PI 3 kinase related kinases-independent proteolysis of BRCA1 regulates Rad51 recruitment during genotoxic stress in human cells

Abstract

Background: The function of BRCA1 in response to ionizing radiation, which directly generates DNA double strand breaks, has been extensively characterized. However previous investigations have produced conflicting data on mutagens that initially induce other classes of DNA adducts. Because of the fundamental and clinical importance of understanding BRCA1 function, we sought to rigorously evaluate the role of this tumor suppressor in response to diverse forms of genotoxic stress.

Methodology/principal findings: We investigated BRCA1 stability and localization in various human cells treated with model mutagens that trigger different DNA damage signaling pathways. We established that, unlike ionizing radiation, either UVC or methylmethanesulfonate (MMS) (generating bulky DNA adducts or alkylated bases respectively) induces a transient downregulation of BRCA1 protein which is neither prevented nor enhanced by inhibition of PIKKs. Moreover, we found that the proteasome mediates early degradation of BRCA1, BARD1, BACH1, and Rad52 implying that critical components of the homologous recombination machinery need to be functionally abrogated as part of the early response to UV or MMS. Significantly, we found that inhibition of BRCA1/BARD1 downregulation is accompanied by the unscheduled recruitment of both proteins to chromatin along with Rad51. Consistently, treatment of cells with MMS engendered complete disassembly of Rad51 from pre-formed ionizing radiation-induced foci. Following the initial phase of BRCA1/BARD1 downregulation, we found that the recovery of these proteins in foci coincides with the formation of RPA and Rad51 foci. This indicates that homologous recombination is reactivated at later stage of the cellular response to MMS, most likely to repair DSBs generated by replication blocks.

Conclusion/significance: Taken together our results demonstrate that (i) the stabilities of BRCA1/BARD1 complexes are regulated in a mutagen-specific manner, and (ii) indicate the existence of mechanisms that may be required to prevent the simultaneous recruitment of conflicting signaling pathways to sites of DNA damage.

Figure 1. Downregulation of BRCA1 protein during genotoxic stress.

A) Top, BRCA1 expression in HeLa cells treated with UVC (30 J/m²) was detected by immunoblotting after harvesting at the indicated times. Bottom, immunostaining of BRCA1 at 4 hrs post-treatment. DNA was counterstained with DAPI. B) BRCA1 levels in HeLa cells treated with IR (10 Gy) for the indicated times. C) BRCA1 levels in HeLa cells treated with the DNA alkylating agent, methylmethanesulfonate (MMS, 200 µM) for the indicated times. D) BRCA1 levels in various cell types treated with 200 µM MMS for the indicated times. All immunoblottings were conducted using total cell extracts. β-actin was detected to ensure equal protein loading.

Figure 2. BRCA1 downregulation is independent of apoptosis and is reversible.

A) Immunostaining of BRCA1 in HeLa cells treated with 200 µM MMS. Cells were harvested at 3 and 6 hrs or changed to MMS free medium for the later times. The nuclei were counterstained with DAPI. B) Top left, immunoblotting of BRCA1 and apoptosis markers, PARP-1 and Caspase-3, in HeLa cells treated as indicated above. Bottom left, BRCA1 band intensity was quantified and data are expressed as percentage of untreated cells. Right, immunoblotting for PARP-1 and Caspase-3 following treatment with high dose of UVC (100 J/m²).

Figure 3. Downregulation of BRCA1 occurs independently of the cell cycle phases.

A) Synchronized HeLa cells, using a thymidine double block (TDB) method, were treated with 200 µM MMS for 3 hrs at various time points post-release. Cell cycle analysis (top panel) and immunoblotting (low panel) were conducted at the indicated time points. B) Downregulation of BRCA1 during cell cycle progression in primary cells. Top, human primary fibroblasts were synchronized in G0/G1 by contact inhibition and were released by replating at low density. Bottom, following UVC (30 J/m²) treatment for the last 2 hrs, cell were harvested at the indicated times for immunoblotting. β-actin immunodetection was used as loading control.

Figure 4. The DNA damage-activated PIKKs family members ATM, ATR and DNA-PK are not required for downregulation of BRCA1.

A) Immunoblotting detection of BRCA1 in HeLa cells treated with 10 Gy IR or 200 µM MMS. B) BRCA1 downregulation is not blocked by caffeine. Immunoblotting detection of BRCA1 in HeLa cells pre-treated with 10 mM caffeine for 30 min prior to 200 µM MMS treatment for 6 hrs. C) The downregulation of BRCA1 is not prevented by the ATM inhibitor (KU-55933). Immunoblotting detection of BRCA1 in HeLa cells pre-treated with 10 µM KU-55933 for 30 min prior to 200 µM MMS treatment for 6 hrs. γH2AX and pChk2 detection were used as controls to confirm inhibition of ATM and/or ATR kinases. D) BRCA1 is downregulated in ATM-deficient human fibroblasts. Cells were treated with 200 µM MMS treatment for 6 hrs and harvested for immunoblotting. E) Depletion of ATR by RNAi does not prevent BRCA1 downregulation by MMS. Left, immunodetection of ATR following shRNA constructs transfection and puromycin selection. Right, ATR-depleted cells were treated with 200 µM MMS and harvested at the indicated times for immunoblotting. F) BRCA1 is downregulated in DNA-PKcs deficient cells. Glioblastoma DNA-PKcs proficient (MO59K) or deficient (MO59J) were treated with 200 µM MMS and harvested at the indicated times for immunoblotting.

Figure 5. The BRCT motif, but not the Ring finger domain, is required for MMS-induced BRCA1 downregulation.

A) Schematic view of the deletion constructs used in this study. B) HeLa cells were transfected with various expression constructs for BRCA1 and 2 days post-transfection, cells were treated with 200 µM MMS and harvested at the indicated times for immunoblotting to detect either endogenous BRCA1 or mutant forms using anti-BRCA1 or anti-GFP respectively. β-actin immunodetection was used as a loading control.

Figure 6. MMS induces a biphasic response of homologous recombination proteins.

A) Immunodetection of various BRCA1-associated and DNA damage/repair proteins following treatment of HeLa cells with 200 µM MMS. HeLa cells were treated with 200 µM MMS and harvested at the indicated times for immunoblotting. The star indicates the specific protein band detected with a given antibody. B) Foci formation of HR proteins following MMS treatment. HeLa cells were treated with 200 µM MMS and harvested at the indicated times for immunostaining. Bottom, cells with more than 10 foci were counted and the data are presented as percentage of cells with foci under each condition. The values represent the average ± SD of three independent experiments.

Figure 7. The proteasome mediates BRCA1 and BARD1 downregulation following MMS treatment.

A) HeLa cells were incubated with 20 µg/ml of cycloheximide alone or with 200 µM MMS (with or without cycloheximide) and harvested at the indicated times for immunoblotting. B) HeLa cells were pre-treated with 20 µM proteasome inhibitor MG132 for 30 min and then incubated with MMS in the presence of the inhibitor and harvested at 6 hrs for immunoblotting. C) Detection of BRCA1 ubiquitination following DNA damage in HEK293T cells. Following MMS treatment for 3 hrs, cell extracts were used for immunoprecipitation with an anti-BRCA1 antibody. A non-related polyclonal antibody was used as a control. The immunoprecipitates were used for immunoblotting using anti-BRCA1 or anti-ubiquitin antibodies. Densitometry quantification was conducted on BRCA1 and ubiquitin and the ratio ubiquitin/BRCA1 is shown.

Figure 8. BRCA1/BARD1 downregulation prevents recruitment of these proteins along with Rad51 to chromatin following MMS treatment.

A) HeLa cells were treated for 6 hrs with IR (10 Gy) or 200 µM MMS (with or without pretreatment with MG132). Chromatin from control or treated cells was prepared as described in material and methods and proteins were detected by western blotting. Histones were stained with coomassie blue to ensure equal loading. B) HeLa cells were treated as in panel A and harvested for immunostaining (left panel). Cells with more than 10 foci were counted and the data are presented as percentage of cells with foci under each condition (right panel). The values represent the average ± SD of three independent experiments. C) HeLa cells were treated as in A except that a permeabilization step was added before fixation and immunostaining (left panel). Cells with more than 10 foci were counted and the data are presented as percentage of cells with foci under each condition (right panel). The values represent the average ± SD of three independent experiments.

Figure 9. The DNA alkylating agent MMS induces the disassembly of BRCA1/BARD1/Rad51 from IRIF.

A) U2OS cells were pre-treated with IR (10 Gy) for 12 hrs and then with or without 200 µM MMS for 6 hrs and harvested for immunostaining. Bottom, cells with more than 10 foci were counted and the data are presented as percentage of cells with foci under each condition. The values represent the average ± SD of three independent experiments. B) Immunoblotting detection of BRCA1, BARD1 and Rad51 in HeLa cells pre-treated with IR (10 Gy) for 12 hrs and then left untreated or exposed to 200 µM MMS for 3 and 6 hrs.

Figure 10. Model indicating a biphasic response of the homologous recombination pathway induced by the alkylating agent MMS.

In response to MMS, human cells induce a signaling pathway that culminates in BRCA1/BARD1 downregulation. This prevents the unwanted assembly of the HR machinery at the early stage of the MMS-induced DNA damage response. At the second stage, recovery and assembly of HR proteins ensure the repair of DSBs generated by replication blocks.