**1. Introduction**

204 DNA Repair

Zhou B.-B.S. & Elledge S.J. (2000) The DNA damage response: putting checkpoints in

Cancer is a leading cause of death in the world. The incidence of cancers is related to environmental factors, behavioral patterns, and genetic disorders. Cancer therapy usually aims to selectively destroy cancer cells while sparing normal tissue. Most chemotherapeutic agents function by damaging cancer cell DNA. The cellular responses to DNA damage are thus critical factors for determining the effectiveness of most cancer therapies (Ashworth, 2008). When normal cells are exposed to damage, DNA repair mechanism is induced. The DNA repair processes are the cellular responses associated with the restoration of the normal DNA nucleotide sequences. The DNA repair activity of the cell is an important determinant of a cells sensitivity to chemotherapeutic agents. It is known that resistance to DNA-damaging agents can be associated with increased cellular repair activities, while defects in DNA repair pathways result in hypersensitivity to damage (Kelley & Fishel, 2008; Quinn et al., 2003, 2009). Several studies have clearly demonstrated that the impairment or absence of genes or proteins responsible for DNA damage repair, frequently causes genomic instability, cell cycle arrest and apoptosis. The importance of these repair pathways is highlighted by the fact that more than 100 genes have been found in mammalian cells that are involved in some way in DNA damage repair pathways. The breast cancer susceptibility gene 1 (*BRCA1*) is a tumor suppressor gene involved in maintaining genomic integrity through multiple functions in DNA damage repair, transcriptional regulation, a cell cycle checkpoint and protein ubiquitination (Brzovic et al., 2001; Hashizume et al., 2001; Mark et al., 2005; Varma et al., 2005; Williams et al., 2004). In cancer cells, damage to *BRCA1* by the anticancer platinum drug cisplatin may lead to a loss of such functions and ultimately results in cancer cell death. In addition, preclinical and clinical studies have recently revealed that inactivation of the BRCA1 protein in cancer cells leads to chemosensitivity. Therefore, approaching the BRCA1 protein as a potential therapeutic target for cisplatin or other such platinum based drugs might be of interest for molecular-targeted cancer therapy. In this chapter, the biophysical characterization and functional consequences of the human *BRCA1* gene and the BRCA1 RING protein induced by cisplatin are described.

#### **2. Breast cancer susceptibility gene 1 (***BRCA1***) and its encoded protein**

In 1990, chromosome 17q21 was identified by linkage analysis as the location of a breast cancer susceptibility gene 1 or *BRCA1* (Hall et al., 1990). The entire gene covers approximately 100 kb

A DNA Repair Protein BRCA1 as a Potentially

**2.2 The large central segment of BRCA1** 

damage (Sy et al., 2009).

**2.3 The BRCA1 C-terminal domain**

Molecular Target for the Anticancer Platinum Drug Cisplatin 207

progression to S phase by aggregation of nuclear BRCA1 and BARD1 implied the importance of both proteins for a DNA repair function (Jin et al., 1997). The BRCA1-BARD1 complex also exhibits enzymatic activity of an E3 ubiquitin ligase that specifically transfers ubiquitin to protein substrates that are essential for cellular viability (Hashizume et al., 2001; Xia et al., 2003). Cancer-predisposing mutations in the Zn2+-binding sites were demonstrated not only to alter the affinity for Zn2+ and the native BRCA1 RING structure but also abolished the interaction with BARD1 and the E3 ligase activity (Morris et al., 2006). The results supported the importance of Zn2+ as a structural component, as it obviously played a critical role in the stabilization of the structure and function of the BRCA1 RING domain.

The central segment of BRCA1 covers exon 11 (approximately 3500 bp) and constitutes approximately 60 percent of the coding region of the gene. Deletion of exon 11 results in removal of the nuclear localization signal of BRCA1. Biophysical characterization revealed that this domain was intrinsically disordered or natively unfolded under physiological conditions. This might potentially allow the BRCA1 central region to act as a long flexible scaffold, to mediate interactions with DNA, and perhaps a number of other proteins involved in the DNA damage response and repair (Mark et al., 2005). The reported binding partners to the central region were c-Myc, RB, p53, FANCA, RAD50, RAD51, JunB, and BRCA2 (Rosen et al., 2003). Recently, the BRCA1 central region has been shown to efficiently interact with p53, and stimulate p53-mediated DNA binding and transcriptional activities (Buck, 2008). This result indicated that the BRCA1 central segment facilitated the induction of cell cycle arrest and apoptosis in response to DNA damage. Furthermore, the association between the central region of BRCA1 and PALB2 (partner and localizer of BRCA2, also known as FANCN) was observed primarily through apolar bonding between their respective coiled-coil domains (Sy et al., 2009). PALB2 binds directly to BRCA1, and serves as the molecular scaffold for the formation of the BRCA1-PALB2-BRCA2 complex. BRCA1 mutations (L1407P and M1411T) identified in cancer patients were shown to disrupt the specific interaction between BRCA1 and PALB2, resulting in a defective homologous recombination (HR) repair and a compromised cell survival after DNA

The C-terminal region (residues 1646-1863) of BRCA1 contains two BRCT (BRCA1 Cterminal) domains in tandem (motif 1: amino acids 1653-1736; motif 2: amino acids 1760- 1855). Each BRCT domain is characterized by a central, parallel four-stranded β-sheet with a pair of α-helices (α1 and α3) packed against one face, and a single α-helix (α2) packed against the opposite face of the sheet (Williams et al., 2001). The two BRCA1-BRCT repeats interact in a head-to-tail fashion. This domain serves as a multipurpose protein-protein interaction module that binds to other BRCT repeats or other protein domains with apparently unrelated structures (Watts & Brissett, 2010). Based on its physical interactions with other proteins, BRCA1 has been implicated in a wide array of cellular functions, including cell cycle regulation, DNA damage response, transcriptional regulation, replication and recombination, and higher chromatin hierarchical control (Starita & Parvin, 2003). The BRCA1-BRCT domain has been identified as a phosphopeptide recognition module, and is demonstrated to bind to the phosphorylated protein partners (BACH1 and

of genomic sequence, and was subsequently cloned four years later (Miki et al., 1994). BRCA1 is a tumor suppressor gene composed of 24 exons, with an mRNA that is 7.8 kb in length, and 22 coding exons that translate into a protein of 1863 amino acids (Fig. 1) with a molecular weight of 220 kDa (Brzovic et al., 1998). It has 3 major domains, including (1) the N-terminal RING finger domain (BRCA1 RING domain), (2) the large central segment with the nuclear localization signal (NLS), and (3) the BRCA1 C-terminal domain (BRCT). The BRCA1 protein plays an essential role in maintaining genomic stability associated with a number of cellular processes, including DNA repair, a cell cycle checkpoint, transcriptional regulation, and protein ubiquitination (Huen et al., 2010; O'Donovan & Livingston, 2010).

Fig. 1. Scheme of BRCA1 mRNA and sites of protein interaction

#### **2.1 The BRCA1 RING domain**

The N-terminal RING finger domain contains the conservative sequences of cysteine and histidine residues (C3HC4) necessary for specific coordination with two Zn2+ ions. The first 109 amino acids of BRCA1 protein constitute a protease-resistance domain. The solution structure of the BRCA1 RING domain revealed the existence of antiparallel α-helices at both ends, flanking the central RING motif (residues 24-64) and was characterized by a short antiparallel three-stranded β-sheet, and two large Zn2+-binding loops, and a central α-helix (Brzovic et al., 2001)The two Zn2+-binding sites are formed in an interleaved fashion in which the first and third pairs of cysteines (Cys24, Cys27, Cys44, and Cys47) form site I, and the second and fourth pairs of cysteines and a histidine (Cys39, His41, Cys61, and Cys64) form site II. It is an important domain since it might mediate a central role in macromolecular interactions to exert the tumor suppression functions. The solution structure together with yeast-two-hybrid studies revealed that the BRCA1 RING domain preferentially formed a heterodimeric complex with another RING domain BARD1 (BRCA1 associated RING domain 1) through an extensive four-helix-bundle interface (Brzovic et al., 2001; Wu et al., 1996). The binding interface is composed of residues 8-22 and 81-96 of BRCA1, and residues 36-48 and 101-116 of BARD1. The BRCA1-BARD1 complex requires each other for their mutual stabilities, and they are co-localized in nuclear dots during S phase but not the G phase of the cell cycle and in nuclear foci (Hashizume et al., 2001). The

of genomic sequence, and was subsequently cloned four years later (Miki et al., 1994). BRCA1 is a tumor suppressor gene composed of 24 exons, with an mRNA that is 7.8 kb in length, and 22 coding exons that translate into a protein of 1863 amino acids (Fig. 1) with a molecular weight of 220 kDa (Brzovic et al., 1998). It has 3 major domains, including (1) the N-terminal RING finger domain (BRCA1 RING domain), (2) the large central segment with the nuclear localization signal (NLS), and (3) the BRCA1 C-terminal domain (BRCT). The BRCA1 protein plays an essential role in maintaining genomic stability associated with a number of cellular processes, including DNA repair, a cell cycle checkpoint, transcriptional regulation, and

protein ubiquitination (Huen et al., 2010; O'Donovan & Livingston, 2010).

Fig. 1. Scheme of BRCA1 mRNA and sites of protein interaction

The N-terminal RING finger domain contains the conservative sequences of cysteine and histidine residues (C3HC4) necessary for specific coordination with two Zn2+ ions. The first 109 amino acids of BRCA1 protein constitute a protease-resistance domain. The solution structure of the BRCA1 RING domain revealed the existence of antiparallel α-helices at both ends, flanking the central RING motif (residues 24-64) and was characterized by a short antiparallel three-stranded β-sheet, and two large Zn2+-binding loops, and a central α-helix (Brzovic et al., 2001)The two Zn2+-binding sites are formed in an interleaved fashion in which the first and third pairs of cysteines (Cys24, Cys27, Cys44, and Cys47) form site I, and the second and fourth pairs of cysteines and a histidine (Cys39, His41, Cys61, and Cys64) form site II. It is an important domain since it might mediate a central role in macromolecular interactions to exert the tumor suppression functions. The solution structure together with yeast-two-hybrid studies revealed that the BRCA1 RING domain preferentially formed a heterodimeric complex with another RING domain BARD1 (BRCA1 associated RING domain 1) through an extensive four-helix-bundle interface (Brzovic et al., 2001; Wu et al., 1996). The binding interface is composed of residues 8-22 and 81-96 of BRCA1, and residues 36-48 and 101-116 of BARD1. The BRCA1-BARD1 complex requires each other for their mutual stabilities, and they are co-localized in nuclear dots during S phase but not the G phase of the cell cycle and in nuclear foci (Hashizume et al., 2001). The

**2.1 The BRCA1 RING domain**

progression to S phase by aggregation of nuclear BRCA1 and BARD1 implied the importance of both proteins for a DNA repair function (Jin et al., 1997). The BRCA1-BARD1 complex also exhibits enzymatic activity of an E3 ubiquitin ligase that specifically transfers ubiquitin to protein substrates that are essential for cellular viability (Hashizume et al., 2001; Xia et al., 2003). Cancer-predisposing mutations in the Zn2+-binding sites were demonstrated not only to alter the affinity for Zn2+ and the native BRCA1 RING structure but also abolished the interaction with BARD1 and the E3 ligase activity (Morris et al., 2006). The results supported the importance of Zn2+ as a structural component, as it obviously played a critical role in the stabilization of the structure and function of the BRCA1 RING domain.

#### **2.2 The large central segment of BRCA1**

The central segment of BRCA1 covers exon 11 (approximately 3500 bp) and constitutes approximately 60 percent of the coding region of the gene. Deletion of exon 11 results in removal of the nuclear localization signal of BRCA1. Biophysical characterization revealed that this domain was intrinsically disordered or natively unfolded under physiological conditions. This might potentially allow the BRCA1 central region to act as a long flexible scaffold, to mediate interactions with DNA, and perhaps a number of other proteins involved in the DNA damage response and repair (Mark et al., 2005). The reported binding partners to the central region were c-Myc, RB, p53, FANCA, RAD50, RAD51, JunB, and BRCA2 (Rosen et al., 2003). Recently, the BRCA1 central region has been shown to efficiently interact with p53, and stimulate p53-mediated DNA binding and transcriptional activities (Buck, 2008). This result indicated that the BRCA1 central segment facilitated the induction of cell cycle arrest and apoptosis in response to DNA damage. Furthermore, the association between the central region of BRCA1 and PALB2 (partner and localizer of BRCA2, also known as FANCN) was observed primarily through apolar bonding between their respective coiled-coil domains (Sy et al., 2009). PALB2 binds directly to BRCA1, and serves as the molecular scaffold for the formation of the BRCA1-PALB2-BRCA2 complex. BRCA1 mutations (L1407P and M1411T) identified in cancer patients were shown to disrupt the specific interaction between BRCA1 and PALB2, resulting in a defective homologous recombination (HR) repair and a compromised cell survival after DNA damage (Sy et al., 2009).

#### **2.3 The BRCA1 C-terminal domain**

The C-terminal region (residues 1646-1863) of BRCA1 contains two BRCT (BRCA1 Cterminal) domains in tandem (motif 1: amino acids 1653-1736; motif 2: amino acids 1760- 1855). Each BRCT domain is characterized by a central, parallel four-stranded β-sheet with a pair of α-helices (α1 and α3) packed against one face, and a single α-helix (α2) packed against the opposite face of the sheet (Williams et al., 2001). The two BRCA1-BRCT repeats interact in a head-to-tail fashion. This domain serves as a multipurpose protein-protein interaction module that binds to other BRCT repeats or other protein domains with apparently unrelated structures (Watts & Brissett, 2010). Based on its physical interactions with other proteins, BRCA1 has been implicated in a wide array of cellular functions, including cell cycle regulation, DNA damage response, transcriptional regulation, replication and recombination, and higher chromatin hierarchical control (Starita & Parvin, 2003). The BRCA1-BRCT domain has been identified as a phosphopeptide recognition module, and is demonstrated to bind to the phosphorylated protein partners (BACH1 and

A DNA Repair Protein BRCA1 as a Potentially

(PRE) of c-Myc.

**5. BRCA1 and protein ubiquitination**

**4. BRCA1 and transcriptional regulation** 

Molecular Target for the Anticancer Platinum Drug Cisplatin 209

As described earlier, BRCA1 contains a C-terminal transactivation domain as was first defined using the yeast two-hybrid system (Chapman & Verma, 1996; Monteiro et al., 1996). The transactivation domain was mapped to the region of the protein encoded by exons 21- 24 using deletion constructs of BRCA1 fused to the GAL4 DNA binding domain. The BRCA1-BRCT domain has been implicated in the regulation of transcription of several genes responsible for DNA damage. The ability of BRCA1 to act as either a co-activator or a corepressor of transcription may involve its ability to recruit the basal transcriptional machinery and other proteins that have been implicated in chromatin remodeling (Mullan et al., 2006). BRCA1 was capable of activating the p21 promoter (Somasundaram et al., 1997). One report claimed that BRCA1 participated in the stabilization of p53 in response to DNA damage, and served as a co-activator for p53 (Zhang et al., 1998). The interaction of BRCA1 and p53 potentially resulted in the redirection of p53-mediated transactivation from a proapoptotic target to genes involved in DNA repair and cell cycle arrest (Zhang et al., 1998). In addition, BRCA1 has been shown to interact with the RNA polymerase II holoenzyme (Scully et al., 1997). However, BRCA1 could repress the transcription of an estrogen receptor α (ERα) and its downstream estrogen responsive genes (Fan et al., 1999). The transcriptional repression activity of BRCA1 for ERα occurs by the association of the N-terminus of BRCA1 (residues 1-300) with the C-terminal activation function (AF-2) of ERα. Breast cancerassociated mutations of BRCA1 were found to abolish its ability to inhibit ERα activity (Fan et al., 2001). The repression activity exerted by BRCA1 involved the ability of BRCA1 to down-regulate levels of the transcriptional coactivator p300, which has also been shown to interact with the AF-2 domain of ERα (Fan et al., 2002). Further investigations revealed that overexpression of BRCA1 could inhibit the recruitment of the co-activators [steroid receptor co-activator 1 (SRC1), and amplified breast cancer 1 (AIB1)], and enhanced the recruitment of a co-repressor [histone deacetylase 1 (HDAC1)] to the progesterone response elements

The BRCA1 protein displays an E3 ubiquitin ligase activity through its RING domain, and this activity is enhanced when it exists as a heterodimer with the BARD1 RING domain (Xia et al., 2003). In vitro and in vivo studies have indicated that the BRCA1-BARD1 complex was capable of autoubiquitination that paradoxically stabilized the protein complex, and that also activated its in vitro E3 ligase activity with other proteins (Chen et al., 2002; Wu-Baer et al., 2010). However, the substrate specificity of the BRCA1 E3 ligase activity and its biological relevance to tumor suppression function are still unknown. Putative substrates for ubiquitination by the BRCA1-BARD1 RING complexes have recently emerged from in vitro and in vivo studies such as the nucleosomal histones H2A and its variant H2AX, RNA polymerase II, γ-tubulin, nucleophosmin/B23, and estrogen receptor α (ERα) (Eakin et al., 2007; Horwitz et al., 2007; Parvin, 2009; Sato et al., 2004; Starita et al., 2005; Thakar et al., 2010). BRCA1 can form a RING heterodimer E3 ligase activity with BARD1, and this is required for the recruitment of BRCA2 and RAD51 to damaged sites for HR repair (Ransburgh et al., 2010). Many cancer-predisposing mutations in the BRCA1 RING domain, that inhibited the E3 ligase activity and its ability to accumulate at damaged sites, were defective in homologous recombination that is critical for tumor suppression (Morris et al.,

CtIP, containing the consensus sequence pSer-X-X-Phe) that is involved mainly in the control of the G2/M phase checkpoint and DNA damage repair (Varma et al., 2005; Williams et al., 2004). Several cancer-predisposing mutations in the BRCA1-BRCT domain resulted in destabilization of the structural integrity at the BRCT active sites, and abolished their affinities to synthetic BACH1 and CtIP phosphopeptides (Rowling et al., 2010). These findings provide a better understanding of the pathogenic BRCA1 mutations on functional mechanisms and tumorigenesis.

#### **3. BRCA1 and DNA damage repair**

A substantial amount of evidence that has implicated BRCA1 in the DNA damage repair pathways has been documented. BRCA1 co-localizes with RAD51 and BARD1 to nuclear foci (sites associated with repair of DNA caused by the damaging agents or γ-irradiation) (Hashizume et al., 2001; Scully et al., 1997). The nuclear foci is marked by the histone variant H2AX that was phosphorylated on Ser139 (known as γH2AX) (Rogakou et al., 1998). γH2AX is one of the initial recruiting factors for various checkpoints and DNA repair proteins, including Abraxas, RAP80, and BRCA1, at sites of DNA breaks (Foulkes, 2010). The H2AX signaling cascade begins to emerge with the finding that MDC1 (mediator of DNA damage checkpoint 1) is the main downstream factor in the pathway, and is required for the damage-induced focal accumulation of a number of DNA damage repair factors at the DNA breaks (Stucki et al., 2005).

BRCA1 plays a role in maintaining genome integrity through its role in DNA damage repair. Several observations have implicated BRCA1 in homologous recombination (HR), non-homologous end-joining (NHEJ and nucleotide excision repair (NER). A role for BRCA1 in HR-mediated repair is involved through its stable complex formation with BRCA2, which has a well-defined role in HR through its direct interaction with RAD51 (Bhattacharyya et al., 2000). RAD51 (the mammalian homolog of the Escherichia coli RecA protein) is a DNA recombinase that catalyzes strand exchange in an early step of HR (Baumann et al., 1996). PALB2 (the partner and localizer of BRCA2) has recently been identified as the bridging factor required for the BRCA1-BRCA2 association (Rahman et al., 2007). The BRCA1-PALB2 interaction was mediated by their respective coiled-coil domains, and was found to promote HR-mediated repair (Rahman et al., 2007). Importantly, missense mutations identified in the PALB2-binding region on BRCA1 disrupted the specific interaction of BRCA1 with PALB2, and compromised DNA repair in a gene conversion assay (Sy et al., 2009). Although these studies have revealed a molecular link between BRCA1 function and HR-mediated repair, the mechanism by which BRCA1 promotes HR through the PALB2-BRCA2-RAD51 axis remains unclear.

As an alternative to HR, there is a growing body of evidences, to indicate that a component of NHEJ is regulated by BRCA1. The exact role of BRCA1 in NHEJ, however, has not been well defined (Zhang & Powell, 2005). In the NHEJ pathway, the DNA-dependent protein kinase catalytic subunit (DNA-PKcs) and a Ku heterodimer of Ku80 and Ku70 are recruited to the sites of DNA DSBs for preparing the DNA ends before ligation by the XRCC4 ligase IV. The most possible explanation for BRCA1 being involved in NHEJ is its association with a NHEJ factor Ku80 (Chiba & Parvin, 2001; Wei et al., 2008). Many studies have provided strong evidences that the NHEJ pathway was impaired, both in vivo and in vitro, in BRCA1-deficient mouse embryonic fibroblasts and in the human breast cancer cell line HCC1937 which carries a homozygous mutation in the BRCA1 gene (Bau et al., 2004; Zhong et al., 2002).
