Association For Molecular Pathology et al v. United States Patent and Trademark Office et al

Filing 188

FILING ERROR - DEFICIENT DOCKET ENTRY - DECLARATION of Laura P. Masurovsky in Support re: 186 MOTION to File Amicus Brief. Document filed by Genetic Alliance. (Attachments: # 1 Exhibit 1, # 2 Exhibit 2, # 3 Exhibit 3, # 4 Exhibit 4, # 5 Exhibit 5, # 6 Exhibit 6)(Masurovsky, Laura) Modified on 12/31/2009 (db).

Association For Molecular Pathology et al v. United States Patent and Trademark Office et al

Doc. 188 Att. 5

Exhibit 5

Dockets.Justia.com

ã

O nc o g e n e (2 0 0 1 ) 2 0 , 4 4 0 ± 4 5 0 2 0 0 1 N a t u r e P u b l i s h in g Gr o u p A l l r i g h t s r e s er v e d 0 9 5 0 ± 9 2 3 2 / 0 1 $ 1 5 . 0 0 www.nature.com/onc

Identi®cation of a novel transcriptional repressor element located in the ®rst intron of the human BRCA1 gene
Ting-Chung Suen1 and Paul E Goss*,1,2
1

Breast Cancer Prevention Program, The Toronto Hospital, Oncology Research Laboratories, Ontario M5G 2M9, Canada; 2Breast Group, Department of Medical Oncology, Princess Margaret Hospital, Toronto, Ontario M5G 2M9, Canada

Loss or lowered expression of BRCA1 in non-familial breast cancer has been shown in several recent studies. Understanding how BRCA1 expression is regulated should provide new insights into the role of BRCA1 in sporadic breast cancer. We have recently identi®ed a critical 18-base pair (bp) DNA element within the minimal BRCA1 promoter whereupon the formation of a speci®c protein-DNA complex and transcription of BRCA1 is dependent. We now report a non tissuespeci®c transcriptional repressor activity, located more than 500 bp into the ®rst intron of BRCA1. Progressive deletions from the 3'-end of intron 1 and reporter gene assays localized the repressor activity to an 83-bp region. Electrophoretic mobility shift assays with this 83 bp DNA and various sub-fragments of it showed binding of nuclear proteins to a 36 bp BstNI ± BseRI fragment. Functional transcriptional repression by this 36 bp DNA could be conferred on a heterologous thymidine kinase promoter. Analysis of multiple reporter gene constructs containing the BRCA1 genomic region driving transcription in both directions suggests that the putative negative regulatory element functions to block transcription only in the BRCA1 direction, although the promoter is shared by the divergently transcribed NBR2 gene. Oncogene ( 20 01 ) 2 0, 44 0 ± 45 0. Keywords: BRCA1; breast cancer; transcriptional regulation; intron; repressor Introduction Mutations of the BRCA1 gene are responsible for the vast majority of breast and ovarian cancer families, and for one-third of breast cancer only families (Miki et al., 1994; Futreal et al., 1994; Ford et al., 1998). BRCA1 encodes a gene product of 1863 amino acid residues (for reviews, see Casey, 1997; Bertwistle and Ashworth, 1998), translating into a 220 kd nuclear phosphoprotein (Chen et al., 1996; Runer and Verma, 1997; Wilson et al., 1999). Recent studies using BRCA1-de®cient mouse embryonic stem cell lines (Moynahan et al., 1999), primary skin ®broblasts (Cressman et al., 1999), or a BRCA1-mutated human breast cancer cell line (Scully et

*Correspondence: PE Goss Received 30 June 2000; revised 2 November 2000; accepted 2 November 2000

al., 1999) have provided strong evidence of a caretaker function for BRCA1 (Deng and Scott, 2000). These studies also demonstrated that loss of p53 function is a growth-promoting event in the transformation process of BRCA1 de®cient cells (Shen et al., 1998; Cressman et al., 1999). BRCA1 can function as a transcriptional activator when fused to the Gal4 protein, both in vitro (Haile and Parvin, 1999) and in vivo (Chapman and Verma, 1996; Monteiro et al., 1996). Its ability to stimulate p21 expression provides direct evidence of its role as a transcription factor (Somasundaram et al., 1997). BRCA1 has been shown to physically associate with p53 and co-activate p53-responsive genes (Ouchi et al., 1998; Zhang et al., 1998). Furthermore, its presence in the RNA polymerase II holoenzyme (Scully et al., 1997; Anderson et al., 1998), its interaction with the histone deacetylase complex (Yarden and Brody, 1999) and the coactivatior CBP/p300 (Pao et al., 2000), ®rmly establish its role in transcription regulation. Its ability to inhibit the transcriptional activating function of estrogen-receptor-a provides a possible explanation as to why the mammary gland is the major target organ of tumorigenesis when BRCA1 is mutated (Fan et al., 1999). Several lines of evidence support the hypothesis that BRCA1 functions as a tumor suppressor protein. Most mutations found in BRCA1 result in truncation of and therefore non-functional protein product (Breast cancer information core, http://www.nhgri.nih.gov/Intramural_research/Lab_transfer/Bic/). In addition, mutations are frequently accompanied by loss of the wild type allele in familial breast and ovarian cancer (Smith et al., 1992; Neuhausen and Marshall, 1994; Cornelis et al., 1995). Increased expression of BRCA1 blocks the induction of tumors in nude mice bearing xenografts of the human MCF7 breast cancer cell line (Holt et al., 1996). Furthermore, reduction of BRCA1 expression by antisense RNA results in an increase in cellular proliferation and transformation of NIH3T3 ®broblasts (Thompson et al., 1995; Rao et al., 1996). The generation of mammary gland speci®c knockout of BRCA1 in mice provides the ultimate evidence for its tumor suppressor activity (Xu et al., 1999). More than 700 mutations have now been reported throughout the entire BRCA1 coding sequence (Breast cancer information core). However, the role of BRCA1 in sporadic cancers is not clear since somatic mutations of the gene are very rare in sporadic breast or ovarian

Transcriptional repressor element of BRCA1 T-C Suen and PE Goss

cancers (Futreal et al., 1994; Hosking et al., 1995; Merajver et al., 1995; Berchuck et al., 1998; Khoo et al., 1999; van der Looij et al., 2000). Loss of, or lowered, BRCA1 expression is frequently found in sporadic breast tumors as compared to surrounding normal tissue (Thompson et al., 1995; Magdinier et al., 1998; Sourvinos and Spandidos, 1998; Wilson et al., 1999). An epigenetic mechanism such as methylation of the promoter has been proposed as a mechanism responsible for lowered BRCA1 expression (Dobrovic and Simpfendorfer, 1997; Mancini et al., 1998; Magdinier et al., 1998; Rice et al., 1998; Catteau et al., 1999). However, alterations in methylation pattern between tumor and normal tissues have been found in only a small percentage (Dobrovic and Simpfendorfer, 1997; Mancini et al., 1998; Rice et al., 1998; Catteau et al., 1999) of cases or not at all (Magdinier et al., 1998). It therefore remains necessary to determine important regulatory sequences of BRCA1 transcription. We (Suen and Goss, 1999) and others (Xu et al., 1997a; Thakur and Croce, 1999) have performed functional studies of the BRCA1 promoter. We recently localized an 18 bp DNA element within the minimal BRCA1 bidirectional promoter whereupon nuclear protein binding is dependent in order for BRCA1 transcription to occur (Suen and Goss, 1999). We report here a second determinant of BRCA1 expression, located more than 500 bp into the ®rst intron of BRCA1. This cis-acting regulatory element functions as a repressor of transcription in multiple cell lines representing various tissues of origin. Protein-DNA complexes formed between this putative repressor element and nuclear extracts isolated from multiple cell lines were detected by electrophoretic mobility shift assays (EMSAs). Additional EMSAs with smaller fragments within the repressor region localized protein binding to a 36 bp DNA, which was able to confer a strong repressor activity on a heterologous thymidine kinase promoter. Analysis of extended promoter constructs going in the opposite direction suggests only transcription in the BRCA1 direction is controlled by this putative repressor element. Transcription in the direction of the neighboring NBR2 gene, which shares the same promoter and transcribes in the opposite direction, is unaected.

direction. Consistent with our previous study (Suen and Goss, 1999), the intergenic sequences between the EcoRI and SstI sites were required for the high level of expression detected with constructs 2, 3 and 7. Only a low background level of expression (construct 8) was observed when this intergenic region was excluded (constructs 4 and 6 in Figure 1). Transcriptional activity was maintained when the promoter was extended 3' into the ®rst exon and also included 172 bp of the ®rst intron of BRCA1 (construct 2, and several other constructs in Suen and Goss, 1999). Complete abolition of promoter activity was however observed, when a further 3' ± 820 bp sequence from the ®rst intron of BRCA1 was included (constructs 1 and 5). These data show that transcriptional activity from the BRCA1 promoter was blocked by DNA sequences between the NruI and XbaI sites. Transcriptional repressor activity could be detected in various cell lines that express BRCA1 To con®rm the intronic transcriptional repressor activity, the four reporter gene constructs shown in Figure 2 were transfected into three dierent cell lines which were shown to express BRCA1 in our previous study (Suen and Goss, 1999). The pBR(SpeI ± XbaI)CAT (construct 1) was found to express 36 ± 48-fold lower activity than pBR(SpeI ± NruI)CAT (construct 2) in all three cell lines. When the activity of pBR (EcoRI ± BamHI)CAT (construct 3) was compared to that of pBR(EcoRI ± SstI) CAT (construct 4), a 12-fold suppression was observed in both Caco2 and HeLa cells, while a less pronounced but signi®cant fourfold suppression was seen in the breast cancer cell line MDA-MB453. Localization of the transcriptional repressor activity to a smaller fragment In order to localize the repressor activity to a smaller region, a series of progressive 3' end deletions were introduced from the XbaI site near the end of intron 1, and their activities were determined in Caco2 cells (Figure 3a). Consist ent with resul ts from the last two experiments, pBR(SpeI ± XbaI)CAT (construct 1) expressed only at background level, similar to the empty vector (construct 9). A strong increase in CAT activity was observed when 3' deletion was extended to the AvaII site at 2255 (compare constructs 3 and 4). A similar pro®le of CAT activities was observed with these constructs when they were analysed in HeLa cells (data not shown). These data suggest that the sequence between the AvaII and A¯II sites was responsible for transcriptional repression of the BRCA1 promoter. Sequence of the putative repressor element A 90 bp NlaIV ± NlaIV (2253 ± 2342) DNA fragment that encompasses the putative repressor activity (as detected in Figure 3a) was cloned. It is located at 553 bp downstream from the ®rst nucleotide of intron

441

Results Detection of a transcriptional blocking activity in the first intron of BRCA1 A 56 bp DNA located within the intergenic region between BRCA1 and its neighboring gene NBR2 could function as a bi-directional minimal promoter (Suen and Goss, 1999). Reporter gene constructs containing the 56 bp minimal region (its relative position is shown by the closed box on the 2.7 kb PstI ± XbaI fragment in Figure 1; nucleotides are numbered the same way as a genomic BRCA1 sequence that was deposited in the GenBank with accession no. U37574) displayed a tissue-speci®c transcriptional activity in the BRCA1

Oncogene

Transcriptional repressor element of BRCA1 T-C Suen and PE Goss

442

Figure 1 The intron 1 of BRCA1 contains a transcriptional blocking activity. The top line shows the genomic organization of BRCA1 and its neighboring gene, NBR2. Exons are marked as open and shaded boxes, and transcription proceed toward the right and left sides for BRCA1 and NBR2, respectively. The translation start site for BRCA1 is located in exon 2, while that of NBR2 is located in exon 3 (not shown). The relative position and restriction map of a 2.7 kb PstI ± XbaI fragment is shown below the genomic scheme. Nucleotides are numbered the same way as a Genbank sequence (accession no. U37574). The closed box represnts a 56 bp minimal region which can function as a bi-directional promoter for the two divergently transcribed genes (Suen and Goss, 1999). The asterisk at position 1702 marks the beginning of intron 1 of BRCA1. Numbered solid rightward pointing arrows (constructs 1 ± 7) correspond to the indicated restriction fragments that were cloned into the empty CAT vector (construct 8) and driving transcription in the BRCA1 direction. These constructs were transfected into HeLa cells and a typical result of CAT assay is shown. Activities of the constructs are shown as a relative number to that of construct 2, which was assigned as 100. Experiments were repeated three times and a s.d.510% was observed

1 of BRCA1 (nucleotide 1702). Interestingly, this putative repressor region contains three GA-rich sequences (each longer than 10 bp and marked by striped boxes above the nucleotides in Figure 3b), which are potential binding sites for both the ets (Sharrocks et al., 1997) and the Sp1 (Philipsen and Suske, 1999) families of transcription factors. Specific nuclear proteins bind to the putative repressor element To determine if there are protein transcription factors that bind to the putative repressor element, the 84 bp NlaIV ± NlaIV DNA (Figure 3b) was labeled and subjected to EMSAs (Figure 4a). Several slow migrating bands representing protein-DNA complexes were observed when the NlaIV ± NlaIV fragment (lane 1) was incubated with nuclear extract isolated from HeLa cells (lane 2). The strong competitive eect of a 100-fold excess of the same unlabeled fragment (lane 3), but not a non-speci®c DNA (ns, lane 4) suggests that a major (marked with a solid arrow in all EMSAs, likely to be equivalent to C3 in Figure 4d) and a minor (marked with open arrowhead, likely to be equivalent to C1 in Figure 4d) bands were protein-DNA complexes that form speci®cally with this fragment. Additional bands representing other nuclear proteins (such as C2 and C4 in Figure 4d) that bind to this DNA were detectable after longer exposure (not shown and Figure 4d). To localize the protein-binding activity to a smaller region, the labeled NlaIV ± NlaIV fragment was cut
Oncogene

with the restriction enzymes BstNI or A¯II; and the four resul ting fragments (Figure 4b, fragmen ts 2 ± 5) were gel-puri®ed and subjected to EMSAs. Speci®c binding similar to those bands detected with the NlaIV ± NlaIV fragment (Figure 4a) was observed with the BstNI ± NlaIV fragment (Figure 4c, lanes 4 ± 6). Weak but speci®c binding was also detected with the NlaIV ± A¯II fragment (lanes 1 ± 3, position marked by a solid arrow. A long exposure of the dotted region (lanes 2 and 3) revealed another band migrating slightly faster than the one marked with the solid arrow (boxed region shown on the left), thus matching the two lower bands (one major and one minor) as detected with the BstNI ± NlaIV fragment (lane 5). The faint band detected with the A¯II ± NlaIV fragment (its position is marked by an asterisk) was determined to be non-speci®c as the addition of the unlabeled competitor had no eect on the intensity of this band (lanes 11 and 12). Speci®c binding of nuclear proteins to the BstNI ± NlaIV fragment was further con®rmed by its strong competitive eect against formation of protein-DNA complexes with the NlaIV ± NlaIV fragment (Figure 4d, lane 3), as compared to that of the NlaIV ± NlaIV fragment itself (lane 2) and the adjacent NlaIV ± BstNI fragment (lane 4). Delineation of the protein-binding activity to a 36 bp fragment We next attempted to delineate a minimal region that could account for the protein-binding activity. The

Transcriptional repressor element of BRCA1 T-C Suen and PE Goss

Figure 2 Putative repressor activity is found in various cell lines. The four CAT constructs (constructs 1 ± 4) were transfected into the three indicated cell lines. The activities of the correspondingly numbered constructs are shown on top of a typical CAT assay. As in Figure 1, the closed box represents the minimal promoter region and the asterisk marks the beginning of intron 1 of BRCA1. Activity of the pBR(EcoRI ± SstI)CAT (construct 4) was assigned as 100 for reference. No cross comparison of activities should be made among the cell lines, as transfection was not normalized as such

62 bp BstNI ± NlaIV fragment (Figure 4b, fragment 3) was cut by BseRI, the resulting 36 bp BstNI ± BseRI and 26 bp BseRI ± NlaIV fragments (fragments 6 and 7, respectively in Figure 4b) were cloned. These fragments were ®rst tested for their ability to compete against formation of protein-DNA complexes with the BstNI ± NlaIV fragment (Figure 5a). In agreement with its strong-competitive eect against formation of proteinDNA complexes between HeLa nuclear extract and the NlaIV ± NlaIV fragment (Figure 4d), the BstNI ± NlaIV fragment yielded the same pattern of binding (proteinDNA complexes labeled as C1 ± C4) as detected with the NlaIV ± NlaIV fragment (compare lane 2 of Figure 5a with lane 1 of Figure 4d). A progressive loss of eectiveness in competition was observed from the BstNI ± NlaIV (lane 3) to BstNI ± BseRI (lane 4), and BseRI ± NlaIV (lane 5) fragments. These data were con®rmed by testing the individual fragments in a separate EMSA (Figure 5b). The labeled BstNI ± BseRI DNA (lanes 1 ± 4) was able to form the two faster migr ating co mplexes C3 and C4 (Figure 5b, lane 2). These complexes were strongly competed away by the inclusion of a 100-fold excess of its unlabeled self (lane 3), and less eectively by a similar quantity of the adjacent BseRI ± NlaIV fragment (lane 4). The crosscompetitive eects of the BstNI ± BseRI and BseRI ± NlaIV fragments and the apparent dierence in their eectiveness in competition against the formation of protei n-DN A comp lexes (Figure 5a and lanes 3 and 4 in Figure 5b) was also re¯ected by the much weaker ability of the BseRI ± NlaIV fragment to form speci®c

443

Figure 3 Localization of a putative repressor region and its sequence. (a) A series of deletion was obtained by cloning the indicated restriction fragments (shown as solid rightward pointing arrows, constructs 1 ± 8) into the CAT vector pMT.IC3 (construct 9). A typical result of a transfection into Caco2 cells and subsequent CAT assay is shown. The activities of all constructs are expressed as a relative number to that of pBR(SpeI ± NruI)CAT (construct 6) which was chosen as a reference and assigned an activity of 100. Note that this construct was used in all CAT assays in this study so that comparison of relative activities of all constructs is possible after simple calculations. (b) The sequence between the AvaII (nucleotide 2255) and BamHI (nucleotide 2337) sites, which was mapped to contain the repressor activity in (a). Nucleotides are numbered the same way as a Genbank sequence (accession no. U37574). Several GA-rich sequences are marked by striped boxes above nucleotides and represent potential binding sites for ets or Sp1 families of transcription factors
Oncogene

Transcriptional repressor element of BRCA1 T-C Suen and PE Goss

444

Figure 4 Nuclear proteins bind to the putative repressor element. (a) An 84 bp NlaIV ± NlaIV fragment which encompasses the putative repressor DNA element (sequence and restriction recognition sites are shown in Figure 3b) was labeled and analysed by an electrophoretic mobility shift assay (EMSA). Slower migrating complexes were detected when the probe (lane 1) was incubated with nuclear extract isolated from HeLa cells (lanes 2 ± 4). No additional DNA (`minus' sign, lanes 1 and 2), a 100-fold excess of the unlabeled self-fragment (`plus' sign, lane 3), or an irrelevant DNA (ns for nonspeci®c , lane 4) was added as competitor to determine the speci®city of the protein-DNA complexes (a major band marked by the solid arrow; a minor band marked by the open arrowhead can be seen more readily with longer exposure). (b) Schematic representation of the dierent fragments used in additional EMSAs. The NlaIV ± NlaIV fragment (fragment 1) is shown on top with positions of the relevant restriction enzyme recognition sites marked. The striped boxes correspond to the positions of the three GA-rich sequences, marked the same way as in Figure 3b. Smaller restriction fragments that were used in additional EMSAs are shown below the NlaIV ± NlaIV fragment (fragments 2 ± 7). (c) The NlaIV ± NlaIV fragment was labeled on both ends, cut with either BstNI or A¯II, and the resulting restriction fragments (fragments 2 ± 5) were separated on a 6% polyacrylamide gel, visualized by autoradiography and puri®ed. An EMSA of the four indicated fragments (fragments 2 ± 5), NlaIV ± A¯II (lanes 1 ± 3), BstNI ± NlaIV (lanes 4 ± 6), NlaIV ± BstNI (lanes 7 ± 9), and A¯II ± NlaIV (lanes 10 ± 12) is shown. No nuclear extract was added for lanes 1, 4, 7, and 10. Either no competitor (lanes with `minus' sign) or a 100-fold excess of the unlabeled NlaIV ± NlaIV fragment (lanes with `plus' sign) was added to the incubation. A long exposure of the dotted region (lanes 2 ± 3) si shown to the left. (d) The NlaIV ± NlaIV fragment was labeled and incubated with nuclear extract isolated from HeLa cells in the absence (lane 1) or presence of an 100-fold excess of the indicated unlabeled DNAs as competitor (lanes 2 ± 4). The solid arrow and open arrowheads indicate the positions of the major and minor speci®c protein-DNA complexes, respectively, in all EMSAs

complexes in the same EMSA (Figure 5b, lanes 5 ± 8, marked by a solid arrow). This complex is likely to be C3 based on the banding pattern and its mobility
Oncogene

which is similar to C3 (minus the dierence in size of the two probes). GA-rich sequences could be found in both the BstNI ± BseRI and the BseRI ± NlaIV frag-

Transcriptional repressor element of BRCA1 T-C Suen and PE Goss

ments (Figure 3b). The vast number of transcription factors in the ets family which can recognize GA-rich sequences (Sharrocks et al., 1997) with dierent anities may underlie the cross-competitive eects of the two fragments against formation of protein-DNA co mplexes with the repres sor DNA (Figure 5). It is also possible that a weaker binding protein can only be detected when a stronger binding protein is competed away from binding to the same GA-rich region. This may explain the strong increase in intensity of a faster migr ating band (Figure 5b, lane 3, band marked with an asterisk) when the BstNI ± BseRI fragment was used to compete for formation of protein-DNA complexes against itself. The 36 bp BstNI ± BseRI fragment conferred a transcriptional repressor activity onto a heterologous promoter The combined results from EMSAs (Figures 4 and 5) suggest that the 36 bp BstNI ± BseRI fragment is responsible for two (a major complex C3 and a minor complex C4) of the four complexes which form on the putative repressor DNA (NlaIV ± NlaIV fragment). It was important to determine if it has functional eects on transcription. A reporter gene construct driven by a thymidine kinase (TK) promoter was used to test if repressor activity could be transferred onto a heterologous promoter. As shown in Figure 6, the BstNI ± BseRI fragment was able to suppress the TK promoter activity (construct 1) by 4 ± 12-fold when it was cloned immediately downstream in either orientation (constructs 2 and 3). This transcriptional repressive eect was however, not observed when the same DNA was cloned upstream to the TK promoter (data not shown). The neighboring gene transcribed in the opposite direction might also be negatively regulated by a repressor element in its intron The above result suggests that the repressor element may function in a position- or oreintation-dependent manner. To characterize the function of the repressor element within its native genomic alignment, we made use of the fact that BRCA1 is in close proximity with its neighboring gene (NBR2) (Xu et al., 1997b) and that they share a common promoter (Xu et al., 1997a; Suen and Goss, 1999). Multiple reporter constructs either including or excluding the putative repressor region were analysed for their activities. As shown in Figure 7a, all constructs excluding the putative repressor element (its position is indicated by a square and labeled with an X) were functional. Consistent with our previous work (Suen and Goss, 1999), promoter constructs transcribing in the NBR2 direction (constructs 4 ± 6) were always more active than their equivalent constructs transcribing in the BRCA1 direction (constructs 1 ± 3). However, when the repressor region was included, only transcription in the BRCA1 direction was stopped (Figure 7b, compare constructs 4 and 5), while transcription in the NBR2

direction was not aected (compare contructs 8 and 9 with construct 10). This suggests that the repressor region may function to stop transcription in the BRCA1 direction speci®cally. The construct which contains extended sequences from the NBR2 intron (construct 1) drove transcription

445

Figure 5 Binding of nuclear proteins is localized to a 36 bp BstNI ± BseRI fragment. (a) The BstNI ± NlaIV fragment (Figure 4b, fragment 3) identi®ed in the previous experiment (Figure 4c, lanes 4 ± 6) was labeled and incubated with HeLa nuclear extract. Similar pattern of binding (lane 2, one major complex C3 and three minor complexes C1, C2 and C4) was observed as with the NlaIV ± NlaIV fragment (Figure 4d). 100-fold excess of unlabeled self (lane 3), the BstNI ± BseRI (lane 4, fragment 6 in Figure 4b), or the adjacent BseRI ± NlaIV fragment (lane 5, fragment 7 in Figure 4b) was included to examine the speci®city of the proteinDNA complexes. (b) The BstNI ± BseRI (lanes 1 ± 4) and the BseRI ± NlaIV (lanes 5 ± 8) fragments were labeled and tested as in (a). The incubation was carried out in the absence (lanes with `minus' sign) or presence of the indicated unlabeled competitor (lanes 3, 4, 7 and 8). The solid arrow marks the same major complex (C3) that was detected in Figures 4 and 5a. An asterisk marks the position of a fast migrating complex that yielded stronger intensity only after self-competition (lane 3)

Figure 6 The 36 bp BstNI ± BseRI fragment is sucient for a repressor activity which could be transferred onto a heterologous promoter. The thymidine kinase (TK) promoter is shown as a solid arrow driving the CAT reporter gene (construct 1). The BstNI ± BseRI (shown as a block arrow) was cloned as in its native (construct 2) or opposite orientation (construct 3) downstream to the TK promoter. The three constructs were transfected into HeLa cells and the result of a typical CAT assay is shown. The activity of the TK promoter was assigned as 100 for comparison
Oncogene

Transcriptional repressor element of BRCA1 T-C Suen and PE Goss

446

Figure 7 Transcriptional repressor element is also found in the ®rst intron of the neighboring NBR2 gene and the two transcriptional repressor elements only aect expression of their respective genes. (a) Comparison of transcriptional activities of multiple constructs in their native genomic con®guration, driving expression in either BRCA1 (rightward pointing arrows, constructs 1 ± 3) or NBR2 direction (leftward pointing arrows, constructs 4 ± 6) when the BRCA1 intronic repressor region (approximate position is indicated by a square labeled with an X) was excluded. (b) Examination of further reporter constructs which include the BRCA1 intronic repressor region and driving in either the BRCA1 (constructs 2 and 4) or the NBR2 direction (constructs 7 ± 9). Constructs extending far into the intron 1 of NBR2 gene were also tested in their ability to drive transcription in either the BRCA1 (constructs 1 and 3) or NBR2 (constructs 6) direction. The intergenic promoter region (EcoRI ± SstI) driving transcription in the BRCA1 direction (construct 5) was used as a reference (activity assigned as 100)

in the BRCA1 direction as eciently as the intergenic sequence (construct 5). As shown in earlier experiments (Figure 1), constru ct 2 and 7 were non-funct ional because the intergenic region was not included (Suen and Goss, 1999). Most interestingly, when the PstI ± SstI (as in construct 1) fragment was tested in the opposite orientation (construct 6), the strong promoter activity in the NBR2 direction was extinguished (compare constructs 6 and 10). These results imply that transcription in the NBR2 direction may be controlled in a similar manner as is transcription in the BRCA1 direction, namely that the ®rst intron of the respective genes plays an important negative regulatory role in their transcription. Discussion An early study of the BRCA1 promoter suggested the existence of an alternative promoter located within the ®rst intron of the BRCA1 gene (Xu et al., 1997a). Recent studies by us (Suen and Goss, 1999) and others (Rice et al., 1998) have shown contradictory results and implied an insigni®cant role of such an alternative promoter. We now present strong evidence that the
Oncogene

®rst intron of BRCA1 functions in a negative regulatory manner in the control of BRCA1 transcription. A transcriptional blocking activity was ®rst located to an 819 bp NruI ± XbaI fragment within the ®rst intron of the BRCA1 gene (Figure 1). This repressor activity appears not to be tissue-speci®c as it could be found in multiple cell lines representing various tissues of origin (Figure 2). Although strong repressor activity could be mapped to a smaller 464 bp NruI ± BamHI fragment, longer sequences might be necessary to achieve the full potential of transcription repression in the breast cancer cell line MDA -MB453 (Figure 2). Further experiments with multiple mammary epithelial cell lines are required to determine if the observed eect is cell line-, tissue-, or cancer-speci®c. Nevertheless, analysis of the activities of a series of 3' deletion constructs con®rmed the existence of the putative repressor element and localized it to an 83 bp AvaII ± BamHI DNA fragment (Figure 3a). Since alternative splicing upstream of exon 2 of the BRCA1 gene has been described (Xu et al., 1995), the repressor element may be considered as either 553 bp downstream from the ®rst nucleotide of intron 1a, or 19 bp from intron 1b. To avoid confusion and allow easy

Transcriptional repressor element of BRCA1 T-C Suen and PE Goss

understanding by others, the BRCA1 genomic sequence deposited in Genbank (accession no. U37574) was chosen as a reference numbering system. In this sense, the AvaII ± BamHI fragment encompasses the sequence be tween 2254 and 2337 (Figure 3b). Analysis of the NlaIV ± A¯II and the BstNI ± NlaIV fragm ents by EMSAs (Figure 4c, lanes 1 ± 6) suggest s that binding of nuclear proteins to the putative repressor region can be localized to the overlapping region, between the BstNI and A¯II sites. However, the strong dierence in their abilities to form the same protein-DNA complexes (Figure 4c, compare lanes 2 and 3 to lanes 5 and 6) also indicates that possible interactions among the dierent fragments might play a role in the binding activity. It is possible that the NlaIV ± BstNI fragment may contain an inhibitory activity against proteins binding to the BstNI ± A¯II region, thus the observed weak binding activity of the NlaIV ± A¯II fragment. On the other hand, the A¯II ± NlaIV fragment may contribute positively to proteins binding to the BstNI ± A¯II region, and thus the observed strong binding to the BstNI ± NlaIV fragment. Although the two possibilities are not mutually exclusive, the fact that the BstNI ± NlaIV was a stronger competitor than the complete NlaIV ± NlaIV fragment in EMSAs (Figure 4d) favors the form er mecha nism. Furthermore, the presence of a GA-rich sequence within the NlaIV ± BstNI, but not the A¯II ± NlaIV fragm ent (Figures 3b and 4b ) may also contri bute to the apparent dierence in the binding activity (also see below, discussion on GA-rich sequences). EMSAs using various smaller fragments within this putative repressor element localized a 36 bp BstNI ± BseRI (2278 ± 2313) fragment which was capable of forming speci®c protein-DNA complexes eectively wi th nuclear extra cts isolated from HeLa cells (Figures 4 and 5) and Caco2 cells (data not shown). The apparent progressive lowering in the ability of the BstNI ± NlaIV and its two sub-fragments to compete against the formation of protein-DNA complexes with the BstNI ± NlaIV fragment (Figure 5a) might be explainable from the analysis of the NlaIV ± NlaIV sequence. GA-rich sequences are found throughout the NlaIV ± NlaIV fragment (Figure 3b) and they can be recognized by numerous transcription factors of the ets (Sharrocks et al., 1997) and Sp1 (Philipsen and Suske, 1999) families. The possible involvement of these proteins and their large range of anities for GA-rich sequences (Sharrocks et al., 1997) may explain the observed cross-competitive eects and protein-binding ability of the individual DNA fragments (Figure 5). It is impossible to speculate which one of these transcription factors may be responsible for the binding activity that we observed. More detailed characterization of the binding sites may provide some clues as to the identity of the binding factors. Cloning of transcription factors interacting with the putative repressor element might be necessary to reveal their identity. Nevertheless, the ability of the 36 bp BstNI ± BseRI fragment to attenuate the transcriptional activity of the heterologous TK promoter con®rmed the repressor

functi on (Figure 6). The apparent posit ional eec ts of this 36 bp repressor element distinguish it from a classical silencer element. Indeed, analyses of multiple reporter gene constructs in their native genomic alignment suggest that the BRCA1 intron repressor element does not aect transcription in the opposite direction (Figure 7). Functional analyses within the context of both the heterologous TK and the native BRCA1 promoters suggest the repressor element is only functional when it is located downstream to the promoter. Most interestingly, intron 1 of the NBR2 gene also appears to contain a repressor activity which blocks only transcription in the NBR2 direction (compare pBR(PstI ± SstI)CAT and pNB(SstI ± PstI)CAT in Figure 7b with constructs 1 ± 6 in Figure 7a). Further studies are necessary to con®rm and localize this repressor activity within intron 1 of NBR2. BRCA1 expression is known to be induced during the S phase of the cell cycle (Gudas et al., 1996; Vaughn et al., 1996) and altered by DNA damaging agents (Andres et al., 1998; Husain et al., 1998). The expression pattern of BRCA1 throughout development also suggests its importance in tissues that undergo rapid proliferation and terminal dierentiation (Lane et al., 1995; Marquis et al., 1995; Blackshear et al., 1998; Magdinier et al., 1999). Removal of the transcriptional block at the position of intron 1 which we have identi®ed in this study would be an eective way to increase BRCA1 levels in a timely manner for subsequent molecular events. Mutations in any important transcriptional regulatory element or its interacting transcription factor(s), regardless of whether it is a positive or a negative element, could result in an absence of BRCA1 expression, which has been reported in non-inherited breast cancer (Wilson et al., 1999; Thompson et al., 1995; Magdinier et al., 1998; Sourvinos and Spandidos, 1998). Recently, elevated levels of Brn-3b have been shown to correlate with reduced BRCA1 expression in mammary tumors (Budhram-Mahadeo et al., 1999). Our numerous BRCA1 promoter-reporter constructs will be very useful in identifying the possible site of interaction of this or other transcription factors that may regulate BRCA1 expression.

447

Materials and methods Enzymes and reagents Restriction enzymes and other DNA modifying enzymes such as T4 kinase, T4 polymerase, T4 ligase, Klenow fragment, and calf intestinal phosphatase were purchased from Life Technologies Inc., New England Biolabs (Mississauga, Ontario, Canada), Roche Molecular Biochemicals, or Amersham Pharmacia Biotech. Chemicals used for the chloramphenicol acetyltransferase (CAT) and b-galactosidase assays were purchased from Sigma-Aldrich Canada (Oakville, Ontario, Canada). Thin Layer Chromatography (TLC) plates were products of Eastman Kodak Co. Cell culture medium and reagents were obtained from Life Technologies Inc. All isotopes were products from Amersham Pharmacia Biotech.
Oncogene

Transcriptional repressor element of BRCA1 T-C Suen and PE Goss

448

Plasmids The plasid pBluescripts(IIKS) (Stratagene, La Jollla, CA, USA) was used for general subcloning purposes. pMT.1C3 is a plasmid containing multiple cloning sites placed upstream of the CAT gene (Suen and Hung, 1990). Most of the BRCA1 DNA restriction fragments were cloned into pBluescripts(IIKS), and were then shued into the matching unique restriction sites on the polylinker of pMT.IC3. DNA fragments were blunt ended with Klenow fragment or T4 polymerase when no appropriate restriction enzymes could be used for directional cloning. In addition, reversed orientation of a subcloned fragment in the pMT.IC3 plasmid could easily be obtained by cutting with HindIII, which ¯ank the polylinker, followed by religation. The 2.7 kb PstI ± XbaI fragment containing intron 1 and upstream sequences of BRCA1 gene has been described in our previous study (Suen and Goss, 1999). Various restriction fragments within this region were subcloned into both pBluescript(IIKS) and pMT.IC3. All the CAT constructs were named according to the direction of transcription followed in parenthesis by their ¯anking restriction sites in a 5'- to 3'-direction. Therefore the pBR- and pNB-series of CAT plasmids indicate a promoter transcribing towards the BRCA1 and NBR2 gene (Xu et al., 1997b), respectively. Putative repressor fragments from the BRCA1 intron were also cloned into the pBLCAT2 (Luckow and Schutz, 1987) reporter plasmid where a possible eect on the heterologous thymidine kinase promoter could be analysed. pCMVb (Clontech, Palo Alto, CA, USA), a plasmid which contains the lacZ gene driven by the cytomegalovirus enhancer (MacGregor and Caskey, 1989), was used for monitoring transfection eciency. Detailed maps of all the plasmids used in this study will be distributed along with the reagents upon request. Sequencing Dideoxy-sequencing of double-stranded plasmids was performed with a T7 polymerase sequencing kit using either [a-35S]dATP or [a-35S]dCTP (Amersham Pharmacia Biotech). Most of the BRCA1 promoter subclones, particularly those with DNA inserts of less than 300 bp in size, were con®rmed by sequencing. Sequencing primers for either the pMT.IC3 or pBluescript(IIKS) series of plasmids has been described previously (Suen and Goss, 1999). Cell culture All cell lines that were used in this study are available from American Type Culture Collection (ATCC, Manassas, VA, USA). This includes HeLa, a human cervical carcinoma cell line; Caco2, a human colon carcinoma cell line; and MDAMB453, a human mammary carcinoma cell line. All cell lines were cultured in Dulbecco's modi®ed Eagle's/F12 medium (Life Technologies Inc.), supplemented with 10% fetal calf serum, and kept in a humidi®ed, 378, 5% CO2 incubator. References Ande rson SF, Sc hleg el BP, Nakajima T, Wolpin ES and Parvin J D. (19 98). Nat. Genet., 19, 254 ± 2 56. An dre s JL , F an S , T urk el GJ , W an g J -A, T wu N-F , Yu an RQ, Lam szus K, G oldber g ID and Rosen EM. (1 998). Onc oge ne, 16, 2 229 ± 22 41.

Transfections and CAT assays A calcium phosphate precipitation method (Chen and Okayama, 1987) was used for transfection as modi®ed and described previously (Suen and Goss, 1999). Brie¯y, cells were split at a predetermined ratio into 100 mm tissue culture dishes (Falcon) the day before transfection. Unless otherwise indicated, 1 mg of pCMVb and 10 mg of a CAT reporter DNA were added to 0.5 ml of 0.25 M CaCl2. This was followed by dropwise addition of 0.5 ml of 26BBS buer (50 mM BES, 280 mM NaCl and 1.5 mM Na2HPO4) and gentle mixing. After 25 min at room temperature, the mixture of DNA-precipitate was added to the cells. The cells were incubated at 378C for 16 ± 20 h, after which they were washed three times with phosphate-buered saline, re-fed with fresh medium, and returned to the 378C incubator. Cells were washed and harvested after 20 ± 24 h and several freeze/thaw/ vortex cycles were carried out to lyse the cells. One-®fth of the cell lysate was used for the b-galactosidase assay using ONPG (O-Nitrophenyl-b-D-Galactopyranoside) as substrate. The results were used to adjust the amount of lysate for CAT assay. The thin layer chromatography (TLC) method of CAT assays was performed as previously described, where the standard [14C]chloramphenicol was replaced with 1-Deoxy[dichloro-acetyl-1-14C]chloramphenicol (Amersham Pharmacia Biotech.) (Suen and Goss, 1999). Electrophoretic mobility shift assays (EMSAs) EMSAs were performed as previously described (Suen and Goss, 2000). Nuclear extract was isolated from the dierent cell lines by means of homogenization under hypotonic conditions (Dignam et al., 1983). DNA fragments were isolated by digesting a plasmid subclone with appropriate restriction enzymes, gel puri®ed, and labeled with [a-32P]dATP or [a-32P]dCTP (depending on the restriction site) by Klenow-fragment. A ®nal volume of 30 ml of reaction mixture was added in the order of H2O, 106 binding buer (1610 mM Tris pH 7.5, 50 mM KCl, 1 mM DTT, 0.1 mM EDTA, 1 mM MgCl2, 5% glycerol), 3 mg of poly(dIdC).poly(dI-dC), 10 mg of nuclear extract, an appropriate amount of unlabeled competitor if desired, and ®nally, 20 000 c.p.m. of probe. The mixture was incubated at room temperature for 25 min, after which it was loaded onto a 6% native polyacrylamide gel. After electrophoresis, the gel was dried under vacuum in a gel dryer, and exposed to a Kodad XOMAT-AR ®lm at 7808C.

Acknowledgments T hi s wo r k w a s su pp or t ed b y a gr a n t f r om t he Br e a s t Cancer Pr evention Progr am of the Toronto H ospital.

Berchu ck A, Heron K-A, C arney M E, La ncaster J M, Fraser EG, V inson V L, Deenbaugh AM, Miron A, Marks J R, Futrea l PA and Fr ank TS. (1998 ). Clin Cancer Res., 4, 243 3 ± 2437 .

Oncogene

Transcriptional repressor element of BRCA1 T-C Suen and PE Goss

Bertwistle D and Ashworth A. (1998). Curr. Opin. Genet. Dev., 8, 14 ± 20 . B l a c k s h e a r P E , Go l d s w o r th y S M , Fo l e y J F , Mc A l l i s t e r KA , Bennett LM, Collins NK, Bunch DO, Brown P, Wiseman RW and Dav is BJ. (1 998). O ncog ene, 16, 61 ± 68 . B ud hra m-Ma had eo V, Ndisa ng D , Wa rd T , We be r BL an d Latchman DS. (199 9). Onc oge ne, 18, 6 684 ± 6 691. Case y G. (19 97). Curr . Opin. Onco l., 9, 88 ± 93 . Cattea u A, Harris WH , Xu C-F a nd Solomon E. (19 99). Onc ogen e, 1 8, 1 957 ± 19 65. Chapm an MS and Ve rma IM. (1 996). Nature, 3 82, 678 ± 6 79. Chen C and Okayam a H. (198 7). Mol. Cell. Biol., 7, 27 45 ± 27 52. C h e n Y , F a r m e r A A , C h e n C - F , J o n e s DC , C h e n P -L an d Lee W-H . (199 6). Canc er Re s., 56 , 31 68 ± 317 2. Cornelis R, N euhausen SL, Joha nsson O, Arason A, Kelsell D, Ponde r BAJ, Tonin P, Ham ann U, Lindblom A, Lalle P , L on g y M , O l a h E, Sch er n ec k S, Bi g n on Y -J , S ob ol H, Cha ng-Claude J, La rsson C, Spurr N, Borg A, Barkard ottir RB , N arod S , D evilee P a nd the B reast Canc er Linkage Consortium . (1995 ). Ge nes Chrom. Can cer, 13, 20 3 ± 210. Cressman VL, B acklund DC, Avrutskaya AV, L eadon SA, Godfre y V and Koller B H. (199 9). M o l . Ce l l . B i o l . , 19, 70 61 ± 707 5. Dignam JD, Leibovitz RM and Roeder RG. (1983). Nucleic Acids Res., 11, 1475 ± 1 489. D eng C-X a nd Sc ott F. (20 00), On coge ne, 19, 1059 ± 1064. D obrovic A and Simpfe ndorfer D. (1 997). Cancer Res., 57, 33 47 ± 335 0. F an S, W an g J-A, Yu an R , Ma Y, Me ng Q, Erd os M R, Pestell RG , Yu an F , Aub orn KJ, G oldb erg ID and Rose n EM. (1999 ). Sc ienc e, 2 84, 1354 ± 1 356. Ford D, Easton DF, S tr atton M , N arod S, G oldgar D, De vilee P, B ish op DT , We ber B , L en oir G, Chan g-Claud e J, Sobol H, Tear e MD, Struewing J , Arason A, Sc hernec k S, Peto J, Rebbec k T R, Tonin P, Neu hausen S, Barka rdottir R, Ey fjord J , Lync h H, Ponder BAJ, Gayther SA, B irch JM, Lindblom A, S toppa-Lyonnet D, Bignon Y, Borg A, Ham ann U, H aites N, Scott R J, Maugard CM, Vase n H and the Breast Ca nce r Linkage Consortium. (199 8). Am. J . Hu m. Gene t., 62 , 676 ± 689. F u t r e a l P A , L i u Q , Sh a t t u c k - E i d e n s D , C o c h r a n C, H a r s h ma n K , T avtigan S, Bennett LM, Ha uge n-Stra no A, S we nse n J, M i ki Y , E dd i ngt o n K, M cC l ure M, F rye C , W eav er-Fe ldh aus J , Ding W , Gh olam i Z, So de rkvis t P, Terry L, Jhanwar S, Be rchuck A, Igle hart JD, Ma rks J, Ballinger DG, Barrett JC, S kolnick MH, Kamb A and Wisema n R. (199 4). Science, 266, 120 ± 122. G ud as JM , L i T, N guye n H, J ense n D, R au s ch er I I I F J an d Cowan K H. (1996). Cell Growth Dier., 7, 7 17 ± 723 . H aile D T a nd P arvin JD. (1 999). J . Bio l . Chem . , 2 74, 21 13 ± 21 17. H olt JT , T ho mpso n M E , Sza bo C, R ob in so n-Be nio n C, Artea ga CL, King M-C and Jensen RA. (1996). N a t . Genet., 12 , 298 ± 302. Hosking L , T rowsdale J, Nicolai H, Solomon E, Foulkes W , Sta mp G, Signe r E and Jeery A . (19 95). Nat. Genet., 9, 34 3 ± 344. H usain A, H e G , Venkatram an E S and Sprigg s DR. (1 998). Canc er Re s., 58 , 11 20 ± 112 3. Khoo U-S, Ozcelik H, Cheung ANY, Chow LWC, Ngan HYS, D one SJ, Lia ng ACT, Chan VWY, Au GK H, Ng W-F, Poon CSP, Leung Y-F, Loong F, Ip P, Chan GSW, Andrulis IL, Lu J and Ho FCS. (1999 ). Oncogene, 18, 46 43 ± 464 6.

L ane T F, Den g C, E lson A, Lyu M S and Ko zak CA. (19 95). Genes Dev., 9 , 271 2 ± 2722 . Luc kow B and Schutz G. (1987 ). Nuc leic Acids Re s., 1 5, 549 0. M a c G r e g o r G R an d C a s k e y C T . ( 1 9 8 9 ) . Nucle ic A cids Re s., 17, 2365 . Mancini D N, Rodenhiser DI, Ainsworth PJ, O'Malley FP, Singh S M, Xing W a nd Arc her TK. (1998). Onc ogen e, 1 6, 116 1 ± 1169 . Magdinier F, Ribieras S, L enoir GM, Fr appart L and Dante R. (19 98). O ncog ene, 17, 3169 ± 3176. Magdinier F, Venezia ND, Lenoir GM, Frappart L and Dante R. (199 9). Onc ogen e, 1 8, 4 039 ± 40 43. Marquis ST, Rajan JV, Wy nshaw-Boris A, Xu J, Yin G-Y, Abe l K J, We ber BL and C hodosh LA. (1 995). Nat. Genet., 11, 17 ± 26 . M e r a j v e r S D , P ham TM , C a d u RF, C h e n M, P oy EL , C oo ney K A , We ber BL , C o ll i n s F S, Joh ns t o n C and Fr ank TS. (1995). Nat. Genet., 9 , 439 ± 443. M i ki Y , Swen sen J, S ha ttu ck -E i de ns D, F utrea l P A, Hars hman K , T avtiga n S, Liu Q, C och ran C, Be nne tt L M, Din g W, Bell R, Rosenthal J, Hussey C, Tran T , M cClur e M, F r y e C, H a t t i e r T , P h e l p s R , H a u g e n - S t r a n o A , K a t c h e r H, Yak um o K, Gh olam i Z, Sh a er D, S ton e S, B ayer S , Wray C, Bogden R, Dayananth P, Ward J, Tonin P, Nar od S, B ristow PK, N orris FH , Helv ering L, Morrison P, Rosteck P, Lai M, Barrett JC, Lewis C, Neuhausen S, Cannon-Albright L, Goldg ar D , Wis eman R, Kamb A and Skolnick MH. (19 94). Science, 26 6, 66 ± 71. Monteiro ANA, Augus t A and H anafusa H . (1996 ). Pro c. Natl. Acad. Sci. USA, 93, 1359 5 ± 135 99. Moy nahan ME, C hiu JW, Koller BH and Jasin M. (19 99). Mol. Cell., 4, 5 11 ± 518 . Ne uhaus en SL and Marshall CJ. (1994). Ca nce r Res., 5 4, 606 9 ± 6072 . O u chi T , M on te i r o A N A , A ugu s t A , A ar o ns o n S A and Hanafus a H. (1 998). Proc. Natl. Acad . Sci. USA, 9 5, 230 2 ± 2306 . Pao GM , Jan kne ch t R, Rune r H , Hu nter T an d Verma IM . (2000 ). Proc. Natl. Aca d. Sci. USA, 97, 1020 ± 1 025. Philipsen S and Suske G. (1999 ). Nucle ic Acid s Res., 2 7, 299 1 ± 3000 . Rao VN, Shao N, Ahmad M and Reddy ESP. (1996). Onc ogen e, 1 2, 52 3 ± 528. Ric e JC, Ma ssey-Brown K S and Futscher BW. (199 8). Onc ogen e, 1 7, 18 07 ± 181 2. Runer H and Verm a IM. (1997 ). Proc . Natl. Aca d. Sci. USA, 94, 7138 ± 7143 . Scully R, Anderson S F, Chao DM, Wei W, Ye L, Young RA, Liv ingston DM and Parv in J D. (19 97). Proc. Natl. Aca d. Sci. USA, 9 4, 56 05 ± 56 10. Scully R, Ganesan S, Vlasakova K, Chen J, Socolovsky M and Livingston D M. (1999). Mol. Cell., 4, 10 93 ± 109 9. S harr ock s A D, Br own A L , L i ng Y a nd Ya te s PR. (19 97). Int. J. Bioc hem. Cell. Biol., 2 9, 1 371 ± 13 87. S hen S -X, W eave r Z , Xu X, L i C, W ein stein M , Chen L , Guan X-Y, Rie d T and Deng C-X. (1998 ). Onc ogen e, 1 7, 311 5 ± 3124 . Smith SA, Easton DF, Evans DGR and Ponder BAJ. (1992). Nat. Genet., 2 , 12 8 ± 131. Somasundaram K, Zhang H, Z eng Y-X, Houvras Y, Peng Y, Zh an g H, Wu G S, L ich t JD , W ebe r B L and E l-Deiry WS . (1997 ). Nature, 38 9, 1 87 ± 190 . Sourvinos G a nd S pandidos DA. (1998 ). Bioche m. Bioph ys. Res. Comm., 24 5, 75 ± 80.

449

Oncogene

Transcriptional repressor element of BRCA1 T-C Suen and PE Goss

450

Suen T-C and Hung M-C. (1990). Mol. Cell. Biol., 10, 6306 ± 63 15. Sue n T-C and Goss PE. ( 199 9). J . B i ol . Ch em . , 274, 312 97 ± 31 304. Sue n T-C and Goss PE. (200 0). J. Bio l. Che m., 275, 66 00 ± 66 07. Tha kur S and Croce CM. (1999). J. Biol. Che m., 27 4, 88 37 ± 88 43. Thompson ME, Jensen RA, Obermiller PS, Page DL and H o l t J T. (1 9 9 5 ) . N a t . G e n e t . , 9, 444 ± 4 50. v an der Lo oij M , Cleto n-Ja nse n A-M , va n E ijk R , Mo rrea u H, van Vliet M, Kuipers-Dijks hoorn N, Olah E, C o r ne l i s s e C J a nd De vi l e e P. (2 0 0 0 ) . G e n e s Ch r o m . Canc er, 27, 2 95 ± 30 2. V a u g h n J P, Da v i s P L, Ja r b oe M D , H up er G, E v a n s A C , Wisema n RW, Ber chuck A, Ig leha rt JD , Futreal PA and M a r k s JR . ( 1 9 9 6 ) . C e l l Gro w t h D i e r . , 7 , 71 1 ± 715. W i l s o n C A , R a m o s L , V i l l a s e n o r MR , A n d e r s K H , Pre s s MF, Clarke K , K arlan B , Chen J-J, Scully R, Livingston D, Zuch RH, Kanter M H, Cohen S, Ca l zone FJ and S l a m o n D J. (1 9 9 9 ) . N a t . G e n e t . , 21 , 236 ± 240.

X u C - F , B r o wn MA , C ha m be r s J A , Gr i t h s B , N i c o l a i H and Solomon E. (199 5). Hum . Mo l. Gene t., 4, 22 59 ± 226 4. Xu C-F, Chambers JA and Solomon E. (1997a). J. Biol. Chem., 272, 209 94 ± 209 97. X u C - F , B r o w n M A , N i c o l a i H, Ch a m b e r s JA , G r i t h s BL and Solom on E. (1997b). Hum. Mol. Ge net., 6, 10 57 ± 106 2. Xu X, Wagner K-U, Larson D, W eaver Z, Li C, Ried T , He nn ig hau sen L, W yn sha w-B oris A and Den g C-X. (199 9). Nat. Ge net., 22, 3 7 ± 43. Ya rden RI a nd Br ody LC. (1 999). Proc . Natl. Acad . S ci. USA, 96, 4983 ± 4988 . Z han g H, Som asu nda ram K, Pe ng Y, T ian H, Z han g H , B i D, Weber BL and El-Deir y WS. (1998 ). O ncog ene, 1 6, 171 3 ± 1721 .

Oncogene