Technical Comments

Independence of R/M/N Focus Formation and the Presence of Intact BRCA1

See allHide authors and affiliations

Science  07 Jul 2000:
Vol. 289, Issue 5476, pp. 11
DOI: 10.1126/science.289.5476.11

The BRCA1 gene product plays a multifunctional role in controlling genome integrity. Embryonic stem cells lacking BRCA1 are reportedly defective in transcription-coupled repair of oxidative DNA damage (1). Cells lacking intact BRCA1 also manifest a mitotic-checkpoint defect and are more likely than their normal counterparts to become aneuploid (2,3). The 1863 residue protein BRCA1 interacts with several nuclear polypeptides, including Rad51, a major participant in double-strand break repair and homologous recombination (4). In this respect, its function contributes to normal double-strand break repair (5, 6).

Zhong et al. (7) argued that a human tumor cell line (HCC1937) that synthesizes a truncated form of BRCA1 is defective in the normal function or functions of the Rad50/Mre11/NBS1 (R/M/N) complex (7). This complex forms nuclear foci (dots) after DNA damage (8,9); Zhong et al. argued that R/M/N focus formation is dramatically reduced in HCC1937 cells and can be restored by reconstitution with wild-type BRCA1(7). Here, we present data that run counter to that claim and suggest that R/M/N foci formation does not change after DNA damage in HCC1937 cells, irrespective of whether they produce ectopically expressed, wild-type BRCA1. These results make it difficult to conclude that BRCA1 is responsible for organizing radiation-induced R/M/N foci.

With one monoclonal (EE15) and one affinity-purified polyclonal (D29) NBS1 antibody, we found that wild-type human fibroblasts produce a 95-kD immunoreactive protein, which was absent from identically prepared extracts of an NBS1 −/− cell line (DST, Fig. 1A). This validates the NBS1 specificity of these antibodies. Moreover, in a pair of human fibroblasts, oneNBS1 −/− (GM07166) and the other a derivative infected with a human NBS1-encoding retrovirus, post-irradiation, D29-staining nuclear foci (dots) were observed only in the latter (Fig. 1B). Thus, any foci observed with this antibody are NBS1 specific. In keeping with the findings of Zhong et al. (7), endogenous R/M/N complexes coprecipitated with both endogenous BRCA1 and BARD1, a known BRCA1-associated protein (not shown). Furthermore, radiation-induced R/M/N foci colocalized with BRCA1 foci (not shown). Thus, these proteins likely interact in vivo and coexist within the same nuclear substructures after DNA damage.

Figure 1

Foci observed with EE15 and D29 antibodies are NBS1 specific. Immunostaining in this figure and in Fig. 2 was performed as described previously (4, 6), except that fixation was performed with 70% methanol and 30% acetone at −20°C for 15 min. (A). Antibodies that react with NBS1 can recognize endogenous NBS1 in a normal human fibroblast strain, but not in an NBS1 −/− cell line (DST). Five hundred micrograms of cell lysate from the normal diploid fibroblast strain, IMR90, and from DST NBS1 −/−B cells were immunoprecipitated with a polyclonal antibody, D29, raised against NBS1 and with identical quantities of preimmune serum from the same rabbit (denoted “Pre”). The immunoprecipitates and matched quantities of cell Lysate (20 μg, denoted “Lysate”) were immunoblotted with a monoclonal antibody, EE15, raised against NBS1. (B). An NBS1-reactive antibody specifically recognizes NBS1 foci inNBS1-reconstituted cells, but not inNBS1 −/− cells. Affinity-purified D29 was used to immunostain an NBS1 −/− fibroblast cell line, GM07166 (a and b) and an NBS1-reconstituted derivative, GM07166+NBS1 (c and d). The latter culture was reconstituted by infection with a human NBS1-encoding retrovirus. In GM07166, no nuclear staining was observed before (a) or 8 hours after (b) γ irradiation (18 Gy); in the NBS1-reconstituted derivative, nuclear staining was detected before γ irradiation (c), and NBS1 nuclear foci were observed 8 hours after γ irradiation (18 Gy) (d). Bottom panels (e through h) show 4′,6′-diamidino-2-phenylindole (DAPI) staining of the above-noted fields.

Figure 1

(A) NBS1 foci or granules are detected by antibody EE15 in HCC1937 cells independent of radiation. HCC1937 cells were untreated (a and b) or γ irradiated (c and d), fixed with 70% methanol and 30% acetone, and stained with antibody EE15 (green) according to the procedures of Wu et al. NBS1 staining detected by EE15 is similar in untreated or irradiated cells. Two groups of foci were detected. The first had a frequency less than 10 dots per nuclei and were in almost every nucleus regardless of treatment; this group of foci likely corresponds to staining in the nucleolar region. The second had a frequency of more than 10 foci per nuclei and exhibited similar distribution in untreated and irradiated cells. This result suggests that NBS1 nuclear focus formation detected by EE15 is independent of the radiation response. Cells fixed with 4% paraformaldehyde produced similar results (e through h). (B) Focus formation detected by commercially available antibodies from Novus Biologicals. NBS1-deficient GM07166 cells were untreated (i and j) or γ irradiated (k and l), fixed with 70% methanol and 30% acetone, and stained with NBS1 antibody (green). Uncharacterized nuclear dots were detected in this NBS cell line. HCC1937 cells were untreated (m and n) or γ irradiated (o and p) and stained with the Novus Mre11 antibody (green). (C) Similar experiments for HCC1937 cells fixed with 4% paraformaldehyde, using monoclonal antibody 12D7 (this lab) against Mre11. (D) Efficient irradiation-induced Rad51 focus formation in HCC1937 cells. Cells were fixed with 4% paraformaldehyde and stained with Oncogene Research Rad51 antibody Ab-1. Compared with untreated cells (u and v), a significant portion of the irradiated nuclei display bright and distinct Rad51 foci (w and x).

In HCC1937 cells, which do not reveal BRCA1 foci (Fig. 2A), there were abundant NBS1 foci 8 hours after exposure to 18 Gy gamma radiation, as revealed by staining with two NBS1 antibodies, D29 (Fig. 2A) and EE15 (Fig. 2B). These foci colocalized with Mre11 foci (Fig. 2B). Identical results were obtained with three different lots of HCC1937 cells. Compared with naı̈ve HCC1937, there was no post-irradiation change in the number of NBS1/Mre11 focus–containing cells (Table 1), the staining intensity of NBS1 and Mre11, or colocalization of NBS1 and Mre11 after BRCA1 reconstitution (Fig. 2, A and B).

Figure 2

R/M/N foci form in the BRCA1-deficient cell line, HCC1937. (A) NBS1 forms nuclear foci after γ irradiation in BRCA1-deficient HCC1937 cells. NBS1 immunostaining is shown in red [polyclonal Ab D29 (a, d, g, and j)], and BRCA1 is shown in green [monoclonal Ab SD118 (b, e, h, and k)]. In HCC1937 (top two rows), no BRCA1 nuclear foci were detected before or 8 hours after γ irradiation (b and e), but NBS1 formed abundant nuclear foci 8 hours after 18-Gy γ irradiation (d). One HCC1937 cell (denoted by white arrows in d and e) after γ irradiation is magnified (m and n) to show NBS1 nuclear focus formation (m) without detectable BRCA1 foci (n). Full-length BRCA1 synthesis was reconstituted in HCC1937 by transfecting HCC1937 with an expression vector for human BRCA1 and selecting a pool of stable, BRCA1-producing cells. The BRCA1-reconstituted cells (bottom two rows) revealed S phase–specific dots before irradiation (h) and post-irradiation foci 8 hours thereafter (k). NBS1 nuclear foci were observed at the same abundance in HCC1937 (d) and itsBRCA1-reconstituted derivative (j) 8 hours after γ irradiation. Two cells from the BRCA1-reconstituted HCC1937 population denoted by arrows (j and k) were magnified (o through s). Those identified with the white arrows (o and p) revealed both BRCA1 and NBS1 nuclear foci, most of which colocalized [merged image (q)]. The cell denoted by yellow arrows (r and s) lacked detectable BRCA1 nuclear foci (s), but revealed clear NBS1 nuclear foci (r). The third column (c, f, i, and l) contains DAPI-stained images of the aforementioned cells. (B) NBS1 and Mre11 irradiation-responsive foci colocalize in HCC1937 cells. NBS1 was immunostained with EE15 [green (a, d, h, and k)], and Mre11 was identified with the polyclonal antibody #59 [red (b, e, i, and l)]. DAPI staining is shown in the fourth column (c, g, j, and n). In HCC1937 (top two rows), NBS1 (d) and Mre11 (e) formed nuclear foci 8 hours after irradiation (18 Gy), and they colocalized [merged image (f)]. BRCA1-reconstituted HCC1937 (bottom two rows) was also detected by anti-NBS1 staining (k) and Mre11 staining (l), and foci also colocalized [merged image (m)]. One HCC1937 cell, denoted by white arrow (d through f), is magnified (o through q); oneBRCA1-reconstituted HCC 1937 cell, denoted by white arrow (k through m), is magnified (r through t).

Table 1

Percentage of BRCA1/NBS1/Mre11 focus formation after γ irradiation. Approximately 500 cells in each culture were analyzed for focus formation before and 8 hours after γ irradiation (18 Gy). Cells containing at least 10 visible nuclear foci were regarded as focus-positive. The table reports the percentage of focus-positive cells in the relevant cultures. Wild-type BRCA1 forms S phase–specific dots before γ irradiation that soon (∼1 hour) disperse and reform by 8 hours thereafter. Post-irradiation foci (NBS1/Mre11) were observed in a large fraction of cells 8 hours after radiation irrespective of the presence of BRCA1. IMR90, normal human fibroblasts; HCC, a BRCA1-deficient cell line; HCC+BRCA1, BRCA1-reconstituted HCC cells. Two right-hand columns show percentage of cells in which BRCA1 and NBS1 or NBS1 and Mre11 were colocalized, as revealed by coimmunostaining.

View this table:
Table 11

Frequency of Mre11 immunoreactive foci in untreated cells (untr.) and cells subject to 12 Gy γ irradiation (γ irr.), detected by rabbit polyclonal Mre11 antibody from Novus Biologicals. Paraformaldehyde, fixation with 4% formaldehyde as described in (1); methanol, fixation with 70% methanol and 30% acetone, as described in comment by Wu et al. Cells with at least 10 visible nuclear foci are viewed as focus positive (focus+).

View this table:
Table 12

Frequency of Mre11 and Rad51 immunoreactive foci in untreated cells (untr.) and cells subject to 18 Gy γ irradiation (γ irr.), detected by monoclonal Mre11 antibody 12D7 (this lab) and rabbit polyclonal Rad51 antibody Ab-1 (Oncogene Research Products, Cambridge, MA). Cells with at least 10 visible nuclear foci were viewed as focus positive (focus+).

View this table:

Taken together, the results described above, which are similar to those obtained by another laboratory (S. Elledge and D. Cortez, personal communication), run counter to the claim that BRCA1 plays a critical role in a signaling pathway that licenses R/M/N focus formation after DNA damage. They do, however, support the claim that these four proteins form a complex that, given its constitution and certain recent findings (5, 6), likely contributes to double-strand break repair and, possibly, other DNA damage responses.

  • * Present address: Department of Oncology, Guggenheim 13, Mayo Clinic, 200 First Street SW, Rochester, MN 55905, USA


Response: Zhong et al. (1) found that BRCA1 coimmunoprecipitated with the endogenous R/M/N complex; that γ–irradiation induced BRCA1 foci assembled 8 hours after irradiation and colocalized with R/M/N foci; and that inBRCA1-mutated HCC1937 cells, BRCA1 focus formation was abrogated and could be restored by ectopic expression of BRCA1. These findings, subsequently confirmed by two other groups (2,3), suggest that BRCA1 forms a functional complex with R/M/N in vivo that may be crucial for DNA double-strand break repair and DNA damage response (1).

Wu et al. conclude, based on immunofluorescence analyses using NBS1 antibodies EE15 and D29, that the formation of γ–irradiation induced R/M/N foci is not affected in HCC1937 cells. To address this discrepancy directly, we performed immunofluorescence experiments using the same antibodies, obtained from Wu et al. HCC1937 cells were grown on coverslips; irradiated at a 12-Gy dose; fixed with 70% methanol and 30% acetone as suggested by Wu et al. and, in a separate experiment, with 4% paraformaldehyde as described in (1); and immunostained with monoclonal antibody EE15 supernatant. Surprisingly, we observed that EE15-immunoreactive foci could be efficiently detected in the nuclei of both untreated and γ-irradiated cells (Fig. 1A). The presence of EE15-immunoreactive foci in the nuclei of untreated cells may reflect the participation of NBS1 in an uncharacterized cellular process independent of the radiation response, and precluded quantitative analysis of irradiation-induced NBS1 focus formation, because irradiation-induced foci could not be effectively distinguished from irradiation-independent foci. For reasons that we cannot explain, the D29 antibody repeatedly failed to reveal any immunoreactive foci in a variety of cells lines that we studied, although it did detect NBS1 protein by immunoblot analysis. We therefore could not evaluate this antibody further.

We performed similar experiments using commercially available antibodies directed against NBS1 and Mre11 (Novus Biologicals, Littleton, CO). The Novus NBS1 antibody could detect nonspecific nuclear dots in NBS1-deficient cells; the Novus Mre11 antibody revealed minimal induction and quantitatively fewer irradiation-induced immunoreactive foci in HCC1937 cells than in T24 and IMR90 cells (Fig 1B; Table 1). This result is consistent with those of analyses using our own Mre11 monoclonal antibody, 12D7 (Fig 1C; Table 2). The Novus Mre11 antibody exhibited a greater propensity to detect nuclear dots in both untreated and γ-irradiated cells, especially when the cells were fixed with 70% methanol. That suggests that variations in foci detection may derive from differences in antibody sources or immunofluorescence fixation procedures.

Although BRCA1-positive irradiation-induced foci were abrogated and R/M/N irradiation-induced foci were reduced in HCC1937 cells, as originally reported (1), we observed no significant reduction in the formation of irradiation-induced foci positive for Rad51, another protein that interacts, albeit indirectly, with BRCA1 (4–6), (Fig 1D; Table 2).

In sum, our work suggests that HCC1937 cells are radiosensitive and that reintroduction of wild-type BRCA1 into these cells restores their radioresistance (1, 2). That finding, in turn, suggests that BRCA1 is crucial for the DNA damage response. Although discrepant observations persist with respect to whether R/M/N irradiation-induced focus formation is proficient in HCC1937 cells, the R/M/N foci detected in these cells are likely to represent functionally impaired repair entities. Clearly, BRCA1 plays a role in DNA damage repair.


Navigate This Article