132 for four hours prior to collection. (b) HeLa cells were synchronized at G1/S by double-thymidine block before trypsinization and release into fresh media. Cells were transfected with either an siRNA targeting cyclin F or a non-targeting (N/T) siRNA between the first and second thymidine treatment, collected at the indicated time points, and immunoblotted as indicated. (c) HEK293T cells were transfected with the indicated constructs. Whole cell extracts were immunoblotted as indicated. (l.e.: long exposure, s.e: short exposure). (d) HEK293T cells were transfected with the indicated constructs. Whole cell extracts were immunoprecipitated with anti-FLAG resin and immunoblotted as indicated. The bracket indicates a ladder of bands corresponding to poly-ubiquitylated SLBP. See also Figure S2

Figure 3. Arg97 and Leu99 in SLBP are necessary for its interaction with cyclin F

(a) Schematic representation of SLBP highlighting three putative CY motifs in SLBP. (b) HEK293T cells were transfected with either an empty vector (EV) or FLAG-STREP-tagged SLBP constructs. Cells were treated with MLN4924 for four hours prior to collection. Whole cell extracts were affinity purified with anti-STREP resin and immunoblotted as indicated. (c) HEK293T cells were transfected with FLAG-tagged SLBP constructs. Whole cell extracts were supplemented with SUPERase-In™ RNase Inhibitor and immunoprecipitated with anti-FLAG resin. The immunoprecipitations were then immunoblotted as indicated. (d) HEK293T cells were transfected with either EV or FLAG-tagged SLBP constructs. Cell lysates were immunoprecipitated with anti-FLAG resin and bound RNA was purified. Random primed cDNAs were prepared from equal volume of isolated RNA and analyzed by qPCR for total HIST1H3H mRNA and total H2AFX mRNA. The data are presented as fold change relative to the EV sample. (e) HEK293T cells were transfected with FLAG-STREP-tagged SLBP constructs. Cells were treated with MLN4924 for four hours prior to collection. Whole cell extracts were affinity purified with anti-STREP resin and immunoblotted as indicated. (f) Alignment of the CY motif in SLBP orthologs. Critical amino acids required for cyclin F (RxL/I) and cyclin A2 (RKL/IL) binding are highlighted in gray. See also Figure S3

Figure 4. SLBP(RL97/99AA) is not degraded in G2 leading to increased H2A.X levels

(a) HeLa cells infected with retroviruses expressing either FLAG-tagged SLBP or FLAG-tagged SLBP(RL97/99AA) were synchronized at G1/S by double-thymidine block before trypsinization and release into fresh medium. Cells were collected at the indicated times, lysed, and immunoblotted as indicated (asterisks denote non-specific bands). (b) U2OS cells infected with lentiviruses expressing untagged SLBP(RL97/99AA) under the control of a doxycycline-inducible promoter were synchronized at G1/S using a double-thymidine block before trypsinization and release into fresh medium. Where indicated, doxycycline was added with the second thymidine treatment and was not re-added after release from double-thymidine block. Cells were collected at the indicated time points, lysed, and immunoblotted as indicated. (c) Processed (top) and polyadenylated (bottom) histone mRNAs with histone open reading frame (ORF), stem-loop, and the histone downstream element (HDE). The cleavage site for the processing of canonical histone and H2AFX mRNAs and the qPCR primers used to evaluate total (top) and polyadenylated (bottom) histone mRNA levels are indicated. The total histone qPCR primer set detects both processed and polyadenylated histone mRNA species. (d) U2OS cells prepared as in (b) were collected 10 hours after release from double-thymidine block and fractionated by sucrose density gradient centrifugation into light (2–3) and heavy (≥4) ribosomes. Random primed cDNAs were prepared from the mRNA isolated in each fraction and analyzed by qPCR for total H2AFX mRNA and total HIST1H3H mRNA presented as a ratio to polyadenylated H2AFX mRNA. Respective mRNAs from the corresponding unfractionated sample were used to normalize each data point. The data are presented as mean ± SD (*p≤0.05, NS: not significant, n=3, each in triplicate). See also Figure S4

Figure 5. SLBP(RL97/99AA) expression in G2 leads to persistent DNA damage response signaling upon genotoxic stress

(a) Schematic representation of the experiment. U2OS cells infected with lentiviruses expressing untagged SLBP(RL97/99AA) under the control of a doxycycline-inducible promoter were synchronized at G1/S using a double-thymidine block before trypsinization and release into fresh medium. Doxycycline was added with the second thymidine treatment and was not re-added after release from double-thymidine. Neocarzinostatin (NCS) was added 10 hours after the release from double-thymidine block. (b) Cells were collected at the indicated time points, lysed, and immunoblotted as indicated (asterisks denote non-specific bands). (c) U2OS cells prepared as in (a) were collected at one and fours hours after treatment with NCS, fixed, stained with γH2A.X antibody, and analyzed by flow cytometry. The data are presented as mean ± SD (*p≤0.05, NS: not significant, n=3). (d) U2OS cells prepared as in (a) were collected at four hours after treatment with NCS, fixed with 4% paraformaldehyde, and stained with a γH2A.X antibody. Images of the γH2A.X foci underwent automated processing with at least 29,000 cells counted per sample. The data are presented as mean ± SEM for one representative experiment (***p≤0.001). (e) U2OS cells prepared as in (a) were collected 1 hour after treatment with NCS and fractionated by sucrose density gradient centrifugation into light (2–3) and heavy (≥4) ribosomes. Random primed cDNAs were prepared from the mRNA isolated in each fraction and analyzed by qPCR for total H2AFX mRNA presented as a ratio to polyadenylated H2AFX mRNA. mRNA from the corresponding unfractionated sample was used to normalize each data point. The data are presented as mean ± SD (*p≤0.05, NS: not significant, n=2, each in triplicate). See also Figures S5 and S6

Figure 6. SLBP(RL97/99AA) expression in G2 leads to increased apoptosis upon genotoxic stress

(a) U2OS cells prepared as in Figure 5A were collected 4 hours after treatment with NCS, stained with Annexin-V Alexa-488 conjugate and propidium iodide, and analyzed by flow cytometry. Data is presented as the percent of cells that stained positive for Annexin-V and negative for propidium iodide (*p≤0.05, n=2). (b) Non-treated (NT) U2OS or U2OS cells treated with NCS (+NCS) were prepared as in Figure 5A, and allowed to grow for approximately 12 days, until colonies could be identified. Each plate was then fixed with 6% glutaraldehyde and stained with crystal violet. The data are presented as mean ± SD (*p≤0.05, NS; not significant, n=3, each in triplicate). (c) A model of the regulation of survival and apoptosis by the cyclin F-SLBP axis. During G2, after the large majority of DNA replication has occurred, cyclin F accumulates, thereby promoting SLBP degradation. Cells in G2 are able to survive moderate genotoxic stress, whereas high levels of genotoxic stress leads to apoptosis. Stabilization of SLBP into G2 induces high expression of H2A.X and sensitizes the cell to genotoxic stress.