**7.2 Inhibition of** *BRCA1* **transcriptional transactivation**

The one hybrid GAL4 transcription assay is used to study the effect of cisplatin on transcriptional transactivation. The level of transcriptional transactivation is inversely proportional to the amount of platinum-*BRCA1* adducts. The results are most likely due to inhibition of transcription of the reporter plasmid that resulted from interstrand crosslinks (Ratanaphan et al., 2009). The transcriptional transactivation activity of *BRCA1* has previously been reported by fusing the C-terminal domain of *BRCA1* to a heterogenous DNA-binding domain (Chapman and Verma, 1996). The BRCT domain (amino acids 1380- 1863) of human *BRCA1* scores positively in transcriptional activation trap experiments using various forms of so-called "one hybrid assay".The *BRCA1*-fused DNA-binding domain activates transcription in a cell-free system to a similar extent as a dose of the powerful activator, VP16 (Scully et al., 1997). A *GAL4:BRCA1* has also been introduced in yeast- and mammalian-based transcription assays to characterize the deleterious mutations in the 3/ terminal region of the *BRCA1* (Vallon-Christersson et al., 2001). The transcriptional activity reflects a tumor-suppressing function of the BRCA1 protein.

In order to investigate whether the drug-damaged *BRCA1* is able to transactivate the expression of a firefly luciferase gene, DNA repair-proficient MCF-7 cells were transiently transfected with the cisplatin-damaged pBIND-BRCT along with the reporter plasmid pG5Luc. The firefly luciferase activity was significantly decreased at a cisplatin concentration of 12.5 μM (Fig. 4).

It has been hypothesized that the BRCT domain could transactivate the expression of another reporter gene. The reporter gene pSV-β-galactosidase was used for this purpose. It was of interest, that the level of transactivation was significantly higher when co-transfected with the pBIND-BRCT than with the parental pBIND (Fig. 5). This indicated that the GAL4-BRCT domain may stimulate the pSV-β-galactosidase. However, the expression of β-galactosidase was decreased to the level of β-galactosidase alone when co-transfected with the platinated pBIND-BRCT. It was again of interest that, β-galactosidase expression was dramatically diminished when both the pSV-β-galactosidase and the pBIND-BRCT were platinated (Fig. 6). Expression of β-galactosidase from the pSV-β-galactosidase can be transactivated both by the GAL4 domain of the pBIND and pBIND-BRCT. Acting upon the GAL4 DNA sequence similarity, the GAL4 protein alone can stimulate the expression of β-galactosidase. However, the degree of transactivation was slightly higher by the pBIND-BRCT. This indicates that the BRCT domain on the fusion protein is able to transactivate the β-galactosidase gene-bearing pSV-β-galactosidase. When platinated pSV-β-galactosidase is co-transfected with the pBIND

A DNA Repair Protein BRCA1 as a Potentially

β**-galactosidase activity (folds)**

0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6

Molecular Target for the Anticancer Platinum Drug Cisplatin 217

**pSVcis pSVcis+pBIND pSVcis+pBIND-**

Fig. 6. Transcriptional transactivation of platinated pBIND-BRCT on platinated pSV-βgalactosidase. The platinated pSV-β-galactosidase (with cisplatin at a concentration of 12.5 μM) was co-transfected with non-platinated pBIND and pBIND-BRCT or platinated pBIND-BRCT. Cell lysates were prepared at 16 h after transfection. β-galactosidase expression was detected using the β-galactosidase assay. The data were derived from four independent

could transcribe the complementary templates which had no DNA lesions on the template strands (Corda et al., 1991). Transcription of globally platinated DNA templates by SP6 and T7 RNA polymerases were also blocked primarily at 1,2-d(GpG) and d(ApG) Pt adducts, and to a lesser extent at the interstrand crosslink (Tornaletti, 2005). Bifunctional Pt-DNA adducts were much more effective at impeding transcription progression than monofunctional DNA adducts (Tornaletti, 2005). Moreover, cisplatin caused a dosedependent inhibition of mRNA synthesis. Treatment of human fibroblast cells with 50 μM cisplatin for 24 h resulted in a 55% decrease in mRNA level and a reduced expression of p21WAF1 protein. This indicated that cisplatin inhibited the transcription of the *p21WAF1* gene (Ljungman et al., 1999). Recently, the processing of site-specific Pt-DNA crosslinks in mammalian cells was investigated (Ang et al., 2010). Site-specific platinated oligonucleotides, containing 1,2-d(GpG) and 1,3-d(GpTpG) adducts, were inserted into an expression vector between its promoter and a luciferase reporter gene. Transcription inhibitions that occurred by blocking passage of the RNA polymerase complex through the 1,2-d(GpG) and 1,3-d(GpTpG) adducts were 50% and 37.7% of the unplatinated controls for vectors, repectively. An X-ray crystal structure of RNA polymerase II showed stalling at the 1,2-intrastrand d(GpG) crosslink to explain the physical block of transcription by the cisplatin-DNA adduct (Damsma et al., 2007). Disruption of chromatin remodeling was another mechanism by which a cisplatin adduct could interfere with transcription. Nucleosomal DNA, containing the 1,2-d(GpG) or 1,3-d(GpTpG) intrastrand crosslinks, enforced a characteristic rotational positioning of the DNA around the histone octamer such that the Pt adduct faced inward towards the histone core (Ober and Lippard, 2008). Increased solvent accessibility of the platinated DNA strand was observed, and this indicated it might be caused by a structural perturbation in proximity of the DNA lesion. In addition, the nucleosomes treated with cisplatin exhibited a significant decrease in heatinduced mobility (Wu et al., 2008). These effects also indicated that a cisplatin assault could inhibit transcription by altering the native nucleosomal organization, and limiting the nucleosomal sliding that protected access of the RNA polymerase to the DNA template.

experiments ± standard deviations (SD) (Ratanaphan et al., 2009).

**BRCT**

**pSVcis+pBIND-BRCTcis**

or the pBIND-BRCT, a relatively lower expression of β-galactosidase was observed. The transcription level of β-galactosidase expression was reduced from 2-2.5 fold to 1.3 fold in both plasmids. Considering the data from the proficiency in repairing cisplatin-*BRCA1* adducts, it demonstrated that over 80% of the DNA lesion was repaired 8 h after cisplatin removal. Thus, it is possible that, during the repair time, RNA polymerase II or other transcriptional machineries may be blocked at any lesion on DNA (Jung & Lippard, 2003, 2006; Tornaletti et al., 2003).

Fig. 4. Time course of firefly luciferase expression. The pBIND-BRCT was incubated with cisplatin at concentrations of 0, 12.5, 25 and 50 μM and then co-transfected with the pG5Luc plasmid into MCF-7 cells. A cell lysate was prepared at 10, 16, 24 and 36 h after transfection. Firefly luciferase expression is detected by the Dual-Luciferase® Reporter Assay System. The data were derived from four independent experiments ± standard deviations (SD) (Ratanaphan et al., 2009).

Fig. 5. Transcriptional transactivation. The pBIND or pBIND-BRCT was co-transfected with pSV-β-galactosidase. Cell lysates were prepared at 16 h after transfection. β-galactosidase activity was detected using the β-galactosidase assay. The data were derived from four independent experiments ± standard deviations (SD) (Ratanaphan et al., 2009).

Several investigations have revealed transcriptional inhibition on DNA templates, containing the site-specific Pt-DNA adducts. The mammalian RNA polymerase II and *E. coli* RNA polymerase did not catalyze the transcriptional reactions when the DNA template strands carried the 1,2-intrastrand d(GpG) and d(ApG) adducts, whereas those polymerases

or the pBIND-BRCT, a relatively lower expression of β-galactosidase was observed. The transcription level of β-galactosidase expression was reduced from 2-2.5 fold to 1.3 fold in both plasmids. Considering the data from the proficiency in repairing cisplatin-*BRCA1* adducts, it demonstrated that over 80% of the DNA lesion was repaired 8 h after cisplatin removal. Thus, it is possible that, during the repair time, RNA polymerase II or other transcriptional machineries

**Firefly luciferase**

**0 µM 12.5 µM 25 µM 50 µM**

0 10 16 24 36 **Time (h)**

Fig. 4. Time course of firefly luciferase expression. The pBIND-BRCT was incubated with cisplatin at concentrations of 0, 12.5, 25 and 50 μM and then co-transfected with the pG5Luc plasmid into MCF-7 cells. A cell lysate was prepared at 10, 16, 24 and 36 h after transfection. Firefly luciferase expression is detected by the Dual-Luciferase® Reporter Assay System. The

data were derived from four independent experiments ± standard deviations (SD)

**pSV pSV+pBIND pSV+pBIND-**

independent experiments ± standard deviations (SD) (Ratanaphan et al., 2009).

Fig. 5. Transcriptional transactivation. The pBIND or pBIND-BRCT was co-transfected with pSV-β-galactosidase. Cell lysates were prepared at 16 h after transfection. β-galactosidase activity was detected using the β-galactosidase assay. The data were derived from four

Several investigations have revealed transcriptional inhibition on DNA templates, containing the site-specific Pt-DNA adducts. The mammalian RNA polymerase II and *E. coli* RNA polymerase did not catalyze the transcriptional reactions when the DNA template strands carried the 1,2-intrastrand d(GpG) and d(ApG) adducts, whereas those polymerases

**BRCT**

**pSV+pBIND-BRCTcis**

may be blocked at any lesion on DNA (Jung & Lippard, 2003, 2006; Tornaletti et al., 2003).

0 0.5 1 1.5 2 2.5 3 3.5

β**-galactosidase activity (folds)**

**Luciferase activity (%)**

(Ratanaphan et al., 2009).

Fig. 6. Transcriptional transactivation of platinated pBIND-BRCT on platinated pSV-βgalactosidase. The platinated pSV-β-galactosidase (with cisplatin at a concentration of 12.5 μM) was co-transfected with non-platinated pBIND and pBIND-BRCT or platinated pBIND-BRCT. Cell lysates were prepared at 16 h after transfection. β-galactosidase expression was detected using the β-galactosidase assay. The data were derived from four independent experiments ± standard deviations (SD) (Ratanaphan et al., 2009).

could transcribe the complementary templates which had no DNA lesions on the template strands (Corda et al., 1991). Transcription of globally platinated DNA templates by SP6 and T7 RNA polymerases were also blocked primarily at 1,2-d(GpG) and d(ApG) Pt adducts, and to a lesser extent at the interstrand crosslink (Tornaletti, 2005). Bifunctional Pt-DNA adducts were much more effective at impeding transcription progression than monofunctional DNA adducts (Tornaletti, 2005). Moreover, cisplatin caused a dosedependent inhibition of mRNA synthesis. Treatment of human fibroblast cells with 50 μM cisplatin for 24 h resulted in a 55% decrease in mRNA level and a reduced expression of p21WAF1 protein. This indicated that cisplatin inhibited the transcription of the *p21WAF1* gene (Ljungman et al., 1999). Recently, the processing of site-specific Pt-DNA crosslinks in mammalian cells was investigated (Ang et al., 2010). Site-specific platinated oligonucleotides, containing 1,2-d(GpG) and 1,3-d(GpTpG) adducts, were inserted into an expression vector between its promoter and a luciferase reporter gene. Transcription inhibitions that occurred by blocking passage of the RNA polymerase complex through the 1,2-d(GpG) and 1,3-d(GpTpG) adducts were 50% and 37.7% of the unplatinated controls for vectors, repectively. An X-ray crystal structure of RNA polymerase II showed stalling at the 1,2-intrastrand d(GpG) crosslink to explain the physical block of transcription by the cisplatin-DNA adduct (Damsma et al., 2007). Disruption of chromatin remodeling was another mechanism by which a cisplatin adduct could interfere with transcription. Nucleosomal DNA, containing the 1,2-d(GpG) or 1,3-d(GpTpG) intrastrand crosslinks, enforced a characteristic rotational positioning of the DNA around the histone octamer such that the Pt adduct faced inward towards the histone core (Ober and Lippard, 2008). Increased solvent accessibility of the platinated DNA strand was observed, and this indicated it might be caused by a structural perturbation in proximity of the DNA lesion. In addition, the nucleosomes treated with cisplatin exhibited a significant decrease in heatinduced mobility (Wu et al., 2008). These effects also indicated that a cisplatin assault could inhibit transcription by altering the native nucleosomal organization, and limiting the nucleosomal sliding that protected access of the RNA polymerase to the DNA template.

A DNA Repair Protein BRCA1 as a Potentially

Price et al., 1992).

Molecular Target for the Anticancer Platinum Drug Cisplatin 219

[Pt(NH3)2(OH)]+ that was attached to a BRCA1 peptide 111ENNSPEHLK119 (Fig. 7) (Berners-

Fig. 7. The product-ion spectrum of the MS/MS analysis for the 656.292+ ion. It indicated that [Pt(NH3)2(OH)]+ is attached to a peptide 111ENNSPEHLK119 of BRCA1 (Atipairin et al., 2010).

Thermal denaturation was monitored by circular dichroism (CD) to follow heat-induced unfolding which determined the effect of cisplatin binding on the stability of the BRCA1 RING domain. The BRCA1(1-139) protein pre-incubated with or without Zn2+ was incubated with cisplatin, and the CD spectra showed identical changes with an increase in ellipticity when the temperature was raised from 15°C to 95°C (Fig. 8). It indicated that the folded proteins gradually lost their ordered structures. When cooling to 20°C after being heated at 95°C, the CD spectrum partially recovered. This indicated that the reversibility of the unfolding/refolding process was incomplete. The melting temperatures of the BRCA1(1- 139) proteins were about 74°C and 83°C in the absence and presence of Zn2+, respectively (Fig. 9). This indicated that the BRCA1 RING domain was more thermostable by about 9°C upon Zn2+-binding. Thus, it supported the important role of Zn2+ in the determination and stabilization of the local secondary structure in the RING domain. It was notable that cisplatin at a concentration of 10 µM had similar melting temperatures to those observed for Zn2+ binding to the BRCA1 RING domain. However, higher melting temperatures were observed at a 10-fold concentration of cisplatin (100 µM). These data indicated that cisplatin binding to the BRCA1 RING domain conferred an enhanced thermostability by 13°C. Resistance to thermal denaturation of the cisplatin-modifed BRCA1 RING domain might result from the favourably intramolecular and intermolecular crosslinks driven by the free

To gain further insights into the functional consequences of cisplatin-induced BRCA1, the BRCA1 RING protein was platinated in vitro by cisplatin at various concentrations. The results showed that the relative E3 ligase activity was inversely proportional to the

**7.4 Thermal stability of the cisplatin-BRCA1 adducts** 

**7.5 Inactivation of BRCA1 E3 ligase activity by cisplatin** 

energy (Atipairin et al., 2010).

It has been suggested that inhibition of transcription by cisplatin was a critical determinant of cell-cycle arrest in the G2 phase because cells could not synthesize the mRNA necessary to pass into mitosis, and this eventually led to apoptosis. Possible mechanisms to explain this inhibitory process can be divided into three categories; (1) hijack of transcription factors (2) physical block of RNA polymerase, and (3) inhibition of chromatin remodeling (Todd & Lippard, 2009). A number of proteins have been identified that specifically recognize the distorted Pt-DNA adducts, including transcription factors. The upstream binding factor (UBF), a member of the HMG-domain proteins, is a ribosomal RNA transcription factor. hUBF can bind the 1,2-intrastrand adducts with a high *Kd* of 60 pM (Jordan & Carmo-Fonseca, 1998). Treatment of DNA with cisplatin inhibited ribosomal RNA synthesis by competing with hUBF for its natural binding site in an in vitro transcription assay (Zhai et al., 1998). The TATA-binding protein (TBP) is a critical transcription factor for all three mammalian RNA polymerases (pol I, II, and III). TBP binding to the DNA duplex, containing the 1,2-intrastrand d(GpG) crosslinks of cisplatin, was similar to that of the TATA-promoter binding in terms of structural and affinity aspects with a *Kd* of 0.3 nM (Jung et al., 2001). It was shown that TBP interacted directly with cisplatin-damaged DNA, and the introduction of exogenous cisplatin-modified DNA into the HeLa whole cell extract could sequester TBP and inhibit transcription 3-to 4-fold more than undamaged DNA (Vichi et al., 1997). Collectively, the failure of RNA synthesis resulted from the hijack of transcription factors by Pt-DNA adducts, that prevented the assembly of transcriptional elongation complexes at their normal promoter sequence and inhibited the transcriptional process. Significant reduction in transcriptional transactivation of cisplatin-modified *BRCA1* in the presence of a second expression vector containing multiple cisplatin-damaged sites could address the lack of or the unavailability of cellular transcription factors at cisplatin-*BRCA1*  lesions. Damage of *BRCA1*, if not properly repaired, may lead to its functional impairment in cancerous cells which ultimately induce programed cell death.

#### **7.3 Cisplatin binding to the BRCA1 RING domain**

The types of adduct formed with cisplatin are distinctive and dependent on the accessibility of the platinum center and protein side-chains (Ivanov et al., 1998; Peleg-Shulman et al., 2002). The BRCA1 RING domain has been found to form favourable intramolecular and intermolecular cross-links caused by cisplatin (Atipairin et al., 2010). Although cisplatin has been demonstrated to induce protein dimerization and has caused perturbations in some protein structures, the secondary structure of the BRCA1 RING domain in the apo-form was maintained and underwent more folded structural rearrangement after increasing cisplatin concentrations as judged by an increase in the negative CD spectra at 208 and 220 nm. It was possible that cisplatin might bind to the unoccupied Zn2+-binding sites and caused the structural changes. The binding constant of the in vitro platination was 3.00 ± 0.11 x 106 M-1, and the free energy of binding (ΔG) was -8.68 kcal Mol-1. In addition, the CD spectra of BRCA1 pre-incubated with Zn2+ gave identical profiles to indicate that cisplatin could interact with other residues rather than the Zn2+-binding sites and barely affected the overall conformation of the Zn2+-bound BRCA1. In order to locate the binding site of cisplatin on the BRCA1 (1-139) protein, in-gel tryptic digestion of the free BRCA1 and the cisplatin-BRCA1 adducts (molar ratio 1:1) were subjected to analysis by LC-MS. A unique fragment ion of 656.292+ was obtained from the cisplatin-BRCA1 adduct digests. Tandem mass spectrometric analyses of this fragment ion indicated that the ion arose from

It has been suggested that inhibition of transcription by cisplatin was a critical determinant of cell-cycle arrest in the G2 phase because cells could not synthesize the mRNA necessary to pass into mitosis, and this eventually led to apoptosis. Possible mechanisms to explain this inhibitory process can be divided into three categories; (1) hijack of transcription factors (2) physical block of RNA polymerase, and (3) inhibition of chromatin remodeling (Todd & Lippard, 2009). A number of proteins have been identified that specifically recognize the distorted Pt-DNA adducts, including transcription factors. The upstream binding factor (UBF), a member of the HMG-domain proteins, is a ribosomal RNA transcription factor. hUBF can bind the 1,2-intrastrand adducts with a high *Kd* of 60 pM (Jordan & Carmo-Fonseca, 1998). Treatment of DNA with cisplatin inhibited ribosomal RNA synthesis by competing with hUBF for its natural binding site in an in vitro transcription assay (Zhai et al., 1998). The TATA-binding protein (TBP) is a critical transcription factor for all three mammalian RNA polymerases (pol I, II, and III). TBP binding to the DNA duplex, containing the 1,2-intrastrand d(GpG) crosslinks of cisplatin, was similar to that of the TATA-promoter binding in terms of structural and affinity aspects with a *Kd* of 0.3 nM (Jung et al., 2001). It was shown that TBP interacted directly with cisplatin-damaged DNA, and the introduction of exogenous cisplatin-modified DNA into the HeLa whole cell extract could sequester TBP and inhibit transcription 3-to 4-fold more than undamaged DNA (Vichi et al., 1997). Collectively, the failure of RNA synthesis resulted from the hijack of transcription factors by Pt-DNA adducts, that prevented the assembly of transcriptional elongation complexes at their normal promoter sequence and inhibited the transcriptional process. Significant reduction in transcriptional transactivation of cisplatin-modified *BRCA1* in the presence of a second expression vector containing multiple cisplatin-damaged sites could address the lack of or the unavailability of cellular transcription factors at cisplatin-*BRCA1*  lesions. Damage of *BRCA1*, if not properly repaired, may lead to its functional impairment

in cancerous cells which ultimately induce programed cell death.

The types of adduct formed with cisplatin are distinctive and dependent on the accessibility of the platinum center and protein side-chains (Ivanov et al., 1998; Peleg-Shulman et al., 2002). The BRCA1 RING domain has been found to form favourable intramolecular and intermolecular cross-links caused by cisplatin (Atipairin et al., 2010). Although cisplatin has been demonstrated to induce protein dimerization and has caused perturbations in some protein structures, the secondary structure of the BRCA1 RING domain in the apo-form was maintained and underwent more folded structural rearrangement after increasing cisplatin concentrations as judged by an increase in the negative CD spectra at 208 and 220 nm. It was possible that cisplatin might bind to the unoccupied Zn2+-binding sites and caused the structural changes. The binding constant of the in vitro platination was 3.00 ± 0.11 x 106 M-1, and the free energy of binding (ΔG) was -8.68 kcal Mol-1. In addition, the CD spectra of BRCA1 pre-incubated with Zn2+ gave identical profiles to indicate that cisplatin could interact with other residues rather than the Zn2+-binding sites and barely affected the overall conformation of the Zn2+-bound BRCA1. In order to locate the binding site of cisplatin on the BRCA1 (1-139) protein, in-gel tryptic digestion of the free BRCA1 and the cisplatin-BRCA1 adducts (molar ratio 1:1) were subjected to analysis by LC-MS. A unique fragment ion of 656.292+ was obtained from the cisplatin-BRCA1 adduct digests. Tandem mass spectrometric analyses of this fragment ion indicated that the ion arose from

**7.3 Cisplatin binding to the BRCA1 RING domain** 

[Pt(NH3)2(OH)]+ that was attached to a BRCA1 peptide 111ENNSPEHLK119 (Fig. 7) (Berners-Price et al., 1992).

Fig. 7. The product-ion spectrum of the MS/MS analysis for the 656.292+ ion. It indicated that [Pt(NH3)2(OH)]+ is attached to a peptide 111ENNSPEHLK119 of BRCA1 (Atipairin et al., 2010).

#### **7.4 Thermal stability of the cisplatin-BRCA1 adducts**

Thermal denaturation was monitored by circular dichroism (CD) to follow heat-induced unfolding which determined the effect of cisplatin binding on the stability of the BRCA1 RING domain. The BRCA1(1-139) protein pre-incubated with or without Zn2+ was incubated with cisplatin, and the CD spectra showed identical changes with an increase in ellipticity when the temperature was raised from 15°C to 95°C (Fig. 8). It indicated that the folded proteins gradually lost their ordered structures. When cooling to 20°C after being heated at 95°C, the CD spectrum partially recovered. This indicated that the reversibility of the unfolding/refolding process was incomplete. The melting temperatures of the BRCA1(1- 139) proteins were about 74°C and 83°C in the absence and presence of Zn2+, respectively (Fig. 9). This indicated that the BRCA1 RING domain was more thermostable by about 9°C upon Zn2+-binding. Thus, it supported the important role of Zn2+ in the determination and stabilization of the local secondary structure in the RING domain. It was notable that cisplatin at a concentration of 10 µM had similar melting temperatures to those observed for Zn2+ binding to the BRCA1 RING domain. However, higher melting temperatures were observed at a 10-fold concentration of cisplatin (100 µM). These data indicated that cisplatin binding to the BRCA1 RING domain conferred an enhanced thermostability by 13°C. Resistance to thermal denaturation of the cisplatin-modifed BRCA1 RING domain might result from the favourably intramolecular and intermolecular crosslinks driven by the free energy (Atipairin et al., 2010).

#### **7.5 Inactivation of BRCA1 E3 ligase activity by cisplatin**

To gain further insights into the functional consequences of cisplatin-induced BRCA1, the BRCA1 RING protein was platinated in vitro by cisplatin at various concentrations. The results showed that the relative E3 ligase activity was inversely proportional to the

A DNA Repair Protein BRCA1 as a Potentially

al., 2011a).

**8. Conclusion** 

Molecular Target for the Anticancer Platinum Drug Cisplatin 221

concentration of the drug (Fig. 10). An increase in platinum concentration was accompanied by a high amount of BRCA1 adducts and a low amount of native BRCA1 protein. To address whether the inhibition of the E3 ligase activity resulted from the formation of BRCA1 adducts or a reduced amount of the BRCA1 subunit, a ten-fold excess amount of the platinated BRCA1 was assayed for the E3 ligase activity. The result demonstrated that platination of BRCA1 was

Fig. 10. *In vitro* ubiquitin ligase activity of cisplatin-BRCA1 complexes. Two µg of the drug-BRCA1 adducts with a number of defined concentrations of cisplatin was assayed for the ubiquitin ligase activitiy. An apparent ubiquitinated product (as indicated by the filled diamond) was markedly reduced as the concentration of platinum increased (Atipairin et

We have demonstrated an in vitro inactivation of *BRCA1*/BRCA1 by the anticancer platinum drug cisplatin. The transcriptional activation of cisplatin-modified *BRCA1*, when tested in a "one-hybrid GAL4 transcriptional assay", was inversely proportional to cisplatin doses and was dramatically diminished in the presence of a second expression vector containing multiple cisplatin-damaged sites. This indicates a repair-mediated transcriptional transactivation of cisplatin-damaged *BRCA1* as well as the lack or unavailability of cellular transcription factors at cisplatin-*BRCA1* lesions. The BRCA1 protein contained a preformed structure in the apo-form with structural changes and more resistance to limited proteolysis after Zn2+ binding. Cisplatin-bound protein exhibited an enhanced thermostability, resulting from the favourable intermolecular crosslinks driven by the free energy. Only the apo-form, not the holo-form, of BRCA1 underwent a more folded structural rearrangement with the retention of protein structure

indeed involved in the inhibition of the E3 ligase activity (Atipairin et al., 2011a).

Fig. 8. Thermal transition of the cisplatin-BRCA1 adducts in the presence of Zn2+. The BRCA1(1-139) proteins (10 µM) after pre-incubation with a 3 molar equivalent ratio of Zn2+ to protein were mixed with cisplatin concentrations of 10 µM. Samples were incubated in the dark at ambient temperature for 24 h. The measurements were performed from 15°C to 95°C with a heating rate of 1°C/min. After heating to 95°C, the measurement at 20°C was also performed. The CD spectra were plotted between the mean residue ellipticity and wavelength (Atipairin et al., 2010).

Fig. 9. Thermal denaturation curves of the cisplatin-BRCA1 adducts. The BRCA1(1-139) protein (10 µM) without Zn2+ and after pre-incubation with a 3 molar equivalent ratio of Zn2+ to protein were mixed with various concentrations of cisplatin (0, 10, and 100 µM). Samples were incubated in the dark at ambient temperature for 24 h before CD measurements. The CD signals at 208 nm were measured, and the unfolded fraction as a function of temperature was plotted (Atipairin et al., 2010).

200 210 220 230 240 250 260

**Wavelength (nm)**

Fig. 8. Thermal transition of the cisplatin-BRCA1 adducts in the presence of Zn2+. The

BRCA1(1-139) proteins (10 µM) after pre-incubation with a 3 molar equivalent ratio of Zn2+ to protein were mixed with cisplatin concentrations of 10 µM. Samples were incubated in the dark at ambient temperature for 24 h. The measurements were performed from 15°C to 95°C with a heating rate of 1°C/min. After heating to 95°C, the measurement at 20°C was also performed. The CD spectra were plotted between the mean residue ellipticity and wavelength

Fig. 9. Thermal denaturation curves of the cisplatin-BRCA1 adducts. The BRCA1(1-139) protein (10 µM) without Zn2+ and after pre-incubation with a 3 molar equivalent ratio of Zn2+ to protein were mixed with various concentrations of cisplatin (0, 10, and 100 µM).

measurements. The CD signals at 208 nm were measured, and the unfolded fraction as a

Samples were incubated in the dark at ambient temperature for 24 h before CD

function of temperature was plotted (Atipairin et al., 2010).

**15°C 20°C 35°C 50°C 65°C 75°C 85°C 90°C 95°C 20°C after 95°C**


**[**θ**] (x103 deg.cm**

(Atipairin et al., 2010).

**2.dmol -1)** concentration of the drug (Fig. 10). An increase in platinum concentration was accompanied by a high amount of BRCA1 adducts and a low amount of native BRCA1 protein. To address whether the inhibition of the E3 ligase activity resulted from the formation of BRCA1 adducts or a reduced amount of the BRCA1 subunit, a ten-fold excess amount of the platinated BRCA1 was assayed for the E3 ligase activity. The result demonstrated that platination of BRCA1 was indeed involved in the inhibition of the E3 ligase activity (Atipairin et al., 2011a).

Fig. 10. *In vitro* ubiquitin ligase activity of cisplatin-BRCA1 complexes. Two µg of the drug-BRCA1 adducts with a number of defined concentrations of cisplatin was assayed for the ubiquitin ligase activitiy. An apparent ubiquitinated product (as indicated by the filled diamond) was markedly reduced as the concentration of platinum increased (Atipairin et al., 2011a).
