Requirement of heterogeneous nuclear ribonucleoprotein C for BRCA gene expression and homologous recombination

Abstract

Background: Heterogeneous nuclear ribonucleoprotein C1/C2 (hnRNP C) is a core component of 40S ribonucleoprotein particles that bind pre-mRNAs and influence their processing, stability and export. Breast cancer tumor suppressors BRCA1, BRCA2 and PALB2 form a complex and play key roles in homologous recombination (HR), DNA double strand break (DSB) repair and cell cycle regulation following DNA damage.

Methods: PALB2 nucleoprotein complexes were isolated using tandem affinity purification from nuclease-solubilized nuclear fraction. Immunofluorescence was used for localization studies of proteins. siRNA-mediated gene silencing and flow cytometry were used for studying DNA repair efficiency and cell cycle distribution/checkpoints. The effect of hnRNP C on mRNA abundance was assayed using quantitative reverse transcriptase PCR.

Results and significance: We identified hnRNP C as a component of a nucleoprotein complex containing breast cancer suppressor proteins PALB2, BRCA2 and BRCA1. Notably, other components of the 40S ribonucleoprotein particle were not present in the complex. hnRNP C was found to undergo significant changes of sub-nuclear localization after ionizing radiation (IR) and to partially localize to DNA damage sites. Depletion of hnRNP C substantially altered the normal balance of repair mechanisms following DSB induction, reducing HR usage in particular, and impaired S phase progression after IR. Moreover, loss of hnRNP C strongly reduced the abundance of key HR proteins BRCA1, BRCA2, RAD51 and BRIP1, which can be attributed, at least in part, to the downregulation of their mRNAs due to aberrant splicing. Our results establish hnRNP C as a key regulator of BRCA gene expression and HR-based DNA repair. They also suggest the existence of an RNA regulatory program at sites of DNA damage, which involves a unique function of hnRNP C that is independent of the 40S ribonucleoprotein particles and most other hnRNP proteins.

Figure 1. Presence of hnRNP C in PALB2-containing nucleoprotein complexes.

A. Schematic diagram of the PALB2 purification procedure. B. Sizes of DNA fragments in solubilized chromatin fractions after digestion of insoluble nuclear structures with micrococcal nuclease (MNase). C. Silver-stained gel showing the components of TAP-purified PALB2 complexes from the solubilized chromatin fraction. D. Protein components of the PALB2 complexes identified by liquid chromatography tandem mass spectrometry (LC-MS/MS). The numbers shown are the averages of the numbers of unique peptides detected for each protein in two independent experiments. E. The interaction between hnRNP C and PALB2 is mediated by RNA. Nuclear pellets of U2OS cells were digested with DNase I or RNase A, and the nuclease-released components were IPed with a PALB2 antibody. The nuclease-released materials and IPed proteins were analyzed by Western blotting.

Figure 2. Critical role of hnRNP C in HR and DSBR.

A. Schematic diagrams of the GFP-based DNA repair reporters used in this study. B–C. DR-U2OS cells containing a stably integrated HR reporter were treated with control or hnRNP C siRNAs for 48 hr and then transfected with an I-SceI expression plasmid (pCBASce) to induce DSB formation and repair. B shows representative downregulation of hnRNP C 72 hr after siRNA transfection and C shows GFP positive cells measured 60–72 hr after pCBASce transfection. D. DR-U2OS cells were treated with control or hnRNP C (629) siRNAs for 72 hr and then co-transfected with pCBASce together with vector, wt hnRNP C or siRNA-resistant hnRNP C plasmids; GFP positive cells were counted 72 hr later. E. U2OS cell lines each harboring a different reporter as indicated were treated with control siRNA or a mixture of the two hnRNP C siRNAs for 48 hr and then transfected with pCBASce, and GFP positive cells were measured 72 hr later. Values shown are averages of at least 3 independent experiments and errors bars represent standard deviations.

Figure 3. Effect of hnRNP C depletion on cell cycle distribution before and after IR.

A. DR-U2OS cells were treated with control, PALB2 or hnRNP C siRNAs for 72 hr and then subjected to 10 Gy of IR. Cells were labeled with BrdU either before or 16 hr post IR, and cell cycle profiles were analyzed by anti-BrdU staining and FACS. Cells in S, G1 and G2/M phases were indicated by upper, lower left and lower right boxes, respectively. Early S and late S phase cells are separated by an arbitrary dotted line and indicated by “ES” and “LS”. B. Quantification of cell cycle distributions in two independent experiments. Error bars represent standard deviations, and the asterisk indicates p≤0.05. C. Cells were treated with the siRNAs and subjected to IR in the same way as in A, and mitotic index was measured by phosphorylated histone H3 staining and FACS. D. Cells were treated with control or hnRNP C siRNAs for 72 hr and then subjected to 10 Gy of IR. Cells were harvested at indicated time points, and cellular abundance of hnRNP C and γH2A.X were analyzed by Western blotting. E. Cells treated with siRNAs and IR as in D were fixed and the abundance and localization of hnRNP C and γH2A.X were analyzed by IF.

Figure 4. Nuclear localization properties of hnRNP C.

A. Control and irradiated DR-U2OS cells as indicated were fixed, permeabilized and double stained with hnRNP C and γH2A.X antibodies. Some of the nuclear foci where the two proteins colocalize are marked by white arrows. B. Control and irradiated cells were fixed, permeabilized, co-stained with hnRNP C and PALB2 antibodies, and analyzed by confocal microscopy. C. Control and irradiated cells were first permeabilized, then treated without or with RNase A and finally fixed for IF analysis.

Figure 5. Selective regulation of DNA repair and replication genes by hnRNP C.

A–B. DR-U2OS cells were treated with transfection reagent alone (labeled as “no siRNA”), control siRNA or hnRNP C siRNAs for 72 hr and protein amounts were analyzed by Western blotting. C. Total RNAs were isolated from cells 48–72 hr after transfection and mRNA amounts of the 6 genes indicated were analyzed by quantitative RT-PCR. Values shown are averages of at least 3 independent experiments and error bars represent standard deviations. P values were calculated with student's t test using GraphPad Prism V5. P values smaller than 0.05 are denoted by one asterisk and those smaller than 0.01 are indicated by two asterisks.

Figure 6. Depletion of hnRNP C leads to Alu element exonization in BRCA1, BRAC2, RAD51 and BRIP1.

Genome browser view of BRCA1, BRAC2, RAD51 and BRIP1 genes displaying RNA-Seq data (overlapping reads per nucleotide; blue) from control and HNRNPC knockdown HeLa cells, that were independently transfected with two different siRNAs (KD1 and KD2), as well as hnRNP C iCLIP data (crosslink events per nucleotide; purple). RefSeq transcript annotations (blue) and Alu elements in antisense orientation (orange) are depicted below. Yellow boxes contain zoomed regions within the four genes where hnRNP C-repressed Alu exonization events were detected (marked by red arrowheads). See ref. #8 for details for data generation and analyses.