Brca2 Is Required for Embryonic Cellular Proliferation in the Mouse

Downloaded from genesdev.cshlp.org on October 4, 2021 - Published by Cold Spring Harbor Laboratory Press

Brca2 is required for embryonic cellular proliferation in the mouse

Akira Suzuki, 1'2'6 Jos6 Luis de la Pompa, 1'2'6 RazqaUah Hakem, 1'2 Andrew Elia, 1'2 Ritsuko Yoshida, 1'2 Rong Mo, 3'4 Hiroshi Nishina, 1'2 Tony Chuang, 1'2 Andrew Wakeham, ~'2 Annick Itie, ~'2 Wilson Koo, ~'2 Phyllis Billia, ~'2 Alexandra Ho, 1'2 Manabu Fukumoto, 5 Chi Chung Hui, 3'4 and Tak W. Mak 1'2"7 1Amgen Institute, Toronto, Ontario, Canada M5G 2C1; 2Ontario Cancer Institute, Department of Medical Biophysics and Immunology, 3Department of Molecular and Medical Genetics, University of Toronto, Toronto, Ontario, Canada; 4program in Developmental Biology and Division of Endocrinology Research Institute, The Hospital for Sick Children, Toronto, Ontario, Canada M5G 1X8; SDepartment of Pathology, Graduate School of Medicine, Kyoto University, Kyoto, Japan 606

Mutations of the tumor suppressor gene BRCA2 are associated with predisposition to breast and other cancers. Homozygous mutant mice in which exons 10 and 11 of the Brca2 gene were deleted by gene targeting (Brca2 I°-11) die before day 9.5 of embryogenesis. Mutant phenotypes range from severely developmentally retarded embryos that do not gastrulate to embryos with reduced size that make mesoderm and survive until 8.5 days of development. Although apoptosis is normal, cellular proliferation is impaired in Brca2 ~°-H mutants, both in vivo and in vitro. In addition, the expression of the cyclin-dependent kinase inhibitor p21 is increased. Thus, Brca2 l°-H mutants are similar in phenotype to Brcal 5-6 mutants but less severely affected. Expression of either of these two genes was unaffected in mutant embryos of the other. This study shows that Brca2, like Brcal, is required for cellular proliferation during embryogenesis. The similarity in phenotype between Brcal and Brca2 mutants suggests that these genes may have cooperative roles or convergent functions during embryogenesis. IKey Words: Brca2 mutant mice; proliferation; embryogenesis; p21] Received March 26, 1997; revised version accepted April 14, 1997.

The control of cell proliferation is achieved through a In the breast, the observation that endocrine factors that balance between negative and positive regulators of control breast development (Kleinberg and Newman growth (for review, see Hunter 1997). Tumor suppressors 1986) also influence breast cancer risk (Siiteri et al. 1986) are negative regulators of growth, and genetic lesions suggests that mammary gland development and carcino- that inactivate them release the cells from normal genesis are processes that are fundamentally related. growth constraints, causing the deregulated proliferation BRCA2 is a breast cancer susceptibility gene recently of cancer cells. Loss of growth control is accompanied by isolated by positional cloning (Wooster et al. 1995). Mu- alterations in normal pathways of differentiation and de- tations in BRCA2 are thought to account for as many as velopment. Tumor suppressor genes have been shown to 35% of all inherited breast cancers, as well as a propor- encode proteins involved in growth control and DNA tion of inherited ovarian cancers (Wooster et al. 1995; repair processes. Germ-line mutations in tumor suppres- Tavtigian et al. 1996). Like BRCA1, BRCA2 appears to be sor genes such as p53 (Malkin et al. 1990; Srivastava et a tumor suppressor gene, because the loss of the wild- al. 1990), Wilms' tumor (WT1)(Gessler et al. 1990), and type BRCA2 allele in heterozygous carriers can result in retinoblastoma (RB) (Hansen et al. 1985) are associated the development of tumors (Collins et al. 1995; Gud- with inherited predispositions to cancer. Targeted muta- mundsson et al. 1995). Few somatic mutations in either tion of these genes in mice causes cancer susceptibility BRCA1 (Futreal et aI. 1994) or BRCA2 (Miki et al. 1996) but also results in abnormalities in cellular proliferation have been identified in sporadic breast tumors, but germ- and differentiation and embryonic development (Done- line mutations in both BRCA1 (Miki et al. 1994) and hower et al. 1992; Lee et al. 1992; Kreidberg et al. 1993). BRCA2 (Phelan et al. 1986) predispose carriers to breast adenocarcinoma. However, unlike BRCA1, germ-line mutations in BRCA2 predispose both males and females to breast cancer, and female carriers of BRCA2 germ-line mutations show a lower incidence of ovarian cancer than 6These authors contributed equally to this work. 7Corresponding author. do BRCA1 carriers (Thorlacius et al. 1995; Wooster et al. E-MAIL [email protected]; FAX (416) 204-5300. 1995).

1242 GENES & DEVELOPMENT 11:1242-1252 © 1997 by Cold Spring Harbor Laboratory Press ISSN 0890-9369/97 $5.00 Downloaded from genesdev.cshlp.org on October 4, 2021 - Published by Cold Spring Harbor Laboratory Press

Brca2 in murine embryonic cellular proliferation

The BRCA2 gene consists of 27 exons that generate an duce homozygous mutant offspring. The genotypes of 11-kb mRNA transcript predicted to encode a novel pro- the mice were confirmed by Southern blot analysis. tein of 3418 amino acids (Tavtigian et al. 1996). Interest- ingly, the BRCA1 and BRCA2 genes are similar in several Phenotype of heterozygous Brca2 ~°-1~ mice respects, despite a marked lack of nucleotide or amino acid sequence homology. Both BRCA1 and BRCA2 are large Mice heterozygous for the Brca2 ~°-~ deletion were phe- AT-rich genes, which include an exon of unusually large notypically normal and fertile. Heterozygous Brca2 ~°-~1 size. Neither gene shows homology to any known protein. mice did not develop any type of cancer up to 7 months The numerous germ-line mutations that have been iden- of age. It remains possible, however, that Brca2 ~°-~ ~ het- tified in both BRCA1 and BRCA2 are distributed through- erozygotes could develop tumors at a more advanced age. out their lengths, and most of them result in protein trun- cation (Gayther et al. 1997). In addition, the spatial and Brca21°-~ 1 mutation results in embryonic lethality temporal pattern of Brca2 mRNA expression in the mouse is strikingly similar to that of Brcal during fetal develop- No viable Brca2 I°-~l pups were identified among 51 ment, and in adult tissues in vivo and in mammary epi- offspring born form heterozygous intercrosses, indicat- thelial cells in vitro (Rajan et al. 1996). Both Brcal and ing that homozygosity for the Brca21°-~ mutation Brca2 are expressed in a cell cycle-dependent manner, causes embryonic lethality (Fig 1B; Table 1). To assess peaking at the G~/S boundary. Nevertheless, because there the consequences of the Brca2 ~°-~ mutation on embry- are no obvious extended regions of nucleotide or amino onic development, we analyzed embryos from heterozy- acid homology between BRCA1 and BRCA2, whether gote intercrosses at different days of gestation (Table 1). these similarities reflect an underlying structural or func- Genomic DNA was isolated from yolk sacs or from tional homology remains to be determined. Taken to- whole embryos, and genotyping was performed by PCR gether, the available data suggest that Brcal and Brca2 may amplification using primers a-d (Fig. 1A, C). At embry- function in overlapping regulatory pathways involved in onic day 6.5 (E6.5) -25% of all embryos were morpho- the control of cell proliferation and differentiation (Rajan et logically abnormal; these were genotyped as mutants. al. 1996, and unpubl.; Vaughn et al. 1996). E6.5 Brca2 ~°-~ mutant embryos were half the size of To determine the role of Brca2 in embryonic develop- their wild-type E6.5 littermates and showed a poorly de- ment and adult cellular physiology, we have generated a fined boundary between the embryonic and the extraem- Brca2-deficient mouse by targeted deletion of exons 10- bryonic regions (Fig. 2A). At E7.5, the difference in size 11 of the Brca2 gene. Heterozygous Brca21°-~ mutant between wild-type and mutant embryos was even more mice are viable and normal, but homozygous Brca2 ~°-~ dramatic (Fig. 2B), and some mutant embryos had com- mutant embryos die during early postimplantation de- menced resorption. At E8.5, most mutant embryos were velopment. Analysis of Brca2 ~°-~ mutant embryos in in resorption (Table 1), although some did develop a head vivo and in vitro indicates that Brca2 is essential for the fold, a primordial neural tube, allantois, and expanded control of the proliferative process that occurs in early yolk sac but no somites (Fig. 2C). All mutant embryos embryonic development. The similarities and differences were dead or in resorption by E9.5. in phenotype between the Brcal and Brca2 mutants, and Mutant mice generated from the two independent tar- their possible functional relationship, are discussed. geted ES cell clones showed identical phenotypes. In addition, the analyses were performed in both an inbred (C57BL/6J) and an outbred (CD1)background that gave Results comparable phenotypes. These results demonstrate that homozygosity for the Brca2 ~°-~ ~ allele results in embry- Generation of Brca2 ~°-1 ~ mutant mice onic lethality before E9.5 and indicate that Brca2 is es- The Brca2 gene was disrupted in embryonic stem (ES) sential for postimplantation development. cells using a targeting vector that deleted the 3' half of exon 10 and the 5' half of exon 11 of Brca2, leading to the Histological analysis of the Brca21°-11 mutant introduction of two termination codons in-frame. A cas- embryos sette containing the neomycin resistance (neo) gene with the phosphoglycerokinase (PGK) promoter and a poly(A) The structural organization of Brca21°-11 mutant em- addition signal was inserted in the sense orientation in bryos was characterized in detail by histological analysis the targeting vector, so that the short arm consisted of of serially sectioned E5.5-E7.5 embryos obtained from 516 bp of exon 10 and the long arm contained 5.0 kb of heterozygous crosses (Fig. 3A-H). Differences in size be- exon 11 (Fig. 1A). The targeting construct was electro- tween wild-type and mutant embryos were apparent at porated into ES cells. PCR and Southern blot analyses the E5.5 egg cylinder stage (Fig. 3A, B). At E6.5, pheno- showed that 3 of 1960 G418-resistant colonies were het- typic differences between wild-type and mutant embryos erozygous for the Brca2 locus, with no random integra- were more obvious (Fig. 3C-E). Wild-type embryos ex- tions. These three ES clones were used to generate chi- hibited a well-organized ectoderm and endoderm, and a meric mice, two of which successfully contributed to nascent mesoderm (Fig. 3C). Brca21°-~1 embryos could germ line. Chimeras were backcrossed to C57BL/6J or be grouped into two phenotypically mutant classes. The CD1 mice, and heterozygotes were intercrossed to pro- less severely affected mutant embryos had ectoderm and

GENES & DEVELOPMENT 1243 Downloaded from genesdev.cshlp.org on October 4, 2021 - Published by Cold Spring Harbor Laboratory Press

Suzuki et al.

H exon 10 exon 11

type locus

sting vector C

TAATGA Bluescript KS+ H Targeted locus +/- +/+ -/- +/-

TAATGA a) (b allele Ranking probe Wild type allele 4.7kb type allele Mutant allele 2.7kb B +/- +1+ +/- +/- +/+ +/+ +/- +/- +/- Figure 1. Targeted disruption of the Brca2 locus. (A)(Top)A portion of the mouse Brca2 wild-type locus showing exons 10-11 and HindIII sites (H). Nild type allele The 4.7-kb HindIII fragment present in the wild-type allele and the PCR primers c and d are shown. (Middle) Targeting vector and the neo gene positioned in the sense orientation to Brca2 gene transcription. The targeting vector was designed such that the neo cassette replaced exons 10 and 1 t of the Brca2 gene. Two stop codons were inserted at the 3' end of the short arm in-frame. (Bottom) The mutated Brca2 locus showing the 2.7-kb Hin- dIII DNA fragment present in the recombinant Brca2 allele. The position of Mutant allele the 5'-flanking probe used for Southern blot analysis and the PCR primers a and b used to identify the mutated allele are shown. (B) Southern blot analysis of representative genomic tail DNA from one litter of Brca21°-11 heterozygous intercrosses. DNA was digested with HindlII and hybridized with the 5' flanking probe. The 4.7-kb band representative of the wild-type allele and the 2.7-kb band representative of the mutated allele are indicated. (C) Representative genotypic analysis of E6.5 embryos from a Brca2 heterozygote breeding. DNA samples were subjected to PCR using the primer pairs a/b or c/d (see Materials and Methods for sequences). PCR amplification of the mutated Brca2 allele by primer pair c/d produces a 602-bp DNA fragment (top), and PCR amplification of the wild-type Brca2 allele by primer pair a/b produces a 433-bp DNA fragment (bottom). endoderm, and formed a primitive streak (Fig. 3D). Em- 9-11 revealed the expression of a Brca2 transcription of bryos in the more severe class showed only epiblast, ll-kb in ES cells, and in E7.5, Ell.5, E15.5, and E17.5 no detectable primitive streak, poorly defined visceral wild-type embryos (Fig. 4) The spatial expression of and parietal endoderm, and a proamniotic cavity (Fig. Brca2 in E6.5 wild-type and mutant embryos was ana- 3E). At E7.5, only some of the Brca2 ~°-~ mutant em- lyzed by in situ hybridization in tissue sections, using an bryos were organized in three embryonic layers (Fig. 3G). antisense probe for Brca2 exons 12-16. Brca2 transcripts Interestingly, cellular debris was observed frequently in could be detected at low levels throughout the epiblast, the proamniotic cavity of E7.5 mutant embryos (Fig. 3 developing mesoderm, visceral endoderm, and extraem- G,H), reminiscent of the debris observed in earlier wild- bryonic ectoderm and endoderm (Fig. 5 A,B). Mutant em- type embryos that is generated during the process of bryos failed to show any Brca2 transcripts (Fig. 5 C,D), cavitation (Coucovanis and Martin 1995). In addition, indicating that the Brca2 m-~ is a null mutation. the amnion was very thin or absent in mutant embryos (Fig. 3G, H). Because of the increasing severity of the Brca2 m-~ mutant phenotype at later stages of develop- Trophoblast and mesoderm development in Brca21°-11 ment, we focused our phenotypic analyses on days E6.5- mutant embryos E7.5 of embryogenesis. At E6.5, Brca21°-11 mutant embryos were poorly developed, showing a drastically reduced embryonic region, Brca2 expression during early embryogenesis although the extraembryonic region appeared normal Northern blot analysis using a probe for Brca2 exons (Fig. 6E). The latter was confirmed by analyzing the ex-

1244 GENES & DEVELOPMENT Downloaded from genesdev.cshlp.org on October 4, 2021 - Published by Cold Spring Harbor Laboratory Press

Brca2 in murine embryonic cellular proliferation

Table 1. Genotypic and phenotypic analyses of neonates and embryos from Brca21°-~1 heterozygous intercrosses A E6;5 Genotype Stage Brca2 ÷/* Brca2 */- Brca2 -/- Total E6.5 18 (0) 36 (0) 11 (11) 65 (11) E7.5 5 (0) 15 (0) 8 (8) 28 (8) E8.5 12 (0) 26 (0) 21 (21)~ 59 (21) J" E9.5 1 (0) 5 (0) 2 (2) 8 (2) 4- Neonate 16 (0) 35 (0) 0 51 (0) Neonates were genotyped by Southern blot analysis using the 5'-flanking probe shown in Fig. lB. The embryos were collected on days 6.5, 7.5, 8.5, and 9.5 of pregnancy from Brca2 ~°-~ heterozygous intercrosses. The embryo genotypes were determined by PCR amplification using DNA extracted from the whole embryo or from the yolk sac, using primers a, b, c, and d. Figures in parentheses indicate the number of phenotypically abnormal embryos. In comparison to their normal-looking control littermates, all homozygous mutant embryos exhibited an abnormal phenotype, characterized by poor embryonic organization and a reduced size. aIncludes embryos in resorption (30%-40%).

Figure 2. Morphology of Brca21°-I1 mutant embryos. (A) At E6.5, Brca2 ~°-z~ mutant embryos (right) are smaller than their pression of the early trophoblast lineage marker Mash-2, wild-type littermates (left), although relatively well organized. a gene expressed in the wild-type embryo throughout The arrows point to the separation between the embryonic and preimplantation development. In wild-type E6.5 em- extraembryonic regions. (B) At E7.5 the mutant embryos have bryos, Mash-2 is expressed in the diploid trophoblast lin- grown very poorly, and the difference in size with their wild- eage (Fig. 5E, F) but not in the trophoblast giant cells. In type littermates has become more apparent. (C) At late E8.5, Brca21°-~ mutant embryos Mash-2 expression was nor- wild-type embryos (left) are well advanced in organogenesis. mal (Fig. 5G,H). Some Brca2 ~°-~ mutant embryos organize an anteroposterior To determine at the molecular level whether meso- axis and some develop a headfold (arrowhead). A poorly devel- derm was formed in mutant embryos, we examined the oped allantois can also be observed in some of them (arrow). Bar, expression of the Brachyury (T) protein by immunohis- 70 pm in A and B and 140 ~m in C. tochemistry in tissue sections. In wild-type E6.5 embryos, the Brachyury protein was expressed in the primitive streak in six of six embryos analyzed (Fig. 5I). How- amined. Blastocysts from heterozygous matings (E3.5) ever, Brca2 ~°-11 embryos at E6.5 could be grouped into were collected and cultured individually in vitro. Mu- two classes: The less severely affected embryos, al- tant blastocysts showed a normal phenotype, indicating though disorganized, did express Brachyury (Fig. 5J). The that the Brca21°-11 mutation does not affect preimplan- more severely affected mutants did not show any tation development (Fig. 6A, B). However, after 7 days in Brachyury expression (three of four embryos analyzed; culture, the ICM did not grow in 90% of the mutant Fig. 5K). These phenotypic differences in Brachyury ex- blastocysts, compared to failure in only -10% of wild- pression were entirely consistent with the histological type and heterozygous blastocysts (Fig. 6C,D; Table 2). analysis and indicated that, unlike Brcal s-6 mutant em- Only trophoblast giant cells developed in cultured mu- bryos, some Brca2 ~°-~ embryos were able to make me- tant embryos (Fig. 6D). In addition, numerous attempts soderm. to generate Brca2 t°-11 homozygous mutant ES cells by increasing the G418 concentration, or by using a second targeting vector with a hygromycin resistance cassette, Brca21°-11 mutation decreases embryonic growth in were unsuccessful (data not shown). vitro and in vivo To directly study the effects of the Brca21°-~ muta- Hakem et al. {1996) and Liu et al. (1996) showed that the tion on cellular proliferation, we examined the incorpo- growth deficit of Brcal mutant embryos was correlated ration of 5-bromo-2'-deoxyuridine (BrdU) into DNA dur- with decreased cell proliferation. The reduced size at ing the S phase of the cell cycle. Analysis was performed E6.5 of Brca2 ~°-t ~ mutant embryos was likely not attrib- at E6.5 and E7.5, the times at which the highest mitotic utable to an excess of apoptosis, as judged by Hoechst activity has been observed in the mouse embryo (Snow dye staining of wild-type and mutant embryos (data not 1977). Whole litters were serially sectioned and embryo shown). To determine whether the proliferative capabil- phenotypes were classified as wild-type or mutant. A ity of Brca21°-~ mutant embryos was intrinsically im- proliferative index based on the ratio of proliferating paired, the growth of the inner cell mass (ICM 1 was ex- cells (BrdU-positive nuclei} to total cell number was cal-

GENES & DEVELOPMENT 1245 Downloaded from genesdev.cshlp.org on October 4, 2021 - Published by Cold Spring Harbor Laboratory Press

Suzuki et al.

E5.5 E6.5 E7.5 erative index of 0.9 _+ 0.03%. Three phenotipically mutant embryos analyzed gave an index of 0.6 + 0.1%. These values were statistically significantly different (P < 0.05, Student's t-test). These results show that the proliferation ability of Brca2 ~°-~1 mutant embryos is impaired at E7.5, at the time when morphological abnormalities become more prominent. The inability of Brca2 ~°-~ mutant blastocysts to grow in vitro and the reduced cellular proliferation of Brca2 ~°-~ mutant embryos in vivo, suggest that Brca2, like Brcal, is required for embryonic cellular proliferation. Hakem et al. (1996) C' 'ig ', j) i showed that the reduced BrdU incorporation observed in i G Brca 1 s 6 mutants was accompanied by decreased expres- , --2' sion of cyclin E. In contrast, no apparent differences in • ::, > i the expression of either cyclin E (Fig. 7) or cyclin A (data not shown) were detected between wild-type and Brca2 l°-~ ~ mutant embryos.

p21 expression is increased in Brca21°-11 mutant embryos

The reduced BrdU incorporation observed in Brca2 ~°-1~ mutant embryos was consistent with an increase in the length of the cell cycle. The interplay between on- cogenic proteins, such as mdm-2, and tumor suppressors, such as p53, is essential for cell cycle progression in eukaryotic cells, mdm-2 expression was down-regulated in Brcal 5-~' mutant embryos, concomitant with increased expression of the cyclin-dependent kinase (cdk) inhibitor p21 (Hakem et al. 1996). The phenotypic similarity be- Figure 3. Histological analysis of Brca2 t°<~ mutant em- tween Brca2 I°-11 and Brca 15-6 mutant embryos led us to bryos. Sagittal sections are shown. (A,B) E5.5; (C-E) E6.5; (F-H) determine, using RT-PCR and in situ hybridization, E7.5. (A) E5.5 wild-type embryo (early egg cylinder), (B) E5.5 whether the expression of mdm-2, p53, and p21 was af- Brca21°-11 mutant embryo. The embryonic region (arrow) is very reduced. (C) E6.5 wild-type embryo (egg-cylinder stage). fected in Brca2 ~°-1~ mutant embryos. For RT-PCR Both the embryonic and extraembryonic regions are well orga- analysis, whole E7.5-dissected embryos were used. Inter- nized. (D) E6.5 Brca21°-11 mutant embryo. The large arrow estingly, while the expression levels of mdm-2 and p53 points to the separation between the embryonic and extraem- were unaffected (Fig. 7A; data not shown), p21 expres- bryonic region. The embryonic region is poorly developed, al- sion was increased two- to fourfold in four E7.5 Brca2 ~°- though the proamniotic cavity is apparent. (E) E6.5 severe ~1 mutant embryos analyzed (Fig. 7A). This result was Brca2 ~°-~ mutant embryo. The embryonic region is very re- supported by in situ hybridization analysis performed in duced. (F) E7.5 wild-type embryo. The three germ layers are three E6.5 wild-type and three Brca2 ~°-~ mutant em- apparent. (G) E7.5 Brca2 ~°-~ mutant embryo. The embryonic bryos (Fig. 7B-E). These results are similar to our previ- region is reduced. Compare the difference in size with the wild- type embryo shown in F. Cellular debris can be observed in the proamniotic cavity (arrow). Mesoderm (m) has been generated. The amnion is absent. (H) Severe Brca2 ~°-~ mutant embyro. ES E7 Ell E15 E17 The embryonic region (arrow) is drastically reduced. (al) Allan- tois; (am)amnion; (ch) chorion; (ee)embryonic ectoderm; (eee) extraembryonic ectoderm; (epc) ectoplacental cone; (m) meso- Brca2 derm; (pa)proamniotic cavity; (ve) visceral endoderm. Bar, 60 gm in A-E; 150 l~m in F-H.

fl-actin culated. Both the extraembryonic and embryonic regions Figure 4. Brca2 is expressed throughout embryogenesis. of the wild-type and mutant Brca2 ~°-~ embryos were Northern blot analysis of poly(A) RNA prepared from ES cells analyzed. No significant differences in mitotic index and from wild-type embryos at 7.5, 11.5, 15.5, and 17.5 days of were observed at E6.5 (data not shown). However, at gestation. Hybridization with 3zP-labeled exon 11 Brca2 probe, E7.5, differences in proliferation rate between wild-type and detected expression of the Brca2 gene in ES cells and in all and mutant embryos were apparent. A total of four wild- analyzed stages of embryonic development. A probe to [3-actin type embryos were analyzed and they showed a prolif- was used as a control for sample loading.

1246 GENES & DEVELOPMENT Downloaded from genesdev.cshlp.org on October 4, 2021 - Published by Cold Spring Harbor Laboratory Press

Brca2 in routine embryonic cellular proliferation

Brca2 Mash2 Brachyury

Figure 5. Spatial expression of Brca2, Mash-2, and Brachyury in E6.5 Brca21°-~1 mutant embryos. (A,C,E,G) In situ hybridization, bright-field view; (B,D,F,H.) dark-field view. (A-D)Brca2 expression; (E-H)Mash-2 expression; (I,J)Brachyury expression. Sagittal sections are shown (A,C,E,G). The arrow points to the separation between the embryonic and extraembryonic regions. (A,B) Brcal expression in an E6.5 wild-type embryo. Sections were hybridized with an antisense Brca2 exon 12-16 probe. Brca2 is expressed ubiquitously in the embryo. (C,D) Brca21°-11 mutant embryo. No Brca2 expression is detected. (E,F) Mash-2 expression in an E6.5 wild-type embryo. Strong expression is detected in the diploid trophoblast of the extraembryonic region of the embryo. (G,H) Mash-2 expression in a Brca2 I°-~ mutant embryo. Expression is normal. (LJ,H) Immunohistochemal analysis of Brachyury expression. (I) Wild-type embryo. Branchyury-positive cells can be observed in the nascent streak. (J) Brca2 ~°<~ mutant embryo. Weak Brachyury expression is detected. The arrows in I and J point to the Brachyury-positive cells. (K) Severe Brca2 ~°-~~ mutant embryo. No Brachyury expression is observed. The arrow points to the embryonic region. Bar, 60 ~m.

ous observations with Brca 15-6 mutant embryos (Hakem Discussion et al. 1996). The Brca2 l°-j 1 mutant phenotype We have generated a null mutation of the Brca2 gene by The Brca21°-11 mutation does not affect Brcal and homologous recombination in mouse ES cells. The Rad51 expression Brca2 ~°-~I mutation affects embryonic growth and caused lethality before day 9.5 of gestation. Mutant em- Brca21°-~ and Brcal 5-6 mutant embryos share many bryos are abnormal as early as E5.5, although some em- phenotypic features, suggesting that these genes may be bryos gastrulate and organize an anterior-posterior pat- functionally related. As an initial step toward determin- tern before dying at around E8.5. We and others have ing whether Brca 1 and Brca2 are indeed related, we used shown that mutation of the Brcal gene causes embry- RT-PCR and in situ hybridization to examine the ex- onic lethality in mice (Gowen et al. 1996; Hakem et al. pression of Brcal in Brca2 ~°-~ mutant embryos and that 1996; Liu et al. 1996). In particular, Brcal s-6 mutant em- of Brca2 in Brcal 5-~ mutant embryos. No differences in bryos show a very severe growth retardation and die be- Brcal expression were detected in Brca2 ~°-1~ mutants fore E7.5 (Hakem et al. 1996). The similarities in pheno- (Fig. 7; data not shown). Similarly, no effect on Brca2 type between Brca2 ~°-~1 and Brcal 5-~ mutant embryos expression was observed in Brcal s-6 mutant embryos suggest that these genes may be involved in the same or (data not shown). The recently reported association be- related processes. tween hRad51, a molecule involved in DNA recombination repair, and BRCA1 (Scully et al. 1997) prompted us to examine whether the expression of mRad51 was af- Brca2 and cellular proliferation fected in Brca21°-~ mutants. No alteration of Rad51 expression in Brca2 ~°-~ ~ mutants was detected by RT-PCR At E7.5, BrdU incorporation was markedly reduced in (Fig. 7). Brca21°-11 mutant embryos, indicating a slow cell cycle

GENES & DEVELOPMENT 1247 Downloaded from genesdev.cshlp.org on October 4, 2021 - Published by Cold Spring Harbor Laboratory Press

Suzuki et al.

WT -/- peared normal, although their proliferative abilities in vitro were reduced. In Brca21°-11 mutant blastocysts, growth retardation was apparent after 3 days of culture, but a fully penetrant growth defect was observed only after 7 days. In the case of Brcal s-6 mutant blastocysts, poor growth or absence of ICM was apparent after 3 days m D of culture (Hakem et al. 1996). In Brca2 ~°-1~ mutant blastocysts, as in Brcal 5-6 blastocysts (Hakem et al. 1996), only giant cells survived after a period of culture, suggesting that this lineage is more resistant to the absence of either the Brcal or Brca2 protein. Interestingly, neither Brcal s-6, Brca21°-11, nor Rad51 mutant embryos (Lim and Hasty 1996) grow in vitro, and only trophoblast I giant cells, in which the DNA endoreplicates, are observed to undergo transient proliferation. It is still for- mally possible that the observed differences in proliferation between wild-type and Brca2 ~°-11 mutant embryos could arise as a secondary effect of the mutation. Gastrulation in the mouse embryo occurs at -6.5 days of gestation, at which time the mesoderm is generated from the epiblast. This process is associated with rapid epiblast cell proliferation, resulting in a 100-fold increase in cell number between E5.5 and E7.5 (Snow 1977; Power and Tam 1993). Three-quarters of the Brca2 l°-1 ~ mutant embryos analyzed did not develop mesoderm at E6.5, as revealed by the absence of Brachyury expression. Con- sidering that the cardinal feature of Brca21°-1~ mutant embryos is their reduced size, the most likely interpretation of this observation is that defective mesoderm development in Brca21°-1 ~ mutant embryos arises as a con- sequence of a primary proliferative defect.

Figure 6. Proliferation xs impaired in vitro and in vivo in Brcal, Brca2, and genetic integrity in the early embryo Brca2 J°-tl mutant embryos. (A) Wild-type E3.5 blastocyst; (B) We have shown that the Brca2 ~°-~1 mutation causes an mutant E3.5 blastocyst; (C) wild-type outgrowth after 7 days of increase in the expression of p21. p21 is a universal cdk culture. The ICM is surrounded by trophoblast giant cells (TG). (D) Mutant outgrowth after 7 days of culture. Very little ICM is inhibitor that impedes the cell cycle at the G 1 transition left. Mostly trophoblast giant cells remain. (E) E7.5 wild-type (Gu et al. 1993; Harper et al. 1993). The transcription of embryo. Strongly BrdU-positive nuclei can be seen throughout p21 is directly regulated by p53 and other factors (E1- the embryo. (F) E7.5 mutant Brca2 ~°-11 embryo. Fewer positive Deiry et al. 1993). As the expression of p53 and mdm-2, cells are detected. (al)Allantois; (am) amnion; (e) ectoderm; (epc) whose protein product regulates p53 activity, appeared ectoplacental cone; (m) mesoderm. Bar, 20/am in A and C; 40 to be unaffected in Brca21°-~ embryos, this would sug- ~m in B and D; 250 ~m in E and F. gest that other regulatory signals downstream of Brca2

Table 2. Genotypic and phenotypic analysis of Brca21°ql mutant blastocysts cultured in vitro progression that ultimately culminated in cell cycle ar- Phenotype rest. Unlike Brcal s-6 mutant embryos, apparently normal cyclin E expression was observed in Brca2 ~°-~ mu- Genotype Total no. normal abnormal tants, suggesting that the cell cycle block in Brca2 ~°-~ Brca2 +/+ 11 10 (90%) 1 (10%) may occur in late G 1 phase, when cyclin E levels are Brca2 ÷/- 22 20 (91%) 2 (9%) already high and thus difficult to distinguish from wild- Brca2 -/- 8 1 (12%) 7 (88%) type levels. In any event, the effect on cell growth of the Brca2 ~°-11 mutation is milder than that of the Brcal s-6 Blastocysts from Brca2 I°-~1 heterozygous intercrosses were col- mutation. This is consistent with the observation that lected at day 3.5 of gestation and individually cultured in ES media without LIF. Embryos grown in vitro for 7 days were the developmental arrest and death of Brca21°-~ 1 mutant photographed daily and genotyped by PCR amplification using embryos occurs almost 2 days later than in Brca 15-6 mu- primers a, b, c, and d as shown in Fig. 1, A and C. Embryo tant embryos. The temporal difference in the onset of phenotype was scored as abnormal if the ICM did not develop detrimental effects caused by these mutations is also re- after 7 days of culture. Numbers in parentheses indicate per- flected in the development of outgrowths of cultured centages of cultured blastocysts with normal and abnormal phe- blastocysts. E3.5 Brca2 ~°-~ and Brcal 5-~ blastocysts ap- notypes.

1248 GENES & DEVELOPMENT Downloaded from genesdev.cshlp.org on October 4, 2021 - Published by Cold Spring Harbor Laboratory Press

Brca2 in murine embryonic cellular proliferation

Brca2 might participate in the same process or be present +l+ -I- in the same complex as Rad51 and Brcal. In this sce- }~iiiL nario, the lethality of Brca21°-11 mutant embryos would p21 be attributable to cell cycle arrest caused by genome er- rors occurring in early postimplantation development, the time at which the rates of mitotic division are the rndm2 highest and the stresses exerted on the maintenance of genetic integrity are greatest. Subsequent activation of checkpoint genome guardian functions would result in p53 cell cycle arrest. It has been shown that p53 is the key 9 cell cycle regulator responding to DNA damage (Cox and cyclinE Lane 1995; Levine 1997). Although p53 expression was unaffected in Brca2 l°-Z ~ mutant embryos, the expression of its downstream transcriptional target p21 was in- Rad51 creased. Because p21 is the most pleiotropic mediator of p53-mediated cell cycle arrest, the enhanced expression of p21 is consistent with the growth arrest observed in Brca I Brca21°-1~ mutant embryos. Interestingly, the survival ~G of Rad51 mutant embryos was increased in a p53 mutant background (Lim and Hasty 1996). ~-actin The possible functional relationship among Brca2, Brcal, and Rad5I may perhaps be dissected by genetic means, that is, by analyzing the phenotypes of mice mu- Figure 7. Brca21°-11 mutant embryos show increased p21 expression. (A) The expression levels of p21, mdrn-2, p53, cyclin E, tated for combinations of these genes. In the simplest Rad51, and Brcal were analyzed in Brca2 ~°-~ mutants using case, if these three proteins belong to the same complex RT-PCR on RNA from individual E7.5 embryos (see Materials or are involved in the same process, a synergistic inter- and Methods). The blots were hybridized to specific cDNA action would be expected in any pairwise combination of probes. While levels of mdm-2, p53, cyclin E, RadS1, and Brcal mutations, resulting in a more severe phenotype in the expression were not affected, the expression of p21 was in- double mutant. However, the extreme severity of the creased two to four times in the E7.5 Brca21°-11 mutant (top). Brcal 5-6 and Rad51 mutant phenotypes would make the ~-Actin was used as a control for RNA sample quantity. (B,C). interpretation of such double mutants very difficult. Bio- In situ hybridization analysis of p21 expression in E6.5 wild- chemical approaches or tissue-specific knockout mu- type embryos, p21 is expressed at low levels throughout the tants will likely prove more fruitful for analyzing the embryo. (D,E) p21 expression in E6.5 Brca2 ~°-~ mutant embryos. Signal is up-regulated. The arrows in B and D point to the function of Brcal and Brca2, and their relation to RadS1. separation between the embryonic and extraembryonic regions. Bar, 60 ~m. Materials and methods Generation of Brca2 mutant ES cells and mutant mice might be resposnsible for the up-regulation of p21 ob- The targeting vector was designed to replace a 5-kb genomic served in Brca2 ~°-~ embryos. fragment containing part of Brca2 exons 10 and 11 with the DNA damage has been shown to inhibit progression PGK-neo resistance expression cassette containig a poly(A) ad- through the cell cycle, thereby demonstrating a relation- dition site in sense orientation to Brca2 transcription. The short ship between DNA lesions and the cell cycle. Tumors arm was engineered by PCR amplification to include two stop can arise form defects in DNA repair (Kolodner 1996). codons in-frame at the 3' end. The construct was linearized by digestion with SacII, and 50 ~ag of DNA was electroporated into Recently, Scully et al. (1997) reported that hRad51 is 1 x 107 E14K ES cells (Bio-Rad Gene Pulser, 0.34 V, 0.25 mF). associated with BRCA1 in mitotic S phase and meiotic E14K ES cells from 129/ola mice were maintained on a layer of zygotene and pachytene, brad51 is a member of a family mitomycin C-treated embryonic fibroblasts in Dulbecco's proteins known to mediate DNA exchange functions modified Eagle's medium (DMEM) supplemented with leuke- leading to normal recombination (Kowalczykowski mia inhibitory factor (LIF), 15% fetal calf serum (FCS), L-gluta- 1991; Radding 1991; Sung 1994; Sung and Robberson mine, and ~-mercaptoethanol. ES cell colonies resitant to G418 1995; Baumann et al. 1996). It has been suggested that (250 ~g/ml; Sigma) were screened for homologous recombina- BRCA1 and other tumor suppressors such as ATM tion by PCR using primers specific for sequences in the Brca2 (Keegan et al. 1996) play a role in the maintenance of and neo resistance genes (sense primer a, 5'-CTGCCTTA- genome integrity. An interaction between Rad family GAAAGCCAATACC-3'; antisense primer b, 5'-GCGCCTC- CCCTACCCGGTAG-3'; Fig. 1A). Recombinant colonies were members and BRCA1 and/or BRCA2 may be required to confirmed by Southern blotting using a 5'-flanking probe and ensure cell cycle progression, high fidelity DNA replica- neo-specific probe. tion, and the maintenance of genome integrity. Chimeric mice were produced by microinjection of targeted The phenotypic similarities among Brca21°-i 1, ES cells into 3.5-day C57BL/6J blastocysts. Chimeras were bred Brcal s-6, and Rad51 mutant embryos would suggest that to C57BL/6J mice (Jackson Laboratories), and germ-line trans-

GENES & DEVELOPMENT 1249 Downloaded from genesdev.cshlp.org on October 4, 2021 - Published by Cold Spring Harbor Laboratory Press

Suzuki et al. mission of the mutant allele was confirmed by genomic South- In vitro culture of preimplantation embryos ern blot analysis of tail DNA from F1 offspring with agouti coat Brca2 I°-II heterozygous males and females were intercrossed color. Two injected ES cells showed successful germ-line trans- for 2 hr, and E3.5 embryos were collected by flushing them from mission. F2 offspring from heterozygous intercrosses were geno- the uterus of the plugged females. Blastocysts were individually typed by PCR or Southern blot analysis. The heterozyous males cultured in 24-well plates in ES cell media without LIF, in 5 % were also crossed with CD1 mice. Mutants derived from both CO2 at 37°C. Photographs of the cultured embryos were taken mouse backgrounds showed the same phenotype. every 24 hr. After 7 days in culture, the morphology of the embryos was noted and their genotype was determined by PCR. PCR analysis of Brca21°-1~ genotypes Genomic DNA from ES cells and neonate tail was isolated and used in PCR amplification. Yolk sacs, E6.5-8.5 embryos, or blas- Histological analysis tocysts cultured in vitro were incubated overnight at 37°C in Uteri from females plugged in a 2-hr mating period were iso- 100 gl of lysis buffer [50 mM KCL, 10 mM Tris (pH 8.3), 2 mM lated in ice-cold PBS at E5.5-E8.5, fixed overnight in 4% para- MgCI2, 0.1 mg/ml of gelatin, 0.45% NP-40, 0.45% Tween-20] formaldehyde at 4°C, dehydrated, and embedded in paraffin. containing 100 ~g/ml of proteinase K. After 10 rain of boiling, Sections 6 gm thick were cut and stained with hematoxylin and 1-5 Ill of the samples was subjected to PCR amplification. Prim- eosin. ers c (5'-CCAGGTGTAACAAGCCAGAAG-3') and d (5'- GTCTCGTCAAGTGGCTCTCTC-3'), specific for the deleted portion of the Brca2 gene, were used to detect the wild-type allele, and primers a and b (see above) were used to detect the In situ hybridization recombinant allele. Temperature cycling conditions were one Uteri were isolated in ice-cold PBS at E6.5 and processed as for inital cycle at 94°C for 7 min, 62°C for 5 min, 72°C for 1 min, histological analysis. The probes used were Brca2 (exons 12-16), followed by 40-50 cycles at 94°C for 1 min, 62°C for 1 min, and Mash-2 (Guillemot et al. 1994), and full-length p53, mdm-2, and 72°C for 1 rain and 30 sec (DNA thermal cycler, Perkin Elmer p2t cDNAs. Probes were labeled with [32p]UTP and processed Cetus). Half of each reaction mixture was electrophoresed on a according to protocols described previously (Hui and Joyner 1.7% agarose gel and stained with ethidium bromide. Primer 1993). pair c/d amplified a 433-bp fragment in both heterozygote and wild-type DNA samples, whereas the primer pair a/b amplified a 602-bp fragment in both heterozygous and homozygous mu- Immunohistochemistry tant DNA. Uteri were isolated in ice-cold PBS at E6.5, fixed in 4% pard- formaldehyde for 4 hr, dehydrated, embedded in wax, and sec- RNA analysis tioned at 6 ~m. The following polyclonal antisera were used: Total RNA was extracted from ES cells or E7.5, Ell.5, E15.5, anti-Brachyury (1:500 dilution)(recognizes amnio acids 328-420 and E17.5 embryos using Trizol (Life Technologies). For North- of the carboxy-terminal part of the T protein; Kispert and Her- ern blot analysis, 1 gg of poly(A) RNA was electrophoresed in a mann 1993); anti-cyclin A (1:200); and anti-cyclin E (1:200). For 1% agarose/formaldehyde gel and transferred to a nylon mem- immunostaining, the protocol described in Hakem et al. (1996) brane (Hybond N+, Amersham). RNA blots were hybridized was used. overnight at 65°C in Church and Gilbert buffer (Church and Gilbert 1984). The filters were hybridized with a cDNA probe specific for exons 9-11 of Brca2. BrdU labeling of embryos cDNAs were generated using random primers (RT-PCR kit, Stratagene) and one-tenth volume, or the total volume, of RNA BrdU labeling of cells in the S phase of the cell cycle was per- from individual wild-type or mutant E7.5 embryos, respec- formed according to the protocol described by Hayashi et al. tively. PCR amplification was performed using the/3-actin-spe- (1988). BrdU (100 lag/gram of body weight) was injected intra- cific primer sets (5'-CACTGCCGCATCCTCTTCCTC-3'/5'- peritoneally into pregnant females at E6.5 and E7.5. The females GCTGTCGCCTTCACCGTTCCA-3', 64°C annealing). RNA were sacrificed 1 hr after injection, the uteri were removed, and from four wild-type and four Brca2 ~°-~ mutants, which showed the decidual swellings were fixed in 4% paraformaldehyde at /3-actin bands of equal intensity, were chosen for PCR amplifi- 4°C overnight and processed for immunohistochemistry. The cation experiments. Specific primers used in the PCR reactions sections were incubated with an anti-BrdU monoclonal anti- were p21 (5'-GACGACCTGGGAGGGGACAAG-3' / 5'-TAAG- body (Boehringer Mannheiml at a 1:10 dilution. Staining was GTTTGGAGACTGGGAGAG-3', 62°C annealing); mdm-2 (5'- performed according to the protocol described by Mishina et al. CCGAGCCTGGGTCTGTGTGAG-3'/5'-GGAAGTCGATG- (1995). GTTGGGAATA-3', 64°C); p53 (5'-GGATAGGAAAGAGCA- CAGAGC-3' / 5'-CCAGTCTTCGGAGAAGCGTGAC-3', 62°C); cyclin E (5'-AGACTCCCACAACATCCAGAC-3'/5'-ACCTA- Acknowledgments CAACACCCGAGCAGAG-3', 62°C); Rad51 (5'GACAAAAT- TCTGACTGAGGCAGCAAAA-3' / 5'-AGCAAGTCGAAGCA- We thank Merrie Jo Johnson, Joseph Dovala, and Laarni Antonio GCATCCTCAGAAA-3', 64°C); or Brcal (5'-GAACTGATCA- for the Brca2 cDNA fragment and DNA sequencing; Hiroki AAGAACCTGTT-3'/5'-GAAGCCTTCTGACACGGTTCC-3', Yoshida, Goichi Matsumoto, David Siderovski, and Marynette 64°C). PCR amplifications were carried out for 22 cycles (/3- Rihanek for help with cell culture, helpful discussions, and actin) or 27 cycles for the other genes. Amplified PCR products searching of DNA sequences; Bernhard Hermann for the were electrophoresed in a 1.7% agarose gel followed by South- Brachyury antiserum; and Mary Saunders for scientific editing. ern blotting, cDNA probes specific for the above genes were The publication costs of this article were defrayed in part by used for hybridization. payment of page charges. This article must therefore be hereby

1250 GENES & DEVELOPMENT Downloaded from genesdev.cshlp.org on October 4, 2021 - Published by Cold Spring Harbor Laboratory Press

Brca2 in murine embryonic cellular proliferation marked "advertisement" in accordance with 18 USC section tent inhibitor of G1 cyclin-dependent kinases. Cell 75: 805- 1734 solely to indicate this fact. 816. Hansen, M.F., A. Koufos, B.L. Gallie, R.A. Phillips, O. Fodstad, A. Brogger, T. Gedde-Dahl, and W.K. Cavenee. 1985. Osteo- scarcoma and retinoblastoma: A shared chromosomal Proc. Natl. References mechanism revealing recessive predisposition. Acad. Sci. 82: 6216-6220. Baumann, P., F.E. Benson, and S.C. West. 1996. Human Rad5t Hayashi, Y., M. Koike, M. Matsunami, and T. Hoshino. 1988. protein promotes ATP-dependent homologous pairing and Effects of fixation time and enzymatic digestion on immu- strand transfer reactions in vitro. Cell 87: 757-766. nohistochemical demonstration of bromodeoxyuridine in Church, G.M. and W. Gilbert. 1984. Genomic sequencing. Pro& formalin-fixed, paraffin-embedded tissue. J. Histochem. Cy- Natl. Acad. Sci. 81: 1991-1995. tochem. 36: 511-514. Collins, N., R. McManus, R. Wooster, J. Mangion, S. Seal, S.R. Hui, C.C. and A. Joyner. 1993. A mouse model for the Greig Lakhani, W. Ormiston, P.A. Daly, D. Ford, D.F. Easton, and cephalopolysyndactyly syndrome: The extra-toes j mutation M.R. Stratton. 1995. Consistent loss of the wild-type allele contains an intragenic deletion of the Gli3 gene. Nature in breast cancers from a family linked to the BRCA2 gene on Genet. 3: 241-246. chromosome 13q12-13. Oncogene 10: 1673-1675. Hunter, T. 1997. Oncoprotein networks. Cell 88: 333-346. Coucovanis, E. and G.R. Martin. 1995. Signals for death and Keegan, K.S., D.A. Holtzman, A.W. Plug, E.R. Christenson, E.E. survival: A two step mechanism for cavitation in the verte- Brainerd, G. Flaggs, N.J. Bentley, E.M. Taylor, M.S. Meyn, brate embryo. Cell 83: 279-287. S.B. Moss, A.M. Carr, T. Ashley, and M.F. Hoekstra. 1996. Cox, L.S. and D.P. Lane. 1995. Tumor supressors, kinases and The Atr and Atm protein kinases associate with different clamps--How p53 regulates the cell-cycle in response to sites along meiotically pairing chromosomes. Genes & Dev. DNA-damage. BioEssays 17: 501-508. 10: 2423-2437. Donehower, L.A., M. Harvey, B.L. Slagle, M.J. McArthur, C.A. Kispert, A. and B.G. Herrmann. 1993. The Brachyury gene en- Montgomery, Jr, J.S. Butel, and A. Bradley. 1992. Mice defi- codes a novel DNA binding protein. EMBO J. 12:3211-3220. cient for p53 are developmentally normal but susceptible to Kleinberg, D.L. and C.B. Newman. 1986. The pituitary gland in spontaneous tumors. Nature 356: 215-221. primate mammary development: Evidence that prolactin is E1-Deiry, W.S., T. Tokina, V.E. Velculescu, D.B. Levy, R. Par- not essential. Ann. N.Y. Acad. Sci 464: 37-43. sons, J.M. Trent, D. Lin, E. Mercer, K.W. Kinzler, and B. Kolodner, R. 1996. Biochemistry and genetics of eukaryotic mis- Vogelstein. 1993. WAF1, a potential mediator of p53 tumor match repair. Genes & Dev. 10: 1433-1442. suppression. Cell 75:817-825. Kowalczykowski, S.C. 1991. Biochemistry of genetic recombi- Futreal, P.A., Q. Liu, D. Shattuck-Eidens, C. Cochran, K. Harsh- nation: Energetics and mechanism of DNA strand exchange. man, S. Tavtigan, L.M. Bennett, A. Haugen-Strano, J. Annu. Rev. Biophys. Chem. 20: 539-575. Swensen, Y. Miki et al. 1994. BRCA1 mutations in primary Kreidberg, J.A., H. Sariola, J.M. Loring, M. Maeda, J. Pelletier, D. breast and ovarian carcinomas. Science 266: 120-122. Housman, and R. Jaenisch. 1993. WT-1 is required for early Gessler, M., A. Poustka, W. Cavenee, R.L. Neve, S.H. Orkin, kidney development. Cell 74: 679-691. and G.A.P. Bruns. 1990. Homozygous deletion in Wilms tu- Lee, E.Y.H.P., C.Y. Chang, N. Hu, Y.C.J. Wang, C.C. Lai, K. mors of a zinc-finger gene identified by chromosome jump- Herrup, W.H. Lee, and A. Bradley. 1992. Mice deficient for ing. Nature 343: 774-778. Rb are nonviable and show defects in neurogenesis and hae- Gayther, S.A., J. Mangion, P. Russel, S. Seal, R. Barfoot, B.A.J. matopoiesis. Nature 359: 288-294. Ponder, M.R. Stratton, and D. Easton. 1997. Variation of Levine, A. 1977. p53, the cellular gatekeeper for growth and risks of breast and ovarian cancer associated with different division. Cell 88: 323-331. germline mutations of the BRCA2 gene. Nature Genet. Lim, D.S. and P. Hasty. 1996. A mutation in mouse rad51 re- 15: 103-105. sults in an early embryonic lethal that is suppressed by a Gowen, L.C., B.L. Johnson, A.M. Latour, K.K. Sulik, and B.H. mutation in p53. Mol. Cell. Biol. 16: 7133-7143. Koller. 1996. Brcal deficiency results in early embryonic Liu, C.Y., A. Fleskennikin, S. Li, Y.Y. Zeng, and W.H. Lee. 1996. lethality characterized by neuroepithelial abnormalities. Inactivation of the mouse brcal gene leads to failure in the Nature Genet. 12: 191-194. morphogenesis of the egg cylinder in early postimplantation Gu, Y., C.W. Turek, and D.O. Morgan. 1993. Inhibition of development. Genes & Dev. 10: 1835-1843. CDK2 activity in vivo by an associated 20k regulatory sub- Markin, D., F.P. Li, L.C. Strong, J.F. Fraumeni, Jr., C.E. Nelson, unit. Nature 366: 707-710. D.H. Kim, J. Kassel, M.A. Gryka, F.Z. Bischoff, M.A. Tain- Gudmundsson, J., G. Johannesdottir, J.T. Bergthorsson, A. Ara- sky, and S.H. Freind. 1990. Germ line p53 mutations in a son, S. Ingvarsson, V. Egilsson, and R.B. Barkardottir. 1995. familial syndrome of breast cancer, sarcomas, and other neo- Different tumor types from BRCA2 carriers show wild-type plasm. Science 250: 1233-1238. chromosome deletions on 13q12q13. Cancer Res. 55: 4830- Miki, Y., J. Swensen, D. Shattuckeidens, P.A. Futreal, K. Harsh- 4832. man, S. Tavtigian, Q.Y. Liu, C. Cochran, L.M. Bennett, W. Guillemot, F., A. Nagy, A. Auerbach, J. Rossan, and A.L. Joyner. Ding, R. Bell, J. Rosenthal, C. Hussey, T. Tran, M. Mcclure, 1994. Essential role of Mash-2 in extraembryonic develop- C. Frye, T. Hattier, R. Phelps, A. Haugenstrano, H. Katcher, ment. Nature 371: 333-336. K. Yakumo, Z. Gholami, D. Shaffer, S. Stone, S. Bayer, C. Hakem, R., J.L. de la Pompa, C. Sirard, R. Mo, M. Woo, A. Wray, R. Bogden, P. Dayananth, J. Ward, P. Tonin, S. Narod, Hakem, A. Wakeham, J. Potter, A. Reitmair, F. Billia, E. P.K. Bristow, F.H. Norris, L. Helvering, P. Morrison, P. Ros- Firpo, C.C. Hui, J. Roberts, J. Rossant, and T.W. Mak. 1996. teck, M. Lai, J.C. Barrett, C. Lewis, S. Neuhausen, L. Can- The tumor-suppressor gene Brcal is required for embryonic nonalbright, D. Goldgar, R. Wiseman, A. Kamb, and M.H. cellular proliferation in the mouse. Cell 85: 1009-1023. Skolnick. 1994. A strong candidate for the breast and ovari- Harper, J.W., G.R. Adami, N. Wei, K. Keyomarsi, and S.J. an-cancer susceptibility gene BRCA1. Science 266: 66-71. Elledge. 1993. The p21-cdk interacting protein Cipl is a po- Miki, Y., T. Katagiri, F. Kasumi, T. Yoshimoto, and Y. Naka-

GENES & DEVELOPMENT 1251 Downloaded from genesdev.cshlp.org on October 4, 2021 - Published by Cold Spring Harbor Laboratory Press

Suzuki et al.

mura. 1996. Mutation analysis in the BRCA2 gene in pri- Wooster, R., G. Bignell, J. Lancaster, S. Swift, S. Seal, J. Man- mary breast cancers. Nature Genet. 13: 245-247. gion, N. Collins, S. Gregory, C. Gumbs, G. Micklem, R. Mishina, Y., A. Suzuki, N. Veno, and R.R. Behringer. 1995. Barfoot, R. Hamoudi, S. Patel, C. Rice, P. Biggs, Y. Hashim, BMPR encodes a type I bone morphogenetic protein-receptor A. Smith, F. Connor, A. Arason, J. Gudmundsson, D. Fice- that is essential for gastrulation during mouse embryogen- nec, D. Kelsell, D. Ford, P. Tonin, D.T. Bishop, N.K. Spurr, B. esis. Genes & Dev. 9" 3027-3037. Ponder, R. Eeles, J. Peto, P. Devilee, C. Cornelisse, H. Lynch, Phelan, C.M., J.M. Lancaster, P. Tonin, C. Gumbs, C. Cochran, S. Narod, G. Lenoir, V. Egilsson, R.B. Barkadottir, D.F. Eas- R. Carter, P. Ghadirian, C. Perret, R. Moslehi, F. Dion, M.C. ton, D.R. Bentley, P.A. Futreal, A. Ashworth, and M.R. Strat- Faucher, K. Dole, S. Karimi, W. Foulkes, H. Lounis, E. ton. 1995. Identification of the breast-cancer susceptibility Warner, P. Goss, D. Anderson, C. Larsson, S.A. Narod, and gene BRCA2. Nature 378: 789-792. P.A. Futreal. 1996. Mutation analysis of the BRCA2 gene in 49 site-specific breast-cancer families. Nature Genet. 13: 120-122. Power, M.A. and P.P.L. Tam. 1993. Oncet of gastrulation, morphogenesis and somitogenesis in mouse embryos displaying compensatory growth. Anat. Embryol. 187: 493-504. Radding, C.M. 1991. Helical interactions in homologous pairing and strand exchange driven by RecA protein. J. Biol. Chem. 266: 5355-5358. Rajan, J.V., M. Wang, S.T. Marquis, and L.A. Chodosh. 1996. Brca2 is coordinately regulated with brcaI during proliferation and differentiation in mammary epithelial-cells. Proc. Natl. Acad. Sci. 93" 13078-13083. Scully, R., J. Chen, A. Plug, Y. Xiao, D. Weaver, J. Feunteun, T. Ashley, and D.M. Livingston. 1977. Association of BRCA1 with Rad51 in mitotic and meiotic cells. Cell 88: 265-275. Siiteri, P.K., N. Simberg, and J. Murai. 1986. Estrogens and breast cancer. Ann. N.Y. Acad. Sci. 464: 100-105. Snow, M.H.L. 1977. Gastrulation in the mouse: Growth and regionalization of the epiblast. J. EmbryoL Exp. Morphol. 42: 293-303. Srivastava, S., Z. Zou, K. Pirollo, W. Blattner, and E.H. Chang. 1990. Germ-line transmission of a mutated p53 gene in a cancer-prone family with Li-Fraumeni syndrome. Nature 348: 747-749. Sung, P. 1994. Catalysis of ATP-dependent homologous DNA pairing and strand exchange by yeast Rad51 protein. Science 265: 1241-1243. Sung, P. and D.L. Robberson. 1995. DNA strand exchange mediated by a Rad51-ssDNA nucleoprotein filament with po- larity opposite to that of RecA. Cell 82: 453-461. Tavtigian, S.V., J. Simard, J. Rommens, F. Couch, D. Shattuck- eidens, S. Neuhausen, S. Merajver, S. Thorlacius, K. Offit, D. Stoppalyonnet, C. Belanger, R. Bell, S. Berry, R. Bogden, Q. Chen, T. Davis, M. Dumot, C. Frye, T. Hattier, S. Jammu- lapati, T. Janecki, P. Jiang, R. Kehrer, J.F. Leblanc, J.T. Mitchell, J. Mcarthurmorrison, K. Nguyn, Y. Peng, C. Sam- son, M. Schroeder, S.C. Snyder, L. Steele, M. Stringfellow, C. Stroup, B. Swedlund, J. Swensen, D. Teng, A. Thomas, T. Tran, T. Tran, M. Tranchant, J. Weaverfeldhaus, A. Wong, H. Shizuya, J.E. Eyfjord, L. Cannonalbright, F. Labrie, M.H. Skolnick, B. Weber, A. Kamb, and D.E. Goldgar. 1996. The complete BRCA2 gene and mutations in chromosome 13q- linked kindreds. Nature Genet. 12: 333-337. Thorlacius, S., L. Tryggbadottir, G.H. Olafsdottir, J.G. Jonasson, H.M. Ogmundsdottir, H. Tulinius, and J.E. Eyfjord. 1995. Linkage to BRCA2 region in hereditary male breast-cancer. Lancet 346: 544-545. Thorlacius, S., G. Olafsdottir, L. Tryggvadottir, S. Neuhausen, J.G. Jonasson, S.V. Tavtigian, H. Tulinius, H.M. Og- mundsdottir, and J.E. Eyfjord. 1996. A single BRCA2 mutation in male and female breast-cancer families from Iceland with varied cancer phenotypes. Nature Genet. 13: 117-119. Vaughn, J.P., F.D. Cirisano, G. Huper, A. Berchuck, P.A. Futreal, J.R. Marks, and J.D. Iglehart. 1996. Cell-cycle control of BRCA2. Cancer Res. 56: 4590-4594.

1252 GENES & DEVELOPMENT Downloaded from genesdev.cshlp.org on October 4, 2021 - Published by Cold Spring Harbor Laboratory Press

Brca2 is required for embryonic cellular proliferation in the mouse.

A Suzuki, J L de la Pompa, R Hakem, et al.

Genes Dev. 1997, 11: Access the most recent version at doi:10.1101/gad.11.10.1242

References This article cites 49 articles, 15 of which can be accessed free at: http://genesdev.cshlp.org/content/11/10/1242.full.html#ref-list-1

License

Email Alerting Receive free email alerts when new articles cite this article - sign up in the box at the top Service right corner of the article or click here.