Centromeric Protein B Null Mice Are Viable with No Apparent Abnormalities

Centromeric Protein B Null Mice Are Viablewith No Apparent Abnormalities

Ana V. Perez-Castro,* Fay L. Shamanski,† Juanito J. Meneses,†TyAnna L. Lovato,* Kathryn G. Vogel,* Robert K. Moyzis,‡and Roger Pedersen†*Department of Biology, University of New Mexico, Albuquerque, New Mexico 87131;†Reproductive Genetics Unit, Department of Obstetrics, Gynecology and ReproductiveSciences, University of California at San Francisco, San Francisco, California 94143;and ‡Department of Biological Chemistry, University of California at Irvine, Irvine,California 92697

The centromere protein B (CENP-B) is a centromeric DNA/binding protein. It recognizes a 17-bp sequence motif called theCENP-B box, which is found in the centromeric region of most chromosomes. It binds DNA through its amino terminusand dimerizes through its carboxy terminus. CENP-B protein has been proposed to perform a vital role in organizingchromatin structures at centromeres. However, other evidence does not agree with this view. For example, CENP-B is foundat inactive centromeres on stable dicentric chromosomes, and also mitotically stable chromosomes lacking a-satellite DNAhave been reported. To address the biological function of CENP-B, we generated mouse null mutants of CENP-B byhomologous recombination. Mice lacking CENP-B were viable and fertile, indicating that mice without CENP-B undergonormal somatic and germline development. Thus, both mitosis and meiosis are able to proceed normally in the absence ofCENP-B. © 1998 Academic Press

Key Words: centromere; CENP-B; gene targeting; mitosis; meiosis.

INTRODUCTION

The centromere plays an important role in chromosomesegregation. This is the region of the chromosome wherethe kinetochore is formed, allowing the binding of thespindle microtubules, which are responsible for the separa-tion of the sister chromatids. In mammals, this centromericregion is composed of different amounts of nontranscribedrepetitive sequences, classically called satellite DNA (Par-due and Gall, 1970; Grady et al., 1992; Pluta et al., 1995).Centromeric proteins (CENP) were initially discoveredwith the aid of autoantibodies from patients with calcino-sis, Raynaud’s phenomenon, esophageal dysmotility,sclerodactyly, telangiectasia (CREST) or with scleroderma(Earnshaw and Rothfield, 1985; Valdivia and Brinkley,1985). Proteins recognized by the autoantibodies CENP-Athrough -C have been localized to the centromere byimmunoelectron microscopy (Cooke et al., 1990). CENP-Bis considered the major centromere antigen, since antibod-ies to it are present in the highest titers in all patients.CENP-B is an 80-kDa protein which binds through its

amino terminus to a 17-bp motif (CENP-B box) in thehuman centromeric a-satellite DNA (Pluta et al., 1992;Muro et al., 1992). Immunoelectron microscopy studiesshow that CENP-B is distributed throughout the hetero-chromatin beneath the kinetochore (Cooke et al., 1990).

The mouse CENP-B gene is an intronless single-copygene (Sullivan and Glass, 1991) highly conserved in itsDNA and protein sequences among species such as human,African green monkey, mouse, and hamster (Sullivan andGlass, 1991; Yoda et al., 1996; Bejarano and Valdivia, 1996).The CENP-B box is the only apparent sequence conservedin the satellite DNA of mammals such as human, Musmusculus, Mus caroli, tree shrews, giant panda, and diverseprimates (Masumoto et al., 1989a; Muro et al., 1992; Pietraset al., 1983; Kipling et al., 1995; Haaf and Ward, 1995; Wuet al., 1990; Haaf et al., 1995). This conservation suggeststhat CENP-B plays an important role in centromere func-tion. Recently, a CENP-B-related centromere DNA-bindingprotein was reported in Schizosaccharomyces pombe, andoverexpression of this protein affects mitotic chromosomestability. In addition, deletion of this yeast homolog results

DEVELOPMENTAL BIOLOGY 201, 135–143 (1998)ARTICLE NO. DB989005

0012-1606/98 $25.00Copyright © 1998 by Academic PressAll rights of reproduction in any form reserved. 135

in meiotic dysfunction and chromosome loss (Halverson etal., 1997). Because of its association to the a-satelliteheterochromatin and its dimerization domain at the car-boxy terminal, CENP-B has been proposed to be importantin organizing specific centromeric structures (Pluta et al.,1992; Kitagawa et al., 1995). Also, certain IgG preparationsof CREST serum that affect kinetochore function andmorphogenesis appeared to recognize only CENP-B in im-munoblots of chromosomal proteins (Bernat et al., 1991).However, other evidence is not consistent with this view.For example, the Y chromosome in humans does not bindCENP-B (Pluta et al., 1995) and mitotically stable chromo-somes lacking a-satellite DNA have been reported (Maio etal., 1981; Heller et al., 1996). To determine if CENP-B isessential in mammalian centromere function, we havegenerated mutant mice carrying a null mutation of theCENP-B gene.

MATERIALS AND METHODS

Isolation and Characterization of the CENP-BMouse Genomic DNA

To generate a probe for the CENP-B gene, an 11.5-day pcembryonic mouse cDNA library (Clontech ML1027a) was screenedby PCR, with primers CENPB7 59-GCTGCTGGACAT-AGCTGTACC-39 and CENPB8 59-CAGATTGGAGGTG-CCTGCTGT designed to produce a 626-bp fragment. PCR ampli-fication was carried out for 40 cycles, with denaturation at 94°C for1 min, annealing at 58°C for 1 min, and elongation at 72°C for 3min. The amplified fragment was directly sequenced by cyclesequencing in an ABI 373A automatic DNA sequencer (AppliedBiosystems, Inc., Foster City, CA) according to the protocolsprovided by the manufacturer, to verify that it was derived from theCENP-B gene. The PCR-generated probe was used to screen ansCos-1 genomic library derived from a 129 strain embryonic stemcell line (kindly provided by Dr. John Mudgett).

Targeted Deletion of the CENP-B Gene in ES Cells

To construct the targeting vector a 7.3-kb NotI–SalI genomicfragment was first subcloned into the O18S1 vector, which is amodification of psP73 vector (Promega). O18S1 has a NotI linker atthe EcoRV site and an SfiI linker at the PvuII site. The resultingsubclone was called V3. An AscI–AscI 660-bp fragment was deletedand a PGK (phosphoglycerate kinase)–neo poly(A)1 cassette wasinserted. The resulting clone containing the neo cassette was calledA183. A PGK–TK cassette was inserted at the 59 end of A183 andthe resulting vector, CENPB1KO, was electroporated into JM-1 EScells as described (Qui et al., 1995).

Animals, Breeding, and Genotyping

C57BL/6J blastocysts (The Jackson Laboratory) were used asrecipients for ES cell injection. Male chimeras generated fromblastocyst injection were mated to Swiss Black females (Tac:N:NIHS-BCfBR; Taconic). Null homozygous mice were generatedfrom heterozygous intercrosses. Crosses between F1, F2, F3, and F4

generations of different genotypes were carefully monitored for

viability and fertility. Genotyping was done by PCR of genomicDNA isolated from tails with primers as shown in Fig. 1A: a,59-CCACTTGTGTAGCGCCAAGTGCCA-39; b, 59-CCTTCGAG-TTGGCAGTGTAGTCGC-3 9 ; and c, 5 9 -CGACTTTCT-GTCGGACCAGGCCTC-39. PCR amplification was carried outfor 30 cycles with denaturation at 94°C for 30 s, annealing at 60°Cfor 30 s, and elongation at 72°C for 30 s. Southern blots were donefrom genomic tail DNA digested with SpeI and fractionated on a0.8% agarose gel. The gel was transferred onto a Hybond1 (Amer-sham) membrane and probed.

Tissue Culture

JM-1 ES cells were cultured and electroporated as previouslydescribed (Qui et al., 1995). CENP-B mutant and normal primarycell fibroblasts were freshly isolated from tissue that was minced in2.5 mg/ml collagenase in CO2-independent medium (Gibco, Cat.No. 18045-088) supplemented with 2% BSA. Minced tissue wasincubated in a shaker at 37°C for 30 min, then aspirated through an18-gauge needle and incubated for another 30 min. Cells wereaspirated again through an 18-gauge needle and then diluted andwashed two times. Cells were resuspended in DMEM completemedium and transferred to a 25- or 75-cm2 flask, depending on thesize (age) of the specimen. Cells were grown for 5 to 8 days beforetheir first passage.

Tubulin Staining

Cells growing on glass coverslips were washed with 13 PBS andfixed for 30 min in 600 mM Pipes, pH 6.9, 25 mM Hepes, 10 mMEGTA, and 2 mM MgCl2 containing 10% formaldehyde and 1%glutaraldehyde. The cells were washed 3 3 5 min with PBS, andfree aldehydes were blocked with 100 mM sodium borohydride inPBS for 20 min. Cells were then washed 3 3 5 min and blocked for30 min with 3% BSA in TST (Tris–HCl 10 mM, NaCl 0.15 M, andTween 20 0.2%). After being washed with TST the cells wereincubated with a 1/200 dilution of the anti-b-tubulin antibody(clone TUB 2.1; Sigma, T-4026) for 1 h, rinsed with TST, and thenincubated with 1/128 dilution of an anti-mouse IgG FITC-conjugated antibody (Sigma F5262) for 1 h. Cells were washed againand mounted with Vectashield.

Testis Histology

Testes samples from F3 homozygous mutants (2/2), heterozy-gotes (1/2), and wild-type (1/1) 17-week-old mice were fixed inBouins overnight at room temperature (RT) with gentle rotation.The tissue was washed several times in PBS and stored at 4°C in70% ethanol. The tissue was dehydrated through a series ofalcohols and then embedded in JB-4 ethyl methacrylate embeddingmedium (Polysciences). Sections (5 mm) were cut on a Sorvall JB4Amicrotome. Sections were baked for 1 h at 75°C, stained withhematoxylin Gill No. 3 and 0.6% eosin Y in distilled water, andmounted with Permount.

Sperm Counts

Male mice were sacrificed by cervical dislocation and the testesand epididymus removed. Surrounding fat was dissected away andthe caudal epididymus was carefully removed. The tissue was thenfinely minced in 1 ml of PBS and allowed to incubate at RT for 15min. The sample was then filtered through a 70-mm nylon mesh to

136 Perez-Castro et al.

Copyright © 1998 by Academic Press. All rights of reproduction in any form reserved.

remove tissue debris and somatic cells. The resulting sperm samplewas diluted in 5% formaldehyde in PBS and counted in a hemacy-tometer. Two counts were done for each sample and the averagewas reported. Counts are given as sperm/epididymus.

RNA Isolation and Northern Blot Analysis

Total RNA was isolated from lung fibroblasts with TRI Reagent(Molecular Research Center, Inc.) according to the manufacturer’sspecifications. Poly(A)1 RNA was isolated using the 5 Prime–3Prime, Inc., minioligo(dT) kit, following the manufacturer’s speci-fications. One microgram of purified poly(A)1 mRNA was fraction-ated on a 1.5% formaldehyde–agarose gel, transferred onto Gene-screen, and hybridized at 42°C with the CENP-B-radiolabeledprobe, following the manufacturer’s specifications. After beingwashed in 0.13 SSC/0.1% SDS at 60°C for 30 min, the filtermembrane was exposed to X-ray film, then rehybridized with afull-length cDNA G3PDH probe (Stratagene) and washed in 0.23SSPE/2% SDS at 65°C for 45 min, and then exposed to X-ray film.

Preparation of Primary Mouse Fibroblasts Nucleiand Nuclear Extracts

Nuclear and cytoplasmic fractions were isolated according toMasumoto et al. (1989b). The nuclear pellet was resuspended inextraction buffer (20 mM Hepes, pH 8.0, 0.5 mM EDTA, 0.1 mMPMSF, 0.5 mg/ml pepstatin A, 0.5 M NaCl, and 15% glycerol) andsonicated 43 for 10 s.

SDS–Page and Immunoblotting

Proteins were fractionated in a 12% SDS–PAGE gel and trans-ferred to nitrocellulose. The membrane was stained with PonceauS and then blocked with 3% dry milk in Block Buffer (10 mMTris–HCl, pH 7.4, 0.15 M NaCl, 0.2% Tween 20) for 1 h. The blotwas incubated for 1 h with a rabbit anti-CENP-B antibody 1:1000dilution (kindly provided by Dr. William Earnshaw) in Block bufferwith 1% dry milk. Five washes of 2 min each were done with Blockbuffer and then the blot was incubated with an HRP-conjugatedgoat anti-rabbit antibody, 1:2000 dilution (Sigma) in block bufferwith 1% dry milk for 1 h. The blot was washed 5 3 2 min anddeveloped with ECL reagents (Amersham).

Statistical Analysis

Probability (P) values were obtained using x2 analysis (Sigmastatversion 1.0) for breeding analysis in two-by-two contingency tableswith 1 degree of freedom. For body weight, testes weight, andsperm counts the nonparametric t test was used, and P values wereobtained using the unpaired Mann–Whitney two-tailed test (Sig-mastat version 1.0).

RESULTS

Targeting of the CENP-B Gene in ES Cells

Using a PCR-amplified mouse CENP-B probe, 8genomic CENP-B clones were obtained from an sCos-1/129 embryonic stem cell line library. Clone 7, whichcontained a .35-kb insert, was further characterized and

subcloned (Fig. 1A). A replacement vector was con-structed by deleting a 660-bp fragment that contained thestart of translation and the basic–Helix I–(loop)–Helix IImotif that binds the CENP-B box, and in its place aPGK-neo poly(A)1 cassette was inserted in the oppositeorientation of transcription to the CENP-B gene. Inaddition, a PGK-TK poly(A)1 cassette was inserted at the59 end of the construct, also in the opposite orientation oftranscription to the CENP-B gene. The resulting CENP-B1KO clone (Fig. 1B) was linearized with SfiI, electropo-rated into 129Sv/J ES cells, and subjected to G418 andFIAU selection. Nine correctly targeted ES cell clones of300 doubly resistant clones were identified by Southernblot analysis. The ES cell clones were expanded andmicroinjected into C57BL/6J blastocysts to generate chi-meric mice, three of which gave germline transmission.

FIG. 1. Disruption of the CENP-B gene. (A) Organization of thewild-type CENP-B mouse allele, the targeting vector, and themutated allele. Solid black box represents the CENP-B codingregion. Restriction sites are as follows: N, NotI; S, SacII; X, XhoI; C,ClaI. Probes 1 and 2 were used for Northern blot analysis and probe3 was used for Southern blot analysis. (B) Southern blot analysis ofthe progeny from a CENP-B heterozygous (1/2) intercross. The14-kb band represents the wild-type allele and the 8.4-kb bandrepresents the mutated allele. (C) PCR analysis of genomic DNA ofprogeny of the different crosses. Primers a and b were used for thedetection of the wild-type allele, yielding a product of 195 bp.Primers b and c were used for the detection of the mutated allele,yielding a product of 240 bp.

137Centromere Protein B-Null Mice


Generation of CENP-B-Deficient Mice

Male chimeras (41, 55, and 59) generated from threeindependent ES cell lines carrying the mutation were matedwith Swiss Black females. Germline transmission was firstfollowed by coat color and later confirmed by PCR andSouthern blot analysis (Figs. 1B and 1C). Male and femaleheterozygous mice were obtained (typically F1). These miceappeared normal and were fertile. The heterozygous micewere mated to obtain F2 homozygous mutants. The geno-type was determined for each pup in 17 F1 matings of thethree lines. One-quarter of the offspring were homozygousmutants (Table 1). CENP-B null mice did not show anyexcessive lethality and displayed no obvious abnormalphenotype. To verify that no CENP-B transcript was beingmade, poly(A)1 mRNA was isolated from wild-type (1/1),heterozygous (1/2), and homozygous mutant (2/2) ani-mals and Northern blot analysis was performed using twodifferent probes. The first probe spanned the deletion andthe second probe spanned 626 bp of the 39 end of the gene(Fig. 1A). No CENP-B message was detected in the homozy-gous mutant (2/2) mice using either probe 1 (Fig. 2) orprobe 2 (data not shown). Primary fibroblasts were culturedfrom wild-type (1/1), heterozygous (1/2), and homozygousmutant (2/2) mice. Western blot analysis failed to detectany CENP-B protein in either cytoplasmic or nuclear prepa-rations of the cultured cells from the homozygous mutant(2/2) mice (Fig. 3). When stained with an anti-b-tubulinantibody, microtubules in the mitotic spindle of homozy-

gous mutant (2/2) cells appeared normal and were nodifferent from those in the heterozygous (1/2) or wild-type(1/1) cells (Fig. 4). Analysis of microtubules in primarycultures of fibroblasts isolated from homozygous mutant(2/2) mice from F3 and F4 generations showed no spindleabnormalities during cell division (Fig. 4).

Breeding Analysis of CENP-B-Deficient Mice

Despite the apparently normal somatic development ofhomozygous mutants, the CENP-B protein could play a rolein meiosis that would impair fertility. To investigate thispossibility several intercrosses between CENP-B heterozy-gous (1/2) and homozygous mutant (2/2) mice of both F2

and F3 parental generations were carried out. The survival,litter size, and genotype for each litter were monitored.There was no significant difference in litter size in any ofthe F2 or F3 crosses (Table 2). Sections of testes from ahomozygous mutant (2/2) F3 male were examined by lightmicroscopy, but no differences in the stages of spermiogen-esis were seen compared to testis from heterozygous (1/2)siblings (Fig. ) or wild-type (1/1) mouse (data not shown).The 21-day survival of the progeny from matings of F2

homozygous mutant mice (2/2) was not significantly dif-ferent from F2 crosses involving heterozygous (1/2) mice(Table 3, B, C, D, E, and F).

Interestingly, the overall survival of offspring from mat-ings of F3 homozygous mutant mice was lower than forsimilar crosses in the F2 generation (Table 3, G versus A,P , 0.01). This could indicate an increased loss of homozy-gous mice in the F3 generation. However, a detailed com-parison of the patterns of F2 and F3 losses did not appear tosupport such a view. Thirty-one litters were producedcrossing F3 homozygous mutant mice. The litter size wasnormal (mean 6.3 6 2.0, range 3–13, Table 2). All of thepups died in 5 of these litters and between 2 and 5 pups diedin 6 other litters. This means that there was completesurvival in 65% of these 31 litters. By way of comparison, in

TABLE 1Genotypes of Offspring from CENP-B F1

Heterozygous Intercrosses

Line 1/1 1/2 2/2

41 12 21 1055 13 28 1659 11 19 8Total 36 68 34

FIG. 2. Northern blot analysis of lung fibroblasts from CENP-Bmutant mice. The blot was first probed for CENP-B and thenreprobed for G3PDH. The size of the G3PDH band is 1.27 kb andthe size of the CENP-B band is 2.9 kb. The locations of the 18S and28S ribosomal RNA are indicated.

FIG. 3. Immunodetection of the CENP-B protein. Extracts fromprimary lung fibroblasts of wild-type (1/1) and homozygous mu-tant (2/2) mice were fractionated into cytoplasmic (C) and nuclear(N) fractions and separated by SDS/PAGE. The Western blot wasdeveloped with an anti-CENP-B antibody. Arrowhead indicates theCENP-B protein in the nuclear fraction of the wild type (1/1) cells.KD, MW markers in kilodaltons.



the case of F2 homozygous mutant crosses, there wascomplete survival in only 50% of the 16 litters, but death ofall pups occurred in only 1 litter. There was completesurvival in 77% of 22 litters from the F2 heterozygouscrosses and death of all pups in 3 litters. Thus, the loweroverall survival in F3 compared with F2 homozygous mat-ings appeared to result from sporadic rather than systematiclosses, with a prevalent outcome of normal development.Interestingly, there was also lower survival of offspring inmatings in which only the F3 male was homozygous mu-tant and the F3 female was heterozygous (Table 3, K). In thisparticular case, all of the pups died in 3 of the 5 litters,making it very unlikely that only the homozygous mutantprogeny were dying. In addition, the fact that survival washigher in the progeny of an F3 homozygous mutant cross

compared to the cross of an F3 male homozygous mutantwith an F3 female heterozygous (Table 3, K) indicates thatF4 homozygous mice were not dying preferentially. Takingall the evidence into consideration, we conclude thatCENP-B null mice are normal in terms of both survival andfertility.

DISCUSSION

From the results of this study we conclude that theCENP-B gene is not essential for viability and that itsabsence does not affect fertility in the mouse. This conclu-sion is based on the following observations of CENP-B nullmice. Northern and Western blot analysis of homozygousmutant mice showed that neither the mRNA for CENP-Bnor the protein itself was present. The ratio of genotypesobtained from F1 heterozygous crosses was 1:2:1 (1/1:1/2:2/2), showing that there was no preferential absence of thehomozygous mutant gametes or progeny in such crosses. Inaddition, we found that litter size did not change in matingsof homozygous mutant F2 and F3 animals, meaning thatthere were no abnormal deaths of the homozygous mutantsduring embryonic development in later generations. Al-though there was somewhat lower survival among theprogeny of F3 homozygous mutant crosses, and even a lowersurvival in an F3 cross between a male homozygote and afemale heterozygote, this could not be clearly attributed tothe homozygous mutant genotype. To date we have ob-served 15 litters of normal size and viability from crossesbetween F4 homozygous mutants, suggesting that the F5

homozygous mutant mice are still viable. These findingsindicate that mitosis and meiosis in the mouse are able toproceed through several generations in the absence of theCENP-B protein.

FIG. 4. Immunostaining of microtubules in mouse primary fibroblasts. Microtubules from primary lung fibroblasts of wild type (1/1) andF3 homozygous mutant (2/2) mice were stained with an anti-b-tubulin antibody. The microtubules appear as fluorescent green.

TABLE 2Mean of Litter Sizes for Crosses of the Different Genotypesand Generations

ParentalGeneration

Genotypes ofcrosses

Number oflitters

Litter size(mean 6 SD)

F2 f2/2 3 m2/2 16 7.0 6 2.0F2 f1/2 3 m1/2 22 7.6 6 2.4F2 f2/2 3 m1/1 8 6.0 6 2.6F2 f2/2 3 m1/2 5 7.8 6 1.3F2 f1/1 3 m2/2 5 7.0 6 2.0F2 f1/2 3 m2/2 12 8.4 6 2.5F3 f2/2 3 m1/2 31 6.3 6 2.0F3 f2/2 3 m1/1 6 7.0 6 2.1F3 f2/2 3 m1/2 3 8.3 6 2.3F3 f1/1 3 m2/2 5 7.8 6 1.6F3 f1/2 3 m2/2 5 6.8 6 1.6



Very recently, Hudson et al. (1998) also reported micelacking CENP-B, with essentially similar conclusions,namely that mitosis and meiosis proceeded normally with-out CENP-B. However, they in addition report lower bodyweight, lower testis weight, and sperm content reduction intheir null mutants. In light of this finding we weighed F3

homozygous (2/2) mutant mice and compared their weightto F3 heterozygous (1/2) and wild-type (1/1) siblings. Themale mice analyzed were either 16 or 22–24 weeks old. Incontrast to the findings of Hudson et al. (1998) we did notfind any significant reduction in body weight in 16- or 22- to24-week-old null mice. The body weight of our malehomozygous mutant mice (2/2) at 16 weeks (N 5 3) was29.3 6 5.79 g (mean 6 SD), not significantly different fromthat of the heterozygous (1/2) and wild-type siblings (1/1)(N 5 4) 34.5 6 2.279 g (P 5 0.4, nonparametric t test). At22–24 weeks male homozygous mutant (2/2) mice (N 5 4)had a mean body weight of 32.5 6 5.09 g, which was notsignificantly different from that of heterozygous (1/2) andwild-type (1/1) siblings (N 5 3) with a mean body weight of

36 6 5.29 g (P 5 0.62). The testes from the same animalswere dissected and weighed, showing values of 0.187 60.039 g for the F3 homozygous (2/2) mutants at 16 weeksand 0.201 6 0.01 g for their heterozygous (1/2) and wild-type (1/1) siblings (P 5 0.4, t test). For the mice between 22and 24 weeks of age, the mean testes weight for F3 homozy-gous mutants (N 5 4) was 0.181 6 0.039 g and for theheterozygous and wild -type siblings (N 5 3) was 0.22 60.049 g (P 5 0.22). Sperm counts done on these same micedid not reveal any significant difference between null miceand their control littermates (P 5 0.22 for 16 weeks and P 50.4 for 22- to 24-week-old mice). Epidydimal sperm countsof 150 6 4.7 3 106 for the 16-week-old F3 homozygousmutants (N 5 4) and of 108 6 2.6 3 106 for the siblingcontrols (N 5 3) were obtained. Epidydimal sperm counts of8.85 6 1.1 3 106 for 22- to 24-week-old homozygousmutant mice and of 12.5 6 4.5 3 106 for the sibling controlswere obtained. Therefore, in our case the CENP-B homozy-gous mutants until at least the F3 generation also did notshow any significant difference from the controls in testisweight or sperm count. From the results of Hudson et al.(1998) this discrepancy could be explained by differences inthe strain of the mice used, as in some other null mutants(e.g., EGF receptor, Miettinen et al., 1995; Sibilia andWagner, 1995; Threadgill et al., 1995).

Among all of the centromeric proteins recognized by theCREST autoantibodies, CENP-B is the best characterized atthe nucleotide and protein levels (Earnshaw et al., 1987;Yoda et al., 1996). CENP-B dimerizes through its carboxyterminus domain, and it is associated through its aminoterminus to the CENP-B box, which is a degenerate motiffound in centromeric a-satellite sequences (Masumoto etal., 1989a; Bernat et al., 1991; Pluta et al., 1992). These factstaken together led to the expectation of an important rolefor CENP-B in establishing the chromatin structure of thecentromere (Masumoto et al., 1989b; Bernat et al., 1991;Pluta et al., 1992; Sugimoto et al., 1994).

However, other evidence conflicts with a functional role forCENP-B in mitosis. Most vertebrate species do not have aCENP-B box in their centromeric DNA, and the human Ychromosome a-satellite DNA array also lacks a CENP-B box(Maio et al., 1981; Broccoli et al., 1990; Moens and Pearlman,1990; Grady et al., 1992; Goldberg et al., 1996). Moreover,functional, mitotically stable human chromosomes devoid ofa-satellite DNA have been observed, and these centromeresdo not bind CENP-B (Heller et al., 1996; Du Sart et al., 1997).Finally, in dicentric chromosomes CENP-B is found at both

FIG. 5. Hematoxylin- and eosin-stained plastic testes tissue sections from F3 homozygous mutant (2/2) and heterozygous (1/2) CENP-Bmale mice (age 17 weeks). (A and B) Variation of the seminiferous tubule sperm content within each testis is apparent. Asterisks indicatetubules with very few mature sperm, and arrowheads indicate tubules filled with mature sperm. (C and D) Higher magnification of a tubulefull of sperm. There was no apparent difference in the stages of sperm differentiation between homozygous mutant (2/2) CENP-B mice orheterozygous (1/2) siblings. MS, mature sperm; PS, primary spermatocyte; RS, round spermatid. Magnification: (A,B) bar indicates 100 mm,(C, D) bar indicates 20 mm.

TABLE 321-Day Survival in Crosses of F2 and F3 Generationsa

Comparison

Parentalgeneration and

genotype Live/total % Survival P value

(1)A F2 f2/2 3 m2/2 103/114 90B F2 f1/2 3 m1/2 153/169 91 n.s.C F2 f2/2 3 m1/1 41/51 80 n.s.D F2 f2/2 3 m1/2 38/39 97 n.s.E F2 f1/1 3 m2/2 27/27 100 n.s.F F2 f1/2 3 m2/2 91/101 90 n.s.

(2)G F3 f2/2 3 m2/2 142/190 75H F3 f2/2 3 m1/1 41/42 97 ,0.01I F3 f2/2 3 m1/2 22/25 88 n.s.J F3 f1/1 3 m2/2 38/39 97 ,0.01K F3 f1/2 3 m2/2 10/34 30 ,0.001

a 21-day survival of offspring from homozygous mutant (2/2)crosses of F2 (1) and F3 (2) generations were compared (by x2

analysis) to homozygous mutant crosses, heterozygous crosses, orcrosses in which only the male or the female was a homozygousmutant. In (1) B through F are compared to A, and in (2) H to K arecompared to G. n.s., not statistical significant, P . 0.05.



functional and nonfunctional centromeres (Earnshaw et al.,1989), suggesting that its presence is not sufficient for centro-meric function in mitosis.

Alternative functions for this gene are suggested by therecent finding, based on sequence similarity, that theCENP-B protein is related to the Tigger transposable ele-ment transposase (Smit and Riggs, 1996). It has been pro-posed that CENP-B may be involved in the evolution ofsatellite DNA arrays by promoting nicking and homologousrecombination (Kipling and Warburton, 1997). Alterna-tively, it may be a DNA binding protein, serendipitouslyrecruited for chromosome condensation, whose functioncan be easily replaced by other proteins. Since CENP-B isnot essential for mitosis or meiosis in mice, it will beinteresting to mate the CENP-B null mice with null mu-tants of other centromeric proteins to test further forfunction and redundancy of these proteins.

ACKNOWLEDGMENTS

We thank Dr. John Mudgett for kindly sharing with us his ESgenomic library, and Dr. William Earnshaw for kindly sharing hisanti-CENP-B antibody. We also thank Margaret Flannery for hertechnical assistance with the plastic tissue sections. This work wassupported by DOE Grant DE-FG03-97ER62485 to R.K.M., NIHGrants HD26732 and ES08750 to R.A.P., HHMI Grant 76296-549901 to UCSF School of Medicine, College of Arts & Sciences ofUNM, and Minority Biomedical Research Support Program of theNIH (GM 52576) to K.G.V. F.L.S. was supported by NIEHS Post-doctoral Training Grant T32 ES07106.

REFERENCES

Bejarano, L. A., and Valdivia, M. M. (1996). Molecular cloning of anintronless gene for the hamster centromere antigen CENP-B.Biochim. Biophys. Acta 1307, 21–25.

Bernat, R. L., Delannoy, M. R., Rothfield, N. M., and Earnshaw,W. C. (1991). Disruption of centromere assembly during inter-phase inhibits kinetochore morphogenesis and function in mito-sis. Cell 66, 1229–1238.

Broccoli, D., Miller, O. J., and Miller, D. A. (1990). Relationship ofmouse minor satellite DNA to centromere activity. Cytogenet.Cell Genet. 54, 182–186.

Cooke, C., Bernat, R. L., and Earnshaw, W. C. (1990). CENP-B: Amajor human centromere protein located beneath the kineto-chore. J. Cell Biol. 110, 1475–1488.

Du Sart, D., Cancilla, M. R., Earle, E., Mao, J., Saffery, R., Tainton,K. M., Kalitsis, P., Martyn, J., Barry, A. E., and Choo, K. H. A.(1997). A functional neo-centromere formed through activationof a latent human centromere and consisting of non-alpha-satellite DNA. Nat. Genet. 16, 144–153.

Earnshaw, W. C., and Rothfield, N. (1985). Identification of a familyof human centromere proteins using autoimmune sera frompatients with scleroderma. Chromosoma 91, 313–321.

Earnshaw, W. C., Sullivan, K. F., Machlin, P. S., Cooke, C. A.,Kaiser, D. A., Pollard, T. D., Rothfield, N. F., and Cleveland,D. W. (1987). Molecular cloning of cDNA for CENP-B, the majorhuman centromere autoantigen. J. Cell Biol. 104, 817–829.

Earnshaw, W. C., Ratrie, H., and Stetten, G. (1989). Visualization ofcentromere proteins CENP-B and CENP-C on stable dicentricchromosome in cytological spreads. Chromosoma 98, 1–12.

Goldberg, I. G., Sawhney, H., Pluta, A. F., Warburton, P. E., andEarnshaw, W. C. (1996). Surprising deficiency of CENP-B bindingsites in African green monkey a-satellite DNA: Implications forCENP-B function at centromeres. Mol. Cell. Biol. 16, 5156–5168.

Grady, D. L., Ratliff, R. L., Robinson, D. L., McCanlies, E. C.,Meyne, J., and Moyzis, R. K. (1992). Highly conserved repetitiveDNA sequences are present at human centromeres. Proc. Natl.Acad. Sci. USA 89, 1695–1699.

Haaf, T., and Ward, D. C. (1995). Rab1 orientation of CENP-B boxsequences in Tupaia belangeri fibroblasts. Cytogenet. CellGenet. 70, 258–262.

Haaf, T., Mater, A. G., Wienberg, J., and Ward, D. C. (1995).Presence and abundance of CENP-B box sequences in great apesubsets of primate-specific alpha satellite DNA. J. Mol. Evol. 41,487–491.

Halverson, D., Baum, M., Stryker, J., Carbon, J., and Clarke, L.(1997). A centromere DNA-binding protein from fission yeastaffects chromosome segregation and has homology to humanCENP-B. J. Cell Biol. 136, 487–500.

Heller, R., Brown, K. E., Burgtorf, C., and Brown, W. R. (1996).Mini-chromosomes derived from the human Y chromosome bytelomere directed chromosome breakage. Proc. Natl. Acad. Sci.USA 93, 7125–7130.

Hudson, D. F., Fowler, K. J., Earle, E., Saffery, R., Kalitsis, P.,Trowell, H., Hill, J., Wreford, N. G., de Krester, D. M., Cancilla,M. R., Howman, E., Hii, L., Cutts, S. M., Irvine, D. V., and Choo,K. H. A. (1998). Centromere protein B null mice are mitoticallyand meiotically normal but have lower body and testis weights.J Cell Biol. 141, 309–319.

Kipling, D., Mitchell, A. R., Masumoto, H., Wilson, H. E., Nicol, L.,and Cooke, H. J. (1995). CENP-B binds a novel centromericsequence in the Asian mouse Mus caroli. Mol. Cell. Biol. 15,4009–4020.

Kipling, D., and Warburton, P. E. (1997). Centromeres, CENP-B andTigger too. Trends Genet. 13, 141–145.

Kitagawa, K., Masumoto, H., Ikeda, M., and Okazaki, T. (1995).Analysis of protein–DNA and protein–protein interactions ofcentromere protein B (CENP-B) and properties of the DNA–CENP-B complex in the cell cycle. Mol. Cell. Biol. 15, 1602–1612.

Maio, J. J., Brown, F. L., and Musich, P. R. (1981). Toward amolecular paleontology of primate genomes. I. The Hind III andEco RI dimer families of alphoid DNAs. Chromosoma 83,103–125.

Masumoto, H., Sugimoto, K., and Okazaki, T. (1989a). Alphoidsatellite DNA is tightly associated with centromere antigens inhuman chromosomes throughout the cell cycle. Exp. Cell Res.181, 181–196.

Masumoto, H., Masukata, H., Muro, Y., Nozaki, N., and Okazaki,T. (1989b). A human centromere antigen (CENP-B) interacts witha short specific sequence in alphoid DNA, a human centromericsatellite. J. Cell Biol. 109, 1963–1973.

Moens, P. B., and Pearlman, R. E. (1990). Telomere and centromereDNA are associated with the cores of meiotic prophase chromo-somes. Chromosoma 100, 8–14.

Miettinen, P. J., Berger, J. E., Meneses, J., Phung, Y., Pedersen, R. A.,Werb, Z., and Derynck, R. (1995). Epithelial immaturity andmultiorgan failure in mice lacking epidermal growth factorreceptor. Nature 376, 337–341.



Muro, Y., Masumoto, H., Nozaki, N., Oshashi, M., and Okazaki, T.(1992). Centromere protein B assembles human centromericalpha-satellite DNA at the 17-bp sequence, CENP-B box. J. CellBiol. 116, 585–596.

Pardue, M. L., and Gall, J. G. (1970). Chromosomal localization ofmouse satellite DNA. Science 168, 1356–1358.

Pietras, D. F., Bennett, K. L., Siracusa, L. D., Woodworth-Gutai, M.,Chapman, V. M., Gross, K. W., Kane-Haas, C., and Hastie, N. D.(1983). Construction of a small Mus musculus repetitive DNAlibrary: Identification of a new satellite sequence in Mus muscu-lus. Nucleic Acids Res. 11, 6965- 6983.

Pluta, A. F., Saitoh, N., Goldberg, I., and Earnshaw, W. C. (1992).Identification of a subdomain of CENP-B that is necessary andsufficient for localization to the human centromere. J. Cell Biol.116, 1081–1093.

Pluta, A. F., Mackay, A. M., Ainsztein, A. M., Goldberg, I. G., andEarnshaw, W. C. (1995). The centromere: Hub of chromosomalactivities. Science 270, 1591–1594.

Qui, M., Bulfone, A., Martinez, S., Meneses, J. J., Shimamura, K.,Pedersen, R. A., and Rubenstein, J. L. R. (1995). Null mutation ofDlx-2 results in abnormal morphogenesis of proximal first andsecond branchial arch derivatives and abnormal differentiation inthe forebrain. Genes Dev. 9, 2523–2538.

Sibilia, M., and Wagner, E. F. (1995). Strain-dependent epithelialdefects in mice lacking the EGF receptor. Science 269, 234–238.

Smit, A. F. A., and Riggs, A. D. (1996). Tiggers and DNA transposonfossils in the human genome. Proc. Natl. Acad. Sci. USA 93,1443–1448.

Sugimoto, K., Hagishita, Y., and Himeno, M. (1994). Functionaldomain structure of human centromere protein B. Implication ofthe internal and C-terminal self-association domains in centro-meric heterochromatin condensation. J. Biol. Chem. 269, 24271–24276.

Sullivan, K. F., and Glass, C. A. (1991). CENP-B is a highlyconserved mammalian centromere protein with homology to thehelix–loop–helix family of proteins. Chromosoma 100, 360–370.

Threadgill, D. W., Dlugosz, A. A., Hansen, L. A., Tennenbaum, T.,Lichti, U., Yee, D., LaMantia, C., Mourton, T., Herrup, K., Harris,R. C., Barnard, J. A., Yuspa, S. H., Coffey, R. J., and Magnuson, T.(1995). Targeted disruption of mouse EGF receptor: Effect ofgenetic background on mutant phenotype. Science 269, 230–234.

Valdivia, M. M., and Brinkley, B. R. (1985). Fractionation and initialcharacterization of the kinetochore from mammalian metaphasechromosomes. J. Cell Biol. 101, 1124–1134.

Wu, Z. A., Liu, W. X., Murphy, C., and Gall, J. (1990). Satellite 1DNA sequence from genomic DNA of the giant panda Ailu-ropoda melanoleuca. Nucleic Acids Res. 18, 1054–1070.

Yoda, K., Nakamura, T., Masumoto, H., Suzuki, N., Kitagawa, K.,Nakano, M., Shinjo, A., and Okazaki, T. (1996). Centromereprotein B of African green monkey cells: Gene structure, cellularexpression, and centromeric localization. Mol. Cell. Biol. 16,5169–5177.

Received for publication April 15, 1998Revised July 1, 1998

Accepted July 1, 1998



Centromeric Protein B Null Mice Are Viable with No Apparent Abnormalities

Documents

Transcript of Centromeric Protein B Null Mice Are Viable with No Apparent Abnormalities