Tata Memorial Centre, Advanced Centre for Treatment, Research and Education in Cancer, Kharghar, Navi Mumbai - 410 210, India
Received date: October 16, 2012; Accepted date: October 27, 2012; Published date: October 29, 2012
Citation: Badgujar DC, Sawant U, Mahadik H, Gadewal N, Varma AK (2012) Pathogenicity of Mutations Discovered in BRCA1 BRCT Domains is Characterized by Destabilizing the Hydrophobic Interactions. J Cancer Sci Ther 4:386-393. doi:10.4172/1948-5956.1000172
Copyright: © 2012 Badgujar DC, et al. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Visit for more related articles at Journal of Cancer Science & Therapy
Background: Breast and ovarian cancer are the most common cancers among hetrogenetically diversified women. It is quite difficult to categorize the population at a high risk of breast cancer using peer genetic information because one particular mutation can be found in the same or in different families. Several mutations have been discovered across the full length of BRCA1 gene, and categorizing their pathogenicity is a major challenge. Carriers of BRCA1 mutations have an increased risk of developing cancer. In the breast cancer database BIC, approximately 1,500 genetic variants have been reported. It is very difficult to characterize each of the reported mutations. Given the complexities in characterizing the mutations, we decided to investigate functional basis associated with the mutations, rather than looking at each mutation.
Materials and methods: BRCA1 BRCT domains were cloned, expressed, and purified using e-coli bacterial expression system. Mutations were generated using site-directed mutagenesis techniques, and all the mutations were sequence verified. The secondary structure of the mutant was characterized by Circular dichroism (CD) and Fluorescence spectroscopy. Molecular dynamics simulations were performed using Desmond software. Hydrophobic interactions and hydrogen bonding of docked molecules were compared using the LigPlot program.
Results: Genetic mutations were discovered throughout BRCA1, and most of the pathogenic mutations were buried in the hydrophobic core and destabilized the BRCA1 BRCT domain. This unstable BRCT domain destabilized the full-length BRCA1, resulting in a loss of function. We conclude that the pathogenicity of each of the mutations in the BRCT domain can be categorized on the basis of its ability to destabilize the hydrophobic interactions. Although such instability is not sufficient to predispose someone to cancer, it provides a basis for formulating a concept for genetic counseling and targeted therapy.
BRCA1 BRCT; Pathogenicity; Hydrophobic interactions; Genetic counseling; Targeted therapy
BRCA1, a breast and ovarian cancer gene, comprises 1,863 amino acids, and it has multifunctional cellular domains involved in the maintenance of genomic integrity [1,2]. Genetic mutations discovered in the BRCA1 gene from a large cohort of patients are helping in clinical management [2-4]. Somatic mutations are present in most patients. Some patients also possess germ-line mutations, which they inherited from their parents . Genetic testing for cancer predisposing germline mutations can help clinicians to select treatment options when performing surgery, radiotherapy, and chemotherapy . Providing information on the importance of genetic testing to high-risk women can help in detecting breast and ovarian cancer at an early stage and, therefore, increase the probability of recovery . At present, an important challenge in genetic counseling is to evaluate the pathogenicity of mutations discovered throughout of the BRCA1 gene. The risk of developing cancer due to carrier of BRCA1 mutations is 40-87% for breast cancer and 16–68% for ovarian cancer . Albeit, there are several other factors such as lifestyle, environmental changes, and hormonal factors that may influence the risk of cancer associated these mutations.
The management of mutations discovered across the world in cancer patients is also a major challenge. Human mutation databases compile information on mutation under one umbrella, e.g., http:// www.biologie.uni-hamburg.de/b-online/library/genomeweb/ GenomeWeb/human-gen-db-mutation.html. The database of the Breast Cancer Information Core (BIC) also serves as a repository of genetic mutations and variants observed in the BRCA1 and BRCA2 genes: http://research.nhgri.nih.gov/bic/. Over 1,500 genetic variants have been reported in the BRCA1/2 genes. These are classified as either known deleterious mutations, polymorphisms without clinical significance, or sequence variations of unknown significance. Several groups have deposited BRCA1/2 mutations in the BIC’s database and categorized their pathogenicity. However, a significant number of reported mutations need further evaluation . Considering the importance of mutations in targeted therapy and drug discovery, we decided to determine the possible locations of pathogenic mutations and evaluate these using multimodel-based structural, in-silico, and biophysical approaches. The challenge is to identify the mutations that cause breast cancer and develop a possible diagnostic marker for the disease. Investigators across the world have different views on mutation-based targeted therapy and biomarkers [10,11]. There are several messages to the scientific community but challenges are in studying cancer biology . For example, are we targeting the correct target in cancer therapy and are we focusing on the mutation-associated function of the protein. Such questions are a direct result of the number of mutations discovered throughout the 1,863 amino acids of the BRCA1 gene and the approaches various groups have adopted to explore these mutations left several unanswered questions.
Another major question is whether mutations discovered in cancer patients should be used for medical management and whether there are any functional aspects associated with these mutations that can help in targeted therapy or finding biomarkers. To answer these questions, we have evaluated reported structures of BRCA1 BRCT extensively and compared the structure-function association with already reported pathogenic mutations. We found that most of the pathogenic mutations in the BRCA1 BRCT domains are located in the hydrophobic core of BRCT. Furthermore, we evaluated the pathogenicity of the mutations using in-vitro and in-silico approaches.
BRCA1 is a well-known tumor suppressor gene, which performs cellular functions through its interaction with proteins participating in DNA repair, cell cycle checkpoint control, and ubiquitination [13,14]. BRCA1 has a ubiquitin-interacting motif at its N-terminus, a DNA-binding domain in the middle, and a transcriptional activation BRCT domain at the C-terminus. The C-terminus of BRCA1 has two tandem repeats of BRCT, each comprising ~90–100 amino acids. The protein data bank (www.pdb.org) lists 124 structures for BRCT. Structurally, both the repeats of BRCA1 BRCT domains are packed together in a head-to-tail manner . There are ~100 BRCT-containing proteins, and some of the BRCTs’ repeats adopt a similar conserved conformation, with a central β-sheet stabilized by four parallel β-sheets (β1–β4) and three α–helices (α1–α3) on each arm of the repeats (Figure 1). BRCA1 BRCT adopts similar secondary structure conformations in the C-terminal of XRCC1 , 53 BP1 , MDC1 , and BARD1 . However, the rmsd of each amino acid reveals that the conserved region of the central β-sheets and the α–helices (α1 and α3) are super-imposable and stabilized by a number of hydrogen bonds and hydrophobic interactions, whereas others such as β1–α1, β2–β3, and β3–α2 and the location of α2 in the central β-sheets are not super-imposable. These variations highlight the fact that the BRCTs of different proteins may have different binding motifs unlike BRCA1 BRCT . Most of the BRCTs do not have much sequence similarity but share three-dimensional structural folding and functions.
The interaction of the C-terminal domain of BRCA1 with phosphopeptide-containing protein targets is well studied . The crystal structure of BRCT repeats that bind to phosphoserine-specific peptides like DNA helicase BACH1 [21-23], CtIP (CtBP-interacting proteins) , and ACC1 (acetyl-CoA carboxylase)  through its conserved binding pocket has been determined. In all reported complex structures of BRCA1 BRCT, binding to phosphopeptides promiscuous about the role of BRCT in the repair of DNA double-strand breaks, cell cycle checkpoint control, and tumor suppressor function. Earlier reported structures of BRCT complexes with BACH1, CtIP have predicted that the pocket near the N-terminus of BRCT harbors phosphoserine and that the interface between repeats has a hydrophobic groove, which stabilizes phenylalanine [21-24,26,27]. Phosphoserine at the zero position (pS0) and phenylalanine at the plus three position (F+3) in the consensus sequence X-pS0-X-X-F3-X, (X-can by any amino acids) binds to BRCT at the dimer interface (BRCT1 and BRCT1) in a two-sided manner. In most BRCT structures, the phosphate group of phosphoserine makes salt bridges and hydrogen bonds with the amide terminus repeats of BRCT. Phenylalanine from the C-terminus of the peptide stabilizes the hydrophobic pocket located at the interface of the repeats and the C-terminus of BRCT. Recent results also predict that Abraxas, which has the sequence motif X-pS-X-pS-X-X-F, binds to BRCA1 BRCT through the mutual exclusion of binding conformation of earlier reported structures of BACH1 and CtIP . This indicates that a dual phosphopeptide recognition site is also present in the BRCT domain and that it may help in maintaining its structural integrity .
To understand the molecular mechanism of the conserved region of the phosphopeptide’s recognition site and the hydrophobic core of BRCTs in evaluating the pathogenicity of mutants for clinical management and targeted therapy, a complete structural and biophysical comparison of BRCA1 BRCT and clinically significant mutants is presented here.
Gene- cloning and site-directed-mutagenesis
The gene encoding BRCA1 BRCT domain (1646-1859) was PCR amplified using BRCT construct in pGEX-kT( generous gift from John A A Ladias). NdeI site was added in forword primer (5’-ATCATATGGCC ATG GTC AAC AAA AGAATGTCC-3’) and BamHI in reverse primer (5’-TAGGATCC TCACTA GGG GAT CTGGGGTAT CAGG-3’). The amplicon was first cloned in pJET (Thermo Scientific, USA) blunt end ligation vector and later sub-cloned in pET3a Vector using T4 DNA ligase from NEB. Site directed mutagenesis for H1686Q (Forward: 5’-CTGAAGAGACTACTCAGGTTGTTATGAAAACAG-3’ Reverse: 5’-CTGTTTTCATA ACAACCTCAGTAGTCTCTTCAG-3’, P1749R (Forward: 5’-GAAACCACCAAGGTCGAA AGCGAGCAAGAG-3’ Reverse: 5’-CTCTTGCTCGCTTTCGACCTTGGTGGTTTC-3’),
S1715 R (Forward: 5’-GAAAATGGGTAGTTAGGTATTTCTGGGTGACC- 3’, Reverse: 5’-GGTCACCCAGAAATACCTAACTACCCATTTTC- 3’) and M1775R Forward: (5’-CCCTT CACCAACAGGCCCACAGATCAAC- 3’) Reverse: (5’-GTTGATCTGTGGGCCTGTTGGTGAAGGG- 3’) M-1775-K Forward: (5’-CCCTTCACCAACAGGCCCACAGATCAAC- 3’) Reverse: (5’-GTTGATCTGTGGGCCTGTTGGTGAAGGG- 3’) was carried out by amplification using pfu polymerase and digestion of template DNA by DpnI. All site-directed- mutations were confirmed by DNA sequencing.
Protein expression and purification
The BRCA1 BRCT variants (H1686Q, P1749R and S1715R) was incorporated in BL21 (pLysS) for overproduction of protein. Cells were grown in LB medium containing 100 mg/ml Ampicillin and 34 mg/ml chloramphenicol at 37°C till OD reaches A600 between 0.6-0.8, then the culture was induced by 0.4 mM IPTG at 18°C, incubated at 250 rpm for 16 hours. All the further protein purification steps were carried out at 4°C.The induced culture were harvested in Sorvall SLC-3000 rotor at 6000 rpm for 10 min. harvested induced cells was resuspended in lysis buffer (20 mM Sodium phosphate pH- 6.0, 50 mM NaCl, 1 mM EDTA, 1 mM βME, 1 mM PMSF and 0.5% Triton X-100) and sonicated for 5-8 times with 50 duty cycles. Cell lysate was centrifuged at 18000 rpm for 30 min. BRCA1 BRCT Mutants S1715R and P1749R were not much soluble and folded properly. However, we could proceed with BRCA1 BRCT variants H1686Q for further purifications. Cleared bacterial cell lysate of BRCA1 BRCT H1686Q was bound to SP sepharose resins pre -calibrated by buffer B (20 mM Sodium phosphate pH- 6.0, 50 mM NaCl, 1 mM EDTA, 1 mM βME, 1 mM PMSF). Bound protein was eluted with a gradient of NaCl (100 mM – 800 mM).The fractions which showed proteins were pooled and concentrated till 2 ml and buffer exchanged using G-50 desalting column to sodium borate buffer of pH 8.7. Variant was then applied on Q sepharose calibrated by 10 mM Sodium borate buffer. Bound protein was eluted using NaCl gradient (100 mM-500 mM). Protein expression and purification was monitored by loading the fractions on SDS PAGE.
Gel filtration chromatography
Gel filtration chromatography was done using superdex-75 Hiload column from GE Healthcare. The elution fractions from Q sepharose which shows protein were pooled down and concentrated till 2 ml and applied on pre-calibrated superdex-75 column using buffer (10 mM Sodium borate pH 8.7, 100 mM NaCl). Fractions were collected by monitoring the absorbance at 280 nm.
Circular dichroism (CD) spectroscopy
Circular dichroism (CD) spectra were collected using JASCO J-715 spectropolarimeter (Jasco, Easton, MD), in the far-UV region (180–260 nm). The JASCO J-715 is well equipped with a Jasco PTC 348 WI temperature controller and sealed quartz cuvettes. In the far-UV region (180–260 nm) BRCA1 BRCT (in buffer 10 μM solution of BRCA1 BRCT and BRCT H1686Q (10 mM Borate buffer, pH 9.0, 300 mM NaCl) was loaded onto a 0.1 cm path-length quartz cuvette (Helma). Seven spectra were accumulated and averaged for each experiment, with a resolution of 1 nm at a scan speed of 50 nm/min and a response time of 1 s. Buffer blank spectra, obtained at identical conditions, have been subtracted from the raw data. The results in all experiments have been expressed as ellipticity [θ] (°cm2 dmol−1).
Emission spectrums were collected on HORIBA FL3-21 spectrofluorometer. The variant and wild type BRCA1 BRCT of 10 μM concentration taken in 1 mm cuvette was excited at 280 nm and checked its emission maxima.
Align-GVGD was carried out using multiple sequence alignment available for BRCA1 from human to frog. The grade of pathogenicity is categorised in different classes beginning from C0, C15, up to C65. Any mutation with category C0 is less likely and C65 most likely pathogenic.
Mutational analysis was performed by building the four mutants (H1668Q, S1715R, M1775R and P1749R) on reported structure BRCA1 BRCT (pdb id: 1Y98), to study their molecular interactions by MD Simulation using Desmond 2010 software package . Optimized Potentials for Liquid Simulations (OPLS) all-atom force field was used to analyze model stability. The protein structures were solvated with the Monte Carlo simulated SPC water model using orthorhombic box. The system was then neutralized replacing water molecules with sodium and chloride counter ions. Subsequently, a maximum of 2,000 energy minimization steps were carried out for all complexes using a steepest descent algorithm with a tolerance of 1000 kJ mol−1 nm−1. A twin-range cutoff was applied to long-range interactions (1.0 nm for van der Waals and electrostatic interactions) using Particle Mesh Ewald (PME). These minimized and solvated systems were considered reasonable structures in terms of geometry and solvent orientations and used in further simulation steps. Equilibration MDS for both temperature (300 K) and pressure (1 atm) was carried out for 100 ps. These pre-equilibrated systems were subsequently used in the 1ns production MDS with a time-step of 2 fs. Structural coordinates were saved every 1.2 ps and analyzed using the analytical tools in the Desmond package. The lowest energy structures were obtained for each mutant structure and molecular interactions of wt and mutant were plotted using Ligplot .
BRCA1 is one of the most studied genes for breast and ovarian cancer. Several investigators have tried to focus on different domains of BRCA1 to characterize its associated functions . Structurally, the C-terminus of BRCA1 has been the focus of much attention to evaluate the pathogenicity of mutants discovered in a large cohort of patients and to facilitate genetic counseling and targeted therapy [30-32]. Most work has concentrated on the conserved phosphoserine-binding motif located at the interface of the BRCTs. However, the association of BRCA1 BRCT with doubly phosphorylated Abraxas indicates the presence of other domains within the BRCT that require further evaluation . Thus, we attempted to study the phosphopeptide binding and the hydrophobic pocket of BRCT wild-type and reported pathogenic mutants. Most of the mutations are categorized as high-grade pathogenic with A-GVGD prediction grade of C65. Although H1686Q has been reported to be pathogenic , its A-GVGD value GV (0.00), GD (24.08) and prediction class C15 suggests that this is less likely pathogenic . All these A-GVGD values are in agreement with results reported in this paper. Most of the pathogenic mutations destabilized the hydrophobic core of the BRCA1 BRCT domain, which may prove useful in designing small molecules for targeted therapy.
Structural basis of the BRCT domains
Mostly, BRCT domains are found in proteins involved in the mediation of DNA damage repair, cell cycle checkpoint control, and tumor suppressor function. There are several deletion/mutations in the BRCT domain that give rise to deleterious/truncated mutants, which, in turn, lead to breast and ovarian cancer. BRCA1, BARD1, and MDC1 have tandem repeats of BRCTs (Figure 1) whereas, PARP1, XRCC1 and DNA ligase III have a single domain. The tandem repeats of BRCA1 BRCT are known to interact with phosphospecific binding partners, where phosphoserine occupies the first N-terminus domain of BRCT and, mostly, phenylalanine occupies the second repeat at the hydrophobic core of C-terminus BRCT . The phospospecific binding domains are present on only selected tandem repeats of BRCT. Despite having the same three-dimensional structural folding, there are no reported binding partners for the single domain of BRCT, but tandem repeats of BRCA1 BRCT and MDC1 BRCT have phosphospecific binding domains. BARD1 BRCT does not have any structurally reported binding partners, whereas in-silico and in-vitro results indicate binding between BARD1 BRCT and its phosphospecific target  (Figure 2A & 2B). Recently, reports suggest that Abraxas, which is a double phosphopeptide, binds to BRCA1 BRCT . All these results have lead to much confusion in the scientific community in terms of translational research. Although the three-dimensional structural folding of BRCA1, MDC1, and BARD1 is very similar, the sequences holding the binding domains are drastically different. Furthermore, the functional aspects of BRCT are characterized by direct interactions of these domains with several cellular binding partners during cell cycle progression. Looking at the tandem repeat structure of BRCT and the single domain of BRCT, it can be concluded that not only protein–protein interactions, but also intramolecular interactions play a vital role in determining the protein’s functions.
Figure 2: Superimposed structure of BRCA1 BRCT of phospho peptide CtIP complex (pdb id: 1Y98) (A) with BARD1 BRCT (pdb id: 2NTE) (B) superimposable amino acids of BARD1 BRCT and (C) MDC1 BRCT (pdb id: 2ETX), Conserved hydrophilic environment around phosphor-Serine of BRCA1 BRCT (pdb id: 1Y98), BARD1BRCT (pdb id: 2NTE) and, MDC1 BRCT (pdb id: 2AZM), dotted red lines are phosphate of BARD1 BRCT. BRCA1 BRCT is colored in pink, peptide CtIP-orange, BARD1 BRCT in marine, MDC1 BRCT is colored in magenta.
It is well known that weak intramolecular interactions play an important role in protein folding . Furthermore, it is well established that hydrogen bonding and hydrophobic interactions play a major role in protein stability . Within the available complex structure of BRCT, we analyzed hydrogen bonding and hydrophobic interactions for mutations discovered to predispose patients to cancer. However, how these weak interactions are associated with the pathogenicity of mutants and targeted therapy has not been explored earlier.
Phosphospecific binding pocket
There are several BRCT complex structures available in the protein data bank (www.pdb.org). The phosphoserine binding pocket is conserved and stabilized by a number of weak inter-and intramolecular interactions. The C-terminal domain of BRCA1 interacts with cellular proteins like phospho-BACH1(DNA helicase BACH1), phospho-CtIP (CtBP interacting proteins) [23,24], and phospho-ACC1 [39-41] through its conserved binding pocket in a structurally diverse manner. Although the pS-X-X-F motif is more or less conserved, the different conformational orientation of the N-terminus and the C-terminus of the binding partners bestow a multifaceted role of the BRCT repeats. The BACH1/CtIP phosphate group of Ser interacts with Lys 1702, Ser 1655, and Gly 1656 and stabilizes this pocket through hydrogen bonding and hydrophobic interactions. The hydrophilic environment around phosphoserine plays a very important role in stabilizing this pocket. Water-mediated hydrogen bonds hold BRCT1 and BRCT2 together (Arg 1699 and Glu 1836 are held together in the same way as the phosphobinding pocket pS-X-X-F holds BRCT1 and 2 together). The N-terminal residues of the phosphospecific peptide bind to the N-terminus of the protein, and the C-terminus of each peptide binds to the C-terminus of the protein, contributing to stabilizing the binding pocket .
The overall location and orientation of the phosphopeptide binding motif of MDC1 BRCT with γ-H2AX is similar to that in BRCA1 BRCT (Figure 2C). The phosphoserine is largely buried in the hydrophilic environment of the protein. γ-H2AX phosphoserine, located at position 139, forms a hydrogen bond with Thr 1898, Gly 1899, and Lys 1936. This binding is further stabilized by intramolecular hydrophobic interactions of the residues Gln 140 and Ala 138 of γ-H2AX. Both the proteins bind the phosphate moiety through direct interactions with side- and main-chain atoms of the structurally conserved residues: Lys 1936, Thr 1898, and Gly 1899 are found in MDC1 BRCT, whereas Thr 1898, Lys 1936, and Arg 1933 are found in BRCA1 BRCT .
The sulfate ion is accommodated in the same location in BARD1 BRCT as the phosphoserine of BACH1/CtIP in the BRCA1 BRCT complex structure, and shows similar kinds of interactions and atomic orientations. The sulfate group forms hydrogen bonds with Ser 575, Thr 617, and Lys 619, which is an anticipated phosphospecific binding pocket of BARD1. The sulfate ion might have enthroned its position during crystallization . On the other hand, the active and proper folded dimer structure of BARD1 BRCT is situated at the phosphospecific binding motif, which inhibits the binding with sequences of pS-X-X-F. The first molecule of the BARD1 dimer (pdb id: 2NTE, chain A, residues from 686-HHKFTGWLY-678) can be superimposed on the phosphospecific CtIP peptide, which is bound to the BRCA1 BRCT (pdb id: 1Y98) structure (Figure 2A & 2B). This may either be an active dimer of BARD1 BRCT that does not have the same consensus sequence binding motif as BRCA1 BRCT, or it may have a monomeric structure at low temperature, supporting the concept of binding with BRIP1 . Structural analysis of BARD1 BRCT at low temperature may help in unraveling the exact molecular mechanism associated with the BARD1 BRCT phosphospecific binding motif.
Hydrophobic phenylalanine binding pocket
The hydrophobic phenylalanine at the +3 position located at the peptide side contains the BRCA1 BRCT repeats. Phe (+3) is fully buried in the hydrophobic environment, which comprises Leu 1701, Phe 1704, Met 1775, and Leu 1839 (Figure 3A). Binding between phenylalanine and BRCA1 BRCT has been reported to be associated with the transcriptional activation function. It is very important to mention that the supramolecular arrangement due to Phe (+3) binding plays a very important role in characterizing the function of the associated domains. Phe (+3) forms hydrophobic interactions and stabilizes the domains. It is very well known that the transactivation function of BRCA1 BRCT is due to binding with the phosphospecific BACH1  and CtIP . However, binding does not occur unless the aromatic ring of Phe is at the +3 position . This provides strong support for the ability of weak intermolecular interactions to shed light on the functional characteristic of proteins. The binding motif is conserved for all the reported peptide complexes, and Arg1699, Asn 1774, Met 1775, Phe 1704, Leu 1701, and Arg 1835 all stabilize the hydrophobic pocket of BRCT. Hydrophobic interactions and hydrogen bonding at the binding interface stabilize the genomic integrity of BRCA1 . Comparing the structure of BRCA1 BRCT with MDC1 BRCT, we found that the Phe +3 hydrophobic pocket is conserved (Figure 3A & 3B). However, thus far, there have been no reports of a complex structure for BARD1 BRCT containing the consensus sequences of the BRCA1 BRCT binding peptide (Figure 3C). This suggests that the mutations that occur at the hydrophobic core of BARD1 BRCT either have different pathological significance than BRCA1 BRCT or may bind to other consensus sequences. The hydrophobic pocket of BARD1 is stabilized by Ser 616, Met 621, His 685, His 686, and Ile764 . Interestingly, MDC1 BRCT senses tyrosine at the +3 position . 53 BP1 BRCT (pdb id: 1GZH) does not have phospho-Serine at putative pSer position (Figure 3D). This indicates that even if the three-dimensional folding is the same, the genomic integrity of each BRCT may differ.
Hydrophobic interactions and mutagenesis of pathogenic residues
It is rather difficult to characterize the large number of mutations available in cancer populations using structural biology and bioinformatics tools. In the C-terminal of BRCA1 BRCT, ~100 mutations are reported in the BIC’s database . Some of these are predicted to be pathogenic mutations with no transcription, and some are characterized but have unknown functions. Most reported pathogenic mutations Met 1775 Arg (Figure 4 i & ii), Met 1775 Lys of BRCA1 BRCT diminished the binding capacity between BRCA1 BRCT and BACH1, CtIP by destabilizing the BRCT domain. Most strikingly, His 1686 Gln is a reported novel pathogenic mutation that has different binding affinity to phosphopeptide target proteins than the BACH1 and CtIP . These findings leave many questions unanswered regarding the functionality of this gene, which predispose high-risk individuals to cancer. Looking at different levels of accuracy in predicting the pathogenicity of mutations to provide novel concepts, which may help in predicting mutation-based biomarkers or designing novel drug leads. We have performed a multimodel in-silico, in-vitro, and biophysical based study to answer major challenging questions pertaining to human genomics.
Proline 1749 to arginine: A structural change in proline (P) from a medium-sized, rigid, and hydrophilic residue to a large-sized and positively charged basic amino acid arginine (R) has been reported to be associated with ovarian cancer, with unknown pathological significance. Proline at position 1749 in BRCT (pdb: 1y98) is >15 Å away from pSer(0) and F(+3) of CtIP and surrounded by hydrophobic interactions of Lys 1750, Glu 1735, Gln 1747, Gly 1748, and Val 1736; a hydrogen bond with Arg 1753 is also present (Figure 4iii). However, mutant BRCT1749 arginine has re-oriented the hydrophobic pocket by forming hydrophobic interactions with Lys 1750, His 1746, Val 1713, and Ile 1707 and by forming hydrogen bonds with Gly 1738, Asp 1739, Tyr 1707, Ala 1843, and Gly 1748 (Figure 4 iv). The structural changes around the hydrophobic environment point to the pathogenicity of the proline to arginine mutation.
Histidine 1686 to glutamine: This mutation has been discovered in Italian breast and ovarian cancer families . It is located on exon 17 of BRCA1 and characterized as pathogenic. Structurally, BRCA1 His 1686 is stabilized by four hydrophobic interactions of Met, 1650, Met 1652, Val 1687, and Trp 1712 and four hydrogen bonds of Met 1650, Thr 1685, Val 1653, and Glu 1731 (Figure 4v) . However, BRCA1-1686Q forms a single hydrophobic interaction with Val 1687 and four hydrogen bonds with Thr 1685, Val 1653, Ser 1651, and Glu 1731 (Figure 4vi). These characteristics differentiate it from the wild-type. It also exhibits significant differences in its three-dimensional structure and genomic stability around the hydrophobic core, pointing to the pathogenicity of the mutation.
Serine 1715 to arginine: Ser 1715 is located on exon 18 of BRCA1. It has been reported on one occasion in a Danish patient, with unknown clinical significance . Mutation from a small size and polar Ser (S) to a large size and basic Arg (R) on that BRCA1-Ser1715 is ~13Å away from pSer(0) and 19 Å from Phe(+3) in BRCA1 CtIP (pdb id: 1Y98) complex structure. There is not much conformational changes occur due to mutation (Figure 4vii & viii). It is stabilized by three hydrophobic interactions Val 1714, Tyr 1716, and Phe 1695 and three hydrogen bonds with Phe 1734, Trp 1718, and Thr 1691. BRCA1-Ser 1715 is located at the head of parallel β-sheets and on the tail of the α-helix. A superimposed model of BRCA1-S1715R destabilized the turn site, moving away from the CtIP binding site. The side chain of Arginine is destabilizing the CtIP binding sites and resulting comparatively less association with the BRCT binding partners.
Figure 4: Schematic representations of weak intramolecular interactions between BRCA1 BRCT wt and reported pathogenic mutations (i) BRCA1 BRCT M1775 (ii) BRCA1 BRCT M1775R (iii) BRCA1 BRCT P1749 (iv) BRCA1 BRCT P1749R (v) BRCA1 BRCT H1686 (vi) BRCA1 BRCT H1686Q (vii) BRCA1BRCT S1715 (viii) BRCA1 BRCT S1715R. for clarity in Ligplot, the mutants are shown as B chain in this figure only.
Comparing the in-vitro results, we used FPLC purified BRCA1 BRCT (1646-1859) His 1686 Gln mutation (Figure 5A & 5B), however Ser 1775 Arg, and Pro 1749 Arg were not soluble and stable. Therefore we have characterized BRCA1 (1646-1859) His 1686 Gln mutation using Circular dichroism (CD) and Fluorescence spectroscopy and found little change in the secondary structure of the wild-type and the mutants (Figure 5C). Tryptophan is buried in the hydrophobic environment of BRCT structure (Figure 5D). Thus, we can conclude that the grade of pathogenicity of the BRCT domain cannot be defined based on the protein’s secondary structure or its three-dimensional folding levels. We can also conclude that hydrophobic interactions play a very important role in unraveling the functional complexities of BRCA1 BRCT mutations. Structural folding, weak intermolecular interactions and microarray analysis have already been reported for the pathogenic mutations M1775R  and M1775K , and to our conclusion we have found these mutations are not only impairing transactivation function but also destabilizing the hydrophobic interactions of BRCA1 BRCT.
Based on a large cohort of patients, breast and ovarian cancer is thought to be genetically heterogeneous. This genetic heterogeneity is indicated by familial occurrences of cancer, with mutations in a particular exon of BRCA1 involved in the same or in different families .Mutations at more than one locus can be found in different families or even in the same case. Based on our multimodal-based approach to characterizing the pathogenicity of mutations discovered in BRCA1 BRCT domains, we can conclude that genomic characterization alone is insufficient for clinical management of high-risk patients predisposed to cancer. There is also still a lack of consensus about whether genetic information can be used as a sole tool for targeted therapy or diagnostic biomarkers. We found that most of the mutations in the BRCA1 gene are located at the hydrophobic core of BRCT. The evaluation of pathogenicity of the mutations depends on how they destabilize the hydrophobic environment of the protein structure. As the hydrophobic region is buried inside the BRCT domains, all the reported pathogenic mutations destabilize the hydrophobic core, thereby further destabilizing the full-length BRCA1. The destabilization of three-dimensional folding as a result of the mutations leads to the loss of hydrophobic interactions and hydrogen bonding. It would be impossible to obtain essential information on the pathogenicity of the mutants without characterizing the associated protein functions. By looking at the structural characteristics, we found that the mutations that destabilize the hydrophobic core evaluate its pathogenicity. The next aim should be to look for small molecules that stabilize the hydrophobic environments of BRCT. These small molecules, which should target specific functions, i.e., causes of cancer, could serve as potential drug leads for targeted therapy. This will help in designing small-molecule compounds that may also help in stabilizing the genomic integrity and reducing the pathogenicity of the mutations.
Funding for this study was supported by DBT (No. BT/PR10765/BRB/664/2008 and BT/PR12565/BID/07/303/2009), Seed in Air fund from TMC. We thank DBT-BTIS facility at ACTREC for providing necessary software to this study. Startup funds of Varma Lab at Tata Memorial Centre, Advanced Centre for Treatment, Research and Education in Cancer, Kharghar to AKV. Hafiza Mahadik thanks Seed in Air fund from TMC for fellowship.
Make the best use of Scientific Research and information from our 700 + peer reviewed, Open Access Journals