A quantitative 14-3-3 interaction screen connects the nuclear exosome targeting complex to the DNA damage response

Through a 14-3-3-interaction screen for DNA damage-induced protein interactions in human cells, Blasius et al. identified protein complexes connected to RNA biology. These include the nuclear exosome targeting (NEXT) complex that regulates turnover of noncoding RNAs termed promoter upstream transcripts (PROMPTs). The NEXT subunit RBM7 is phosphorylated upon DNA damage by the MAPKAPK2 kinase, and this mediates 14-3-3 binding and decreases PROMPT binding.

Transcription occurs ;0.5-2.5 kb upstream of mammalian RNA polymerase I, II, and III promoters, producing promoter upstream transcripts (PROMPTs) (Preker et al. 2008(Preker et al. , 2011. PROMPTs are degraded by the RNA exosome (Chlebowski et al. 2013), and because exosome depletion caused changes in promoter DNA methylation, it is proposed that PROMPTs influence this methylation to alter gene expression (Preker et al. 2008). Recently, the human nuclear exosome targeting (NEXT) complex was shown to mediate PROMPT degradation by delivering them to the RNA exosome (Lubas et al. 2011). NEXT comprises RBM7, which contains an RNA-binding motif (Guo et al. 2003); ZCCHC8 (Gustafson et al. 2005); and the putative RNA helicase MTR4/SKIV2L2. The precise functions for PROMPTs and the NEXT complex and whether they are regulated remain unknown. Here, we show that ultraviolet light (UV) exposure leads to MK2-mediated RBM7 phosphorylation, triggering 14-3-3 binding and controlling PROMPT turnover, thus highlighting PROMPT regulation as a new facet of the DDR.

UV induces PROMPTs and decreases their binding to RBM7
RBM7 depletion causes PROMPT accumulation (Lubas et al. 2011), suggesting that without RBM7, the NEXT complex cannot deliver PROMPTs to the RNA exosome for degradation. We reasoned that if UV-dependent RBM7 phosphorylation and 14-3-3 binding affected PROMPT targeting, PROMPT levels would change upon UV. Indeed, UV irradiation of HeLa cells markedly increased Figure 3. RBM7 is phosphorylated by MK2 in response to UV. (A, top) Peptides identified by phosphopeptide mapping. UV-induced phosphorylated residues are marked by an asterisk, and phosphorylated residue positions are in brackets. 14-3-3-binding motifs are underlined, and consensus 14-3-3-binding motifs are shown above. (Bottom) RBM7 domain structure. RNA recognition motif (RRM; black) and residues phosphorylated upon UV are in bold. (B) GST-14-3-3 pull-downs for the indicated GFP-RBM7 derivatives. (C) Time course of interaction between 14-3-3 and RBM7. U2OS cells stably expressing GFP-RBM7 were UV-C-treated and harvested at the indicated times, GST-14-3-3 pull-downs were done, and interactions with GFP-RBM7 and ZCCHC8 were monitored. (D) U2OS cells stably expressing GFP-RBM7 (lanes 2-7) or empty vector (lane 1) were UV-C-treated in the presence or absence of the indicated inhibitors, and extracts were used for GST-14-3-3 pull-down assays. (E) U2OS cells stably expressing GFP-RBM7 were transfected with control siRNA (siCont) or siRNA against MK2 (siMK2) and irradiated with 40 J/m 2 UV-C 2 h prior to harvesting and GST-14-3-3 pull-downs.
The above data suggested that UV-induced RBM7 phosphorylation decreases RBM7 binding to PROMPTs and targeting them to the exosome. Indeed, immunoprecipitation studies revealed that UV treatment decreased PROMPT binding to Flag-RBM7 but not Flag-RBM7-S136A/S204A (binding to FLAG-RBM7-S136A/S204A actually increased, likely reflecting PROMPT induction after UV) (Fig. 4B). Similar effects were observed when we used GFP-tagged RBM7 constructs (Supplemental Fig.  S3C). In line with these findings, overexpression of RBM7-S136A/S204A but not RBM7-WT prevented UVinduced PROMPT accumulation (Supplemental Fig. S3E). Furthermore, consistent with these RBM7-mediated responses playing a functional role in response to DNA damage, RBM7 depletion by two different siRNAs caused hypersensitivity to the UV-mimicking drug 4nitroquinoline 1-oxide (4-NQO) (Fig. 4C). As shown in Supplemental Figure S3D, 4-NQO hypersensitivity was also seen upon RBM7 depletion in U2OS cells and was partially rescued by expression of GFP-RBM7-WT. In contrast, expression of RBM7-S136A/S204A, which is not controlled by MK2, increased cell survival, suggesting that lower PROMPT levels promote survival of damaged cells.
In summary, through proteomic screening, we identified various factors whose interactions with 14-3-3 proteins are altered by UV exposure. Although we focused follow-up work on pathways connected to the DDR, we note that RNA damage could also contribute to the responses that we observed. It is striking that the majority of proteins identified displaying the most pronounced UV-induced changes in 14-3-3 binding have intimate connections to RNA. Our data are thus in line with other work (Sette 2010;Reinhardt et al. 2011;Beli et al. 2012;Bhatia et al. 2014;Britton et al. 2014) identifying RNAassociated proteins as impacting on the DDR, highlighting how controlling RNA metabolism and functions is likely to represent important but as-yet relatively unexplored aspects of the DDR. Our findings also suggest new linkages between responses mediated by p38/MK2 and events such as RNA polyadenylation, transcriptional elongation, and translational control that could now be explored. Exemplifying the potential for such work, our studies on the NEXT component RBM7 has led to a model in which RBM7 binds PROMPTs in unchallenged cells, targeting them for degradation by the nuclear RNA exosome. Upon DNA (and potentially RNA) damage created by UV or other stresses, MK2 phosphorylates RBM7 on Ser136 and Ser204, creating a binding site for 14-3-3 proteins that impairs RBM7 RNA binding, preventing the NEXT complex from delivering PROMPTs and possibly other RNAs for degradation. Consequently, PROMPT levels increase upon UV, potentially enhancing cell survival via changes in gene expression (Fig. 5). It will be interesting to see whether RBM7 is involved in other aspects of RNA metabolism. Indeed, as RBM7 also interacts with splicing factors and the nuclear proteasome (Supplemental Tables 2, 3; Lubas et al. 2011), it will be worthwhile assessing whether these interactions are affected by MK2-mediated RBM7 phosphorylation and 14-3-3 binding. Additionally, it will be interesting to establish how RBM7 phosphorylation and 14-3-3 interactions affect PROMPT binding and how UV-induced changes in PROMPT levels or other readouts of NEXT complex activity affect cell physiology.

Materials and methods
Additional methods are described in the Supplemental Material.  RBM7-associated PROMPTs were quantified by qPCR, and PROMPT levels from nontreated (NT) cells were set as 1. (C) Cytotoxicity of HCT116 cells in response to chronic 4-nitroquinoline 1-oxide (4-NQO) treatment measured by SRB assay. Error bars show standard deviation from three independent experiments. (*) P-value < 0.1; (**) P-value < 0.01.

Protein-RNA coimmunoprecipitations
[Roche]) and sonicated for 5 sec at 30% amplitude, and extracts were centrifuged for 20 min at maximum speed. Protein extract was used for GFP immunoprecipitation (GFP-Trap agarose beads [ChromoTek]) or Flag immunoprecipitation (anti-Flag M2 affinity gel [Sigma-Aldrich]) using 20 mL of beads. After 1 h at 4°C, beads were washed three times with buffer A. Retained RNA was purified by an RNeasy Mini Kit (Qiagen), reverse-transcribed with high-capacity cDNA reverse transcription kit (Applied Biosystems), and quantified by quantitative PCR. Results were quantified by DDCt method, normalizing first to GAPDH and then to corresponding nontreated control.