Functional analyses of rare germline BRCA1 variants by transcriptional activation and homologous recombination repair assays

The 11 BRCA1 variants (listed in Tables 1 and 2) analysed in this study were selected from our previous study “BRCA1 Norway”, which is a collection of all BRCA1 variants detected in the four diagnostic genetic laboratories in Norway [21]. The variants were classified as VUSs by one or more of the Norwegian medical genetic departments and/or in the ClinVar database at the time of selection.

For the HDR assay, the plasmid pcDNA5-HBT-BRCA1-WT, hereafter called His-BRCA1 WT, was used as full-length template for generation of plasmid variants according to Supplementary Table 1. His-BRCA1 WT and the negative control variant plasmid pcDNA5-HBT-BRCA1-Leu1786Pro were kindly provided by Prof. Jeffrey Parvin (Ohio State University, Columbus, OH, USA) [22]. The fusion plasmid construct GAL4 DBD:BRCA1 (amino acids 1396–1863), hereafter called DBD-BRCT WT, used for the TA assay, was kindly provided by Alvaro N. A. Monteiro (H. Lee Moffitt Cancer Center and Research Institute, Tampa, FL, USA). This plasmid expresses the C-terminal part of BRCA1 (amino acids 1396–1863) including the BRCT domain fused to the GAL4 DNA Binding Domain (DBD). In addition, the pGAL4-e1b-Luc (firefly luciferase reporter), phRG-TK (Renilla luciferase reporter) and pcDNA3.1 (empty vector) were used in the TA assay.

The 11 BRCA1 variants of interest, as well as benign and pathogenic control variants, were introduced in both His-BRCA1 WT and DBD-BRCT WT plasmids by site-directed mutagenesis, using the QuikChange XL Site-Directed Mutagenesis Kit (Agilent). The sequences of the primers (Sigma-Aldrich) are presented in Supplementary table 1. The plasmids were purified using the NucleoBond Xtra Midi Plus Kit for transfection-grade plasmid DNA (MACHEREY–NAGEL) or the QIAfilter Plasmid Maxi Kit (Qiagen), following the manufacturer’s instructions. The presence of the variants, in addition to the whole BRCA1 insert, were verified by Sanger sequencing. Nomenclature was used according to the Human Genome Variation Society (HGVS) with the BRCA1 reference sequence NM_007294.3 [23].

Protein expression levels of the BRCA1 variants (expressed from both His-BRCA1 and DBD-BRCT plasmid constructs) were assessed in HEK293FT cells (Invitrogen). The HRR activity was assessed in HeLa-DR-GFP cells, kindly provided by Prof. Jeffrey Parvin (Ohio State University, Columbus, OH, USA) [22]. These cells are characterised by the presence of two differently inactivated copies of green fluorescent protein (encoded by GFP) integrated in a single locus in the genome (Fig. 1). In one of the GFP copies, the recognition element for the I-SceI endonuclease has been incorporated. The second GFP copy is truncated at both ends. Transfection of HeLa-DR-GFP cells with a plasmid encoding the I-SceI restriction enzyme will cause a double-stranded break in the first GFP copy, activating HRR which subsequently utilises the second inactive copy of GFP for repair. The repair results in formation of a functional GFP gene encoding an active protein, and the recombination can be detected through green fluorescent cells. To evaluate TA activity, HEK293 cells (ATCC) were used. Both HeLa-DR-GFP and HEK293 cells were cultured in Dulbecco’s modified Eagle’s medium (DMEM) supplemented with 10% fetal bovine serum (FBS) (both from Gibco) and 1.5% penicillin–streptomycin (Life Technologies). HEK293FT cells, used to evaluate protein expression levels, were cultured in DMEM high glucose GlutaMAX™ medium (Thermo Fisher Scientific) supplemented with 10% FBS and 1% PenStrep (Sigma-Aldrich).

Due to the large difference in protein size resulting from the construct expressing the BRCT domain fused to DBD (DBD-BRCT WT, 69 kDa) and the full-length plasmid (His-BRCA1 WT, 221 kDa), two different systems (Invitrogen and BioRad) were used for western blotting to evaluate protein expression of the BRCA1 variants. For both systems, HEK293FT cells were lysed in RIPA buffer (supplemented with cOmplete™, EDTA-free Protease Inhibitor Cocktail, Merck) 48 h post transfection, and centrifuged at 13 000 g for 10 min at 4 °C. Protein concentration of the supernatant was measured using the Pierce BCA protein assay kit (Thermo Fisher Scientific). For western blot analysis of the cell lysates transfected with the His-BRCA1 plasmid, 5 µg total protein was mixed with loading buffer and reducing agent (Invitrogen). After 10 min of denaturation at 70 °C, samples were separated on SDS-PAGE using 3–5% Tris–Acetate SDS-PAGE gels (150 V, 75 min). Subsequently, proteins were transferred to a nitrocellulose membrane (30 V, 60 min). To detect the BRCA1 protein, anti-BRCA1 (sc-6954, Santa Cruz) was used as primary antibody and m-IgGκ BP-HRP (sc-516102, Santa Cruz) as secondary antibody. Anti-β-Actin (sc-47778, Santa Cruz) antibody was used as a loading control. Proteins were visualised using SuperSignal™ West Pico PLUS Chemiluminescent Substrate (Thermo Scientific) and the ChemiDOC™ MP imaging system. The signals were quantified using the ImageLab Software (Biorad, version 6.0), and normalised by dividing the densitometric values from the His-BRCA1 band to the densitometric values of the corresponding actin band. Target protein and loading control signals were assured to be in their linear range. The average relative protein expression of the variants compared to the WT (set to 100%) were calculated.

The HDR assay was assessed as previously described [22, 24] with some modifications. HeLa-DR-GFP cells (40 000/well) were seeded in 24-well plates and grown in DMEM supplemented with 10% FBS without antibiotics. Cells were co-transfected at 60–70% confluency with the appropriate BRCA1 plasmids (His-BRCA1 WT or variants) in combination with specific 3’UTR siRNA (GCUCCUCUCACUCUUCAGU) to block endogenous BRCA1 mRNA, using a JetPrime:DNA ratio of 4:1 (Polyplus Transfection) following the manufacturer’s instructions. Twenty-four hours after transfection, the cells were reseeded into 6-well plates in DMEM supplemented with 10% FBS without antibiotics. After another 24 h, the cells were re-transfected with the appropriate BRCA1 variant plasmids in combination with siRNA and the plasmid encoding endonuclease (I-SceI). Three days after the second transfection, the presence of GFP cells was confirmed by confocal microscopy. Subsequently, cells were treated with trypsin (Sigma-Aldrich), collected, and resuspended in 1 ml of DMEM supplemented with 10% FBS without antibiotics. In total, 10 000 cells from each well were counted by flow cytometry (FC) using a BD LSRFortessa cell analyser (BD Biosciences), and the proportion (%) of GFP-positive (GFP⁺) cells was calculated. All BRCA1 variants were tested in triplicate in three independent experiments. The results were expressed as mean percentage of GFP⁺ cells ± standard deviation (SD) of the nine obtained values relative to the WT (set to 100%). As experimental controls, the following samples were included in each independent experiment: 1) untransfected cells to evaluate the background fluorescence from the HeLa-DR-GFP cells, 2) cells transfected exclusively with I-Scel and siRNA to ensure that there is no contribution from remaining endogenous BRCA1, 3) cells transfected with I-Scel, siRNA and empty vector (EV) as a vector control, 4) cells transfected with I-Scel and scrambled siRNA (sc siRNA) as control for the WT BRCA1 specific 3’UTR siRNA, and 5) cells transfected exclusively with I-Scel as control for the maximum expected value of GFP⁺ cells from endogenous BRCA1. Both BRCA1 specific 3’UTR and scrambled siRNAs were purchased from Eurofins.

A luciferase reporter assay was performed to measure TA activity of the DBD-BRCT WT and variants using the Dual-Luciferase Reporter Assay System from Promega. HEK293 cells were seeded in 24-well plates and co-transfected with the plasmid encoding the fusion protein DBD-BRCT WT/variants (0.36 μg), and the two reporter plasmids encoding the firefly luciferase (pGAL4-e1b-Luc, 0.36 μg) and Renilla luciferase (phRG-TK, 36 ng) using a JetPrime:DNA ratio of 2:1. After 24 h, cells were washed in cold PBS and lysed in 100 μl passive lysis buffer (Promega). The lysates were incubated on a rocking platform for 15 min, before clearing by centrifugation at 13 000 rpm for 10 min at 4 °C. The luciferase reporter assay was then performed in a white 96-well plate following the manufacturer’s instructions. Before measuring the luminescence signals, the dual injectors were washed 10 times with ultrapure water and primed with Luciferase Assay Reagent II and Stop & Glo reagent. The luminometer injection volume (Microplate luminometer, Centro XS3 LB 960, Berthold) was 100 μl as per manufacturer guidelines, and the injection speed was set to medium. The measured raw signals were normalised by conversion into firefly/Renilla ratios. The TA assay was performed in three biological replicates in independent transfections set-ups. Within a given biological replicate, each variant’s luminescence signal was measured in three technical replicates. The mean WT ratio was calculated for each biological replicate, and the mean percentage of TA activity for each variant relative to WT (set to 100%) and standard deviations were calculated.

The Alamut Software (Version 2.15, SOPHiA GENETICS) and the Human Gene Mutation Database (HGMD) professional 2022.1 (QIAGEN) was used for gathering information on the BRCA1 variants. Reinterpretation of the variants was performed based on new knowledge using the BRCA1/BRCA2 gene-specific criteria by CanVIG-UK [19, 20]. If there were evidence elements for both pathogenicity and benignity when combining the total CanVIG-UK criteria for a variant, the recommendations from Garrett et al. were applied [25].

HEK293FT cells were transfected with the His-BRCA1 WT/variants and DBD-BRCT WT/variant plasmids followed by western blotting. For both the full-length protein and the fusion protein variants, bands corresponding to the expected size for His-BRCA1 (221 kDa) and DBD-BRCT (69 kDa) were detected (Supplementary Figs. 1 and 4). After quantification, the expression levels of BRCA1 protein variants relative to the respective WT protein were calculated. As shown in Fig. 2A (DBD-BRCT) and 2B (His-BRCA1), the two variants p.Ala1708Val and p.Trp1718Ser were found to be expressed at levels similar to the pathogenic controls for both the fusion protein DBD-BRCT and the full-length BRCA1. The variant p.Phe1668Leu was expressed at intermediate levels, i.e. levels between the benign and pathogenic control variants, from both plasmid constructs. The six variants p.Leu1439Phe, p.Glu1535Lys, p.Met1628Ile, p.Leu1701Met, p.Pro1749Ala and p.Arg1835Gln were expressed at levels similar to the WT and/or benign controls in both DBD-BRCT and full-length BRCA1. The two variants p.Gly1709Arg and p.Lys1711Gln were expressed at intermediate protein levels in DBD-BRCT, but at levels similar to the pathogenic controls in His-BRCA1.

The HRR activity was assessed for the full-length WT protein (His-BRCA1 WT), EV, benign and pathogenic control variants, the variants of interest, and a selection of experimental controls by performing HDR assay as described in the methods. As expected, siRNA transfected in combination with I-Scel completely abolished endogenous HRR (Fig. 3, column 2) and showed only a few GFP⁺ cells similar to untransfected cells (Fig. 3, column 1), which has no HRR activation due to lack of I-Scel. This implicates that the endogenous BRCA1 was completely silenced, and that only the effects of the BRCA1 protein variants expressed from the plasmids were observed in all other samples. The cells co-transfected with I-SceI and scrambled siRNA showed a HRR activity of approximately 20% (Fig. 3, column 4) compared to the cells co-transfected with I-SceI, siRNA and WT BRCA1 (Fig. 3, column 6). The reduced number of GFP⁺ cells is due to differences in the amount of endogenous BRCA1 protein already present in the cells, compared to the much higher expression of the transiently transfected BRCA1 WT protein. The scrambled siRNA is expected not to affect the endogenous BRCA1, and one would expect a number of GFP⁺ cells comparable to that of cells transfected exclusively with I-Scel endonuclease (Fig. 3, column 5). However, our results show that the co-transfection of I-Scel and scrambled siRNA (Fig. 3, column 4) might impair the transfection efficiency of the endonuclease. This would however not affect the results for the variants of interest, as these are co-transfected with the same additional plasmid constructs as the WT (same total DNA input). The raw data output from flow cytometry is reported in Supplementary Fig. 2. Only one variant (p.Pro1749Ala) exhibited HRR activity comparable to the WT BRCA1 (101% of WT activity) and the benign control variants (Fig. 3). All other variants displayed HRR activity from 16–22% of the WT, similarly to the pathogenic control variants which showed HHR activity of 22–25% of the WT. Fluorescence microscopy analyses confirmed these results (Supplementary Fig. 3). Taken together, these data demonstrate that all but one of the variants impaired the BRCA1 HRR activity.

The TA assay was performed in three biological replicates (each consisting of three technical replicates) measuring each variant’s luminescence signal (Fig. 4). As expected, the TA activity of the pathogenic control variants was low compared to the WT protein (1–16%). The benign controls exhibited TA activity in the range from 46–97% compared to DBD-BRCT WT. Two variants (p.Ala1708Val and p.Trp1718Ser) showed TA activity levels similar to the pathogenic controls (16% and 2%, respectively), while the variants p.Gly1709Arg and p.Lys1711Gln displayed reduced TA activity levels of respectively 43% and 42% compared to the WT, but still similar to the benign control p.Arg1751Gln (46%). The remaining seven variants (p.Leu1439Phe, p.Glu1535Lys, p.Met1628Ile, p.Phe1668Leu, p.Leu1701Met, p.Pro1749Ala and p.Arg1835Gln) exhibited TA activities comparable to or higher than WT BRCA1 and/or benign controls ranging from 68 to 167%. Table 1 summarises the numeric values corresponding to all the functional assays.