DNA damage-induced dynamic changes in abundance and cytosol-nuclear translocation of proteins involved in translational processes, metabolism, and autophagy

Abstract

Ionizing radiation (IR) causes DNA double-strand breaks (DSBs) and activates a versatile cellular response regulating DNA repair, cell-cycle progression, transcription, DNA replication and other processes. In recent years proteomics has emerged as a powerful tool deepening our understanding of this multifaceted response. In this study we use SILAC-based proteomics to specifically investigate dynamic changes in cytoplasmic protein abundance after ionizing radiation; we present in-depth bioinformatics analysis and show that levels of proteins involved in autophagy (cathepsins and other lysosomal proteins), proteasomal degradation (Ubiquitin-related proteins), energy metabolism (mitochondrial proteins) and particularly translation (ribosomal proteins and translation factors) are regulated after cellular exposure to ionizing radiation. Downregulation of no less than 68 ribosomal proteins shows rapid changes in the translation pattern after IR. Additionally, we provide evidence of compartmental cytosol-nuclear translocation of numerous DNA damage related proteins using protein correlation profiling. In conclusion, these results highlight unexpected cytoplasmic processes actively orchestrated after genotoxic insults and protein translocation from the cytoplasm to the nucleus as a fundamental regulatory mechanism employed to aid cell survival and preservation of genome integrity.

Keywords: DNA damage response (DDR); cytoplasm; ionizing radiation (IR); protein regulation; protein synthesis; quantitative proteomics.

Figure 1.

Experimental strategy for determination of protein abundance dynamics in the cytoplasm. Two series of three SILAC-encoded cell populations were irradiated with 6Gy and grown in culture for different lengths of time as indicated before harvest. Cell population in each time series were mixed 1:1:1. Proteins from the cytoplasmic fraction were precipitated with acetone, subsequently digested and subjected to affinity enrichment of phosphopeptides by ERLIC and TiO2 chromatography (phospho-data is not included in this manuscript). A small part of the digest was separated by isoelectric focusing for protein normalization. All peptides were analyzed by LC-MS/MS and raw data were processed by MaxQuant. Processed data were analyzed using various computational statistics and bioinformatics tools.

Figure 2.

Clustering, validation and deciphering of temporal profiles for proteins regulated during DDR. (a), Distribution of log2-transformed ratios of all proteins identified in the cytoplasmic fraction quantified at all five time points. The dashed lines indicate the boundaries for the 1.25 fold change regions. (b), Graphs showing the five temporal clusters identified using the soft clustering algorithm, Mfuzz. The color code shows the membership value of proteins in individual clusters and a bar diagram display the number of proteins in each clusters.

Figure 3.

Cluster-wise enrichment of functional and compartmental categories. Heatmap of gene ontologies significantly enriched among the different temporal clusters. Statistical significance is based on hypergeometric testing using Fisher’s Exact test and Benjamini-Hochberg adjustment of p-values. An enrichment score is defined as -logp. Ontologies with p < 0.001 and p > 0.05 were given a fixed score of 3 and 1, respectively. (a), Gene ontology biological processes. (b), Gene ontology cellular components.

Figure 4.

Immunofluorescence, immunoblotting and IHC analyses of protein cytoplasm-nuclear translocation dynamics in human cells exposed to IR.(a, b) Immunofluorescence staining illustrates partial and transient translocation of NFĸB p65 from the cytoplasm into nuclei of HeLa, BJ (a) and GM00130 (b) cells, with the maximum nuclear presence of p65 at 1 h after irradiation. Scale bar, 20 µm. (c), Immunoblotting analysis of fractionated cytoplasmic versus nuclear proteins isolated from the GM00130 cells at the indicated times after irradiation. The efficiency of the fractionation procedure is documented by monitoring selective presence of MEK1 (a cytoplasmic marker) and SMC1 (nuclear marker) proteins in cytosolic and nuclear fractions, respectively. Phosphorylation of SMC1 at serine 957 (pS957), an established ATM-mediated phosphorylation event induced by IR, documents the peak activity of DNA damage signaling at 1h, and partial recovery by 8 h post-IR. Note partial decrease of cytoplasmic NFĸB p65 and its correspondingly increased abundance in the nuclear fraction at 1 h post-IR, consistent with the immunofluorescence staining in (a, b). (d), Quantification of immunoblotting data on dynamic changes in the relative abundance of NFĸB p65 in cytoplasm versus nuclei of GM00130 cells, based on integrated intensity (see methods), normalized to the 1-hour cytosolic value as the reference point. Error bars represent the SD from three independent irradiation and fractionation experiments. (e), NFĸB p65 staining in human medulloblastoma confirming cytoplasmic-nuclear translocation after IR exposure. (f), quantification of NFĸB p65 stainings in xenografts derived from two different medulloblastoma cell lines, D324 and D283.

Figure 5.

Significant linear motif in translocation candidate proteins.(a), three significant linear motifs with a central lysine (K). Score and fold increase are indicated in the table below. (b), four significant linear motifs with a central arginine (R), the SR motif being highly abundant in hnRNP-protein sequences.

Figure 6.

Significant subnetworks within the probabilistic network. Using the network motif algorithm MCODE significant subnetworks within the STRING predicted network were found. The five highest-ranking subnetworks are shown here. Node color indicate assigned cluster membership and connecting lines are determined by membership value.

Figure 7.

Schematic model illustrating time-course of cytoplasmic responses after IR. Changes in protein abundance over time illustrated for functional classes of proteins assigned to cluster I, II, III and IV.