Genetic Analysis of Jumbled Spine and Ribs (Jsr) Mutation Affecting the Vertebral Development in Mice

P1: ZCK

Biochemical Genetics [bigi] pp605-bigi-450082 September 6, 2002 13:39 Style file version June 7th, 2002

Biochemical Genetics, Vol. 40, Nos. 9/10, October 2002 (C© 2002)

Genetic Analysis of Jumbled Spine and Ribs (Jsr)Mutation Affecting the Vertebral Developmentin Mice

Shinya Okano,1 Atsushi Asano,1 Yasuhiro Kon,1 Hiroyuki Miyoshi, 1

and Tomomasa Watanabe1,2

Received 18 October 2001—Final 15 February 2002

The jumbled spine and ribs (Jsr) mouse was derived from a spontaneous muta-tion. As the phenotype, a shortened trunk and kinky tail are characteristicJsrtraits. In this study, on high resolution mapping it was found thatLunatic fringe(Lfng) mapped at the same position asJsr. Lfng was identified as the candidategene forJsr, but sequence analysis of this gene revealed no substitution in thecoding region of cDNA. Therefore, we adopted the strategy of positional cloningfor Jsrusing a mouse bacterial artificial chromosome (BAC) library. A BAC con-tig was constructed from three BAC clones showing positive signals ofLfng and11MMHAP75FRD8.seqnear theJsrlocus on chromosome 5. Based on the geneticmapping of both T7 and sp6 ends of a clone of BAC382-O-7 (BAC382), theJsrgene was considered to exist in BAC382 and to be positioned near the sp6 side.

KEY WORDS: Jsr; vertebral development;Lfng; BAC.

INTRODUCTION

Abnormalities of the axial skeleton are observed in various animals, includinghumans and mice. The genes related to their development are likely to be similarin basic construction in species from insects to vertebrates, for example homeoboxgenes. This suggests that the genes for some developmental phenomena havecommon ancestors in all animals. However, the process of vertebral development

1 Department of Experimental Animal Science, Graduate School of Veterinary Medicine, HokkaidoUniversity, Sapporo, Japan.

2 To whom correspondence should be addressed; e-mail: [email protected].

311

0006-2928/02/1000-0311/0C© 2002 Plenum Publishing Corporation

P1: ZCK


312 Okano, Asano, Kon, Miyoshi, and Watanabe

in vertebrates, beginning with somite formation, is very different from that ofDrosophila. The developmental genes associated with the Notch signaling pathwayhave recently been studied by using knockout mice. The results from these micehave made it clear that the genes involved in the Notch signaling pathway play animportant role during somitogenesis (Aulehla and Johnson, 1999; Barranteset al.,1999; Dale and Pourquie, 2000; Rawlset al., 2000).Lunatic fringe(Lfng) is oneof the genes associated with the Notch signaling pathway and definitely influencessomitogenesis from the abnormal structure ofLfng knockout mice (Evrardet al.,1998; Moloneyet al., 2000; Zhang and Gridley, 1998).

The jumbled spine and ribs (Jsr) mutation in mice, producing a short trunk andkinky tail, originated spontaneously in the mouse strain CTA/Idr and its characterhas been described previously (Miyoshiet al., 1999). A mating experiment showedthat theseJsrabnormalities were due to a single autosomal dominant gene. In thisstudy, a candidate gene approach concerningLfng, which is related to somiteformation and located nearJsr, was carried out forJsr. Furthermore, to find thecausal gene ofJsr, the strategy of positional cloning was adopted with a bacterialartificial chromosome (BAC) contig.

MATERIALS AND METHODS

Mice

The CKH strain carrying theJsr mutation is maintained as an inbred strain inour laboratory. The maintenance is usually done by sib mating between mutant(Jsr/+) and normal (+/+) mice. The MOG strain derived from Japanese wildmice (Mus m. molossinus) and the NJL strain derived from Danish wild mice(Mus m. musculus) are also maintained as inbred strains in our laboratory. Wedesignated the normal alleles corresponding toJsras+C for CKH,+M for MOG,and+N for NJL. The intersubspecific backcross progeny used to determine thechromosomal position of theJsr locus were generated between F1 carrying theJsrallele from the cross of CKH-Jsr/+C×MOG-+M/+M and CKH-+C/+C mice.NJL-Jsr/+N congenic mice were also produced in our laboratory and used togenerateJsr homozygous mice.

Restriction Fragment Length Polymorphism (RFLP)

For RFLP ofLfngand sequence tagged site (STS) marker11.MMHAP75FRD8.seq(11MMseq), genomic DNA was digested by restriction enzymeDraI, electro-phoresed, transferred to a Hybond-N+membrane (Amersham Pharmacia BiotechUK Ltd., England), and hybridized with each appropriate probe amplified by PCR.The probe ofLfngwas amplified using the forward primer 5′-CCGTCAAGACCACCAGAAGTT (bp 454-474) and the reverse primer 5′-TCATACTCCACAGCCAT

P1: ZCK


Genetic Analysis of Jumbled Spine and Ribs Mutation 313

CTTGC (bp 658-637). The probe of11MMseqwas amplified using the forwardprimer 5′-AGGGGACTTGACACAGTTGG and the reverse primer 5′-GCAATGGAGAGACTCTTCGC, expected to give a product of 181 bp. For RFLP of the sp6end of a mouse genomic BAC clone of BAC382-O-7 (BAC382), genomic DNA wasdigested byEcoRI and hybridized with its sp6 end fragment, which was amplifiedby PCR; the forward primer sequence was 5′-TGGCTGCAAGACCTTACCTGand the reverse one was 5′-CATCCTTGCTTAGTCCCAAC, giving a product of474 bp.

Simple Sequence Length Polymorphism (SSLP)

PCR amplifications were performed to detect SSLPs of microsatellite markerD5Mit37 and the T7 end of the mouse genomic BAC382 clone. PCR primersfor D5Mit327were purchased from Research Genetics (Huntsville, AL) and usedas recommended (Dietrichet al., 1992).D5Mit327was used as a genetic markerfor distinguishing between alleles of CKH and NJL to generate CKH-Jsr/+N con-genic mice. Primers for SSLPs of the T7 end of BAC382 were synthesized in thisstudy according to the 470-bp sequence of BAC ends. The sequence of the forwardprimer was 5′-ATGAACCCTAAGTCAGCAAC and that of the reverse one was5′-CTCAGACTCACACTATTACAAG. The PCR cycling profile was composedof an initial denaturating step for 2 min at 95◦C, followed by 35 cycles at 95◦C for1 min, 57◦C for 1 min, 72◦C for 1 min, and final extention at 72◦C for 4 min.

Reverse Transcription-Polymerase Chain Reaction (RT-PCR)

Total RNA from the adult spleen and brain was isolated by use of an Isogen kit(Nippon Gene, Japan) and 3µg was used for synthesis of first-strand cDNA byusing Super Script II (Life Technologies, Rockville, MD). TheLfngcoding regionwas amplified with PCR from spleen cDNA. To amplify a partial region ofLfng, theforward primer was 5′-TGGCACTGGGATAGATATTA (bp 66-85) and the reverseprimer was 5′-GCCTCTACAATGTAGCCAATGG (bp 997-976). For the secondpartial region the forward primer used was 5′-TCGATCTGCTGTTCGAGACC(bp 490-509) and the reverse one was 5′-ACGACTGCTAGAAGATGGCG(bp 1248-1229), and for the third region the sequence of the forward primer was5′-CATTGTAGAGGCTCTGCTGG (bp 986-1005), and that of the reverse onewas 5′-GAGACTGCTACAATCAGCACC (bp 1571-1551).

DNA Sequencing

Coding regions ofLfng cDNA, and each end of the BAC382, BAC433-D-8(BAC433), and BAC489-P-1 (BAC489) clones were sequenced in this study.pGEM-T Easy vector (Promega, Madison, WI) was used for cloning after PCR

P1: ZCK



amplification. Sequencing was performed with an ABI Prism 377 DNA Sequencer(Perkin Elmer Cetus, Norwalk, CT) and an ABI Prism Big Dye Terminator CycleSequencing Ready Reaction Kit (Perkin Elmer Cetus).

Isolation of Mouse BAC Clones

A mouse BAC library (Research Genetics) was first screened with PCR ampli-fication for 11MMseq, which was mapped to the same position asJsr. To de-tect positive signals for11MMseqand Lfng, PCR was performed on BAC382,BAC433, and BAC489 clones with their specific primer sets. The primer se-quences for11MMseqwere described above. The forward primer forLfng was5′-CTGTTGGCTTGTCTCCTGG (bp 153-173), and the reverse primer was5′-AGGTCTCGAACAGCAGATCG (bp 510-491). The molecular length of eachBAC clone showing positive signals for11MMseqandLfng was determined bypulse field gel electrophoresis (PFGE). PFGE was carried out for 10–20 s at thepulse time on 1.5% agarose gel at 14◦C for 15 h. Then Southern blot analysis onthe membrane transferred from PFGE was performed with an11MMseqprobe toconfirm its internal signal in the BAC clones.

B1-Interspersed Repetitive Sequence PCR (B1 IRS-PCR) Experiment

To determine the overlaps of the BAC clones, a B1 IRS-PCR experiment wasperformed. The B1 repetitive sequence is specific to rodents (Nocket al., 1999),and the BAC clones used in this study had to have its sequence. B1 IRS-PCR wascarried out with the specific primer (5′-AGTTCCAGGACAGCCAGGG) for B1repetitive sequences.

Ligation Mediated-PCR (LM-PCR)

Linkers for LM-PCR, consisting of both long (5′-GCGGTGACCCGGGAGATCTGAATTC) and short (5′-GAATTCAGATC) polynucleotides, were used. Beforeligation, both linkers were incubated and annealed for 15 min at 65◦C and furtherfor 30 min at room temperature. The BAC clones were digested by a blunt-end-type restriction enzyme (DraI, EcoRV or HaeIII) and ligated to the linkers withLM-PCR. The ligated construct between the T7 or sp6 end of BAC clones and thelinker was amplified with PCR primers specific to the external region. The primerfor the external region of the T7 end was 5′-GTCACGACGTTGTAAAACGACGGCCAGT, and that for external region of sp6 end was 5′-AGGCTTTACACTTTATGCTTCCGGCTCG. The linker primer was the long polynucleotide described above.Furthermore, PCR products after dilution were amplified by nested PCR withprimers specific to the internal region of the ligated construct between the T7 or

P1: ZCK



sp6 end of BAC clones and the linker. The primer for the internal region of the T7end was 5′-ACGACTCACTATAGGGCGAATTCGAG, and that for the sp6 endwas 5′-TTTCACACAGGAAACAGCTATGA. The linker primer was the same asdescribed above. These PCR products were subcloned into pGEM-T Easy Vector(Promega, Madison, WI) and sequenced by the method previously described.

PCR Amplification for Each BAC End

To determine the identification and direction of each BAC clone, the PCR primerswere designed to amplify both the T7 and sp6 ends of BAC382 and BAC489.The specific primers for PCR of the T7 end of BAC382 have already been described.The forward primer for the sp6 end of BAC382 was 5′-ATGAACCCTAAGTCAGCAAC and the reverse primer was 5′-CTCAGACTCACACTATTACAAG. Theforward primer for the T7 end of BAC489 was 5′-TCTGCAGGGCAGTTTCTCTGand the reverse primer was 5′-GGCTAAGAAGCCATCAGGAC. For the sp6 endof BAC489 the forward and reverse primers were 5′-ATACACCTGAGGGGGTGAAG and 5′-TGAATTCTCTAGTTAACGCTGC, respectively.

RESULTS

High-Resolution Mapping

We previously reported a high-resolution map near theJsr locus, using 1026 back-cross progeny from (CKH-Jsr/+C×MOG-+M/+M ) Fl-Jsr/+M× CKH-+C/+C

(Miyoshi et al., 1999). Furthermore, in this study, we revised the map with addi-tional markers ofLfngand11MMseqas a step toward positional cloning ofJsr. Ofthe 1026 backcross progeny, 13 mice carrying the recombination betweenMafk(=Nfe2u) andD5Mit327were examined in detail (Fig. 1). As the result of mappinganalysis,Lfng and11MMseqwere mapped to the same position asJsr. The newgenetic map is shown in Fig. 1(B).

Analysis ofLfng in Jsr/Jsr Mice

SinceLfng was suggested to be a candidate gene for theJsr gene from the abovegenetic mapping, and the morphological defects ofLfng knockout mice (Evrardet al., 1998) resembled those ofJsr, mRNA expression and cDNA sequence ofLfng were analyzed inJsr mice. The expression ofLfng mRNA was confirmedin the heart, kidney, liver, lung, pancreas, skin, spleen, and thymus in adult miceby RT-PCR. No difference of expression was observed in any tissue examinedby RT-PCR betweenJsr/Jsr and+N/+N . Because the expression ofLfng mRNAwas detected strongly in the spleen, sequencing analysis was performed usingLfng cDNA of spleen from bothJsr/Jsr and+N/+N adult mice. The results of

P1: ZCK



Fig. 1. Linkage analysis ofJsron mouse chromosome 5. Segregation profiles of 13 micecarrying the recombination betweenMafk (=Nfe2u) andD5Mit327out of 1026 back-cross progeny of (CKH-Jsr/+C×MOG-+M/+M ) Fl-Jsr/+M× CKH-+C/+C . Thefilled boxes represent the CKH allele, and open boxes mean the MOG allele (A). Esti-mated linkage map and intergenomic distances are shown between the loci in cM plusstandard errors (B).

sequencing analysis ofLfng cDNA were compared with the cDNA sequence ofthe C3H/He strain reported by Cohenet al. (1997), and no difference betweenthem was observed in the amino acid sequence of the coding region.

Construction of BAC Contig Near Jsr

Four positive clones, BAC382, BAC433, BAC489, and BAC20-H-20 (BAC20),were detected by PCR screening for11MMseqprimers (Fig. 2). Also, the BACclones all showed positive PCR signals forLfng-specific primers. The molecular

P1: ZCK



Fig. 2. Agarose gel electrophoresis for the detection ofLfngand11MMseqfrom BAC382, BAC433,and BAC489 clones by PCR. BothLfngand11MMseqsignals were detected in all three BAC clones.

lengths of BAC clones were determined by PFGE, and calculated to be 250 kb forBAC382, 240 kb for BAC489, 130 kb for BAC433, and 70 kb for BAC20.

BAC382, BAC489, and BAC433 clones were used for the construction of aBAC contig nearJsr, because BAC20 had a very small length. As the result ofPCR amplification of both T7 and sp6 ends of BAC382 and BAC489 clones, thesecontained their own ends (Fig. 3). BAC382 did not contain either end of BAC489,but BAC489 included both the T7 and sp6 ends of BAC382. However, the length ofBAC489 (240 kb) was shorter than that of BAC382 (250 kb). Therefore, BAC489was thought to have a deleted region as shown in Fig. 3(B).

Furthermore, a B1 IRS-PCR experiment was carried out to examine the over-laps of the three BAC clones. As shown in Fig. 4, BAC382 contained bands 1, 2,3, 4, and 5, BAC433 included bands 1, 2, 3, and 4, and BAC489 had only bands 1and 6. The lengths of bands 1, 2, 3, 4, 5, and 6 were estimated to be 2.5, 1.8, 1.0,0.8, 0.7, and 0.6 kb respectively. From the above results, the overlap of the BACclones related to these bands was depicted as shown in Fig. 4(B).

Genetic Mapping of BAC Ends

After LM-PCR, sequences of about 500 bp of both T7 and sp6 ends from BAC382and BAC489 were determined. On the other hand, neither the T7 nor the sp6 end ofBAC433 was analyzed in the mapping study because primers could not be designedowing to GT-rich sequence in the T7 end and nine repeats of TTTA sequence inthe sp6 end. To detect SSLP or RFLP of the BAC382 and BAC489 ends, PCR andSouthern blot analysis were performed using primers and probes, respectively,made from the sequence information of the BAC ends. SSLP was found out on the

P1: ZCK



Fig. 3. Agarose gel electrophoresis for the detection of T7 and sp6 ends of BAC382 and BAC489clones by PCR (A). Both BAC clones contained their own ends. BAC382 did not contain either endof BAC489, but BAC489 included both T7 and sp6 ends of BAC382. Schematic representation of theoverlap of two BAC clones (B).

T7 end of BAC382, and RFLP withEcoRI digestion was detected on the sp6 endof BAC382, but neither SSLP nor RFLP was detected on BAC489. Twelve micecarrying the recombination betweenMafk(=Nfe2u) andCdx2(Fig. 1) were used tomap the T7 and sp6 ends of BAC382. Figure 5 shows the genetic typing of both T7and sp6 ends of BAC382 by SSLP and RFLP experiments. Recombination betweenthe T7 end andJsr was observed in 10 mice, and that betweenJsr and the sp6end was observed in only 1. In another mouse there was no recombination among

P1: ZCK



Fig. 4. Agarose gel electrophoresis of B11RS-PCR. In total, six bands of fragments were detectedin the clones from BAC382, BAC433, and BAC489 (A). BAC382 contained bands 1, 2, 3, 4, and5; BAC433 included bands 1, 2, 3, and 4; and BAC489 had only bands 1 and 6. The overlap ofBAC clones based on their bands of repeated sequences was estimated (B).

the T7 end,Jsr, and the sp6 end. Consequently, theJsr gene must be included inBAC382 and be positioned near the sp6 side (Fig. 5(B)), because recombinationbetweenJsrand the sp6 end occurred only in 1 mouse but recombination betweenthe T7 end andJsroccurred in 10. From the results of this mapping study, a BACcontig nearJsr was finally constructed as shown in Fig. 6.

DISCUSSION

High resolution mapping revealed thatLfngwas located at the same position asJsr(Fig. 1), because no recombination between them was observed in 1026 backcrossprogeny.Lfngalters regional expression periodically in the presomitic mesoderm

P1: ZCK



Fig. 5. Upper photograph shows genetic typing of the T7 end of BAC382 by SSLP in 12 micecarrying recombination nearJsrout of 1026 backcross progeny of (CKH-Jsr/+C×MOG-+M/+M )Fl-Jsr/+M× CKH-+C/+C (A). The band of 300 bp is derived from the MOG (M) allele and thatof 150 bp is from the CKH (C) allele. Middle lane represents theJsr morphological genotypes of+M/+C (M) andJsr/+C (C). Lower photograph shows typing of sp6 end of BAC382 by Southernblot analysis. The band of 3.9 kbp is from the MOG (M) allele and that of 3.3 kbp is from the CKH (C)allele.∗ represents a recombination point. The linkage analysis amongJsrand the ends of BAC382 issummarized (B). The filled boxes represent the CKH allele, and open boxes correspond to the MOGallele.

(Aulehla and Johnson, 1999). Furthermore,Lfngobstructs the interaction betweenNotchandSerrate, which are a receptor and a ligand of the Notch signaling path-way, respectively, and promotes the relationship between Notch and another ligandDelta (Hrabeet al., 1997). These signals are thought to play an important role

P1: ZCK



Fig. 6. The revised BAC contig nearJsr. ∗ represents a recombination point betweenMafkandD5Mit292 in 1026 backcross mice.Lfng and11MMseqare included in all three BAC clones.Jsr is considered to be located near the sp6 end of BAC382.

during somitogenesis in the regulation of the Notch signaling pathway (Johnstonet al., 1997; Moloneyet al., 2000).Lfng knockout mice have a phenotype verysimilar toJsr mice (Evrardet al., 1998; Zhang and Gridley, 1998). As highLfngmRNA expression was seen in the spleen of adult mice by RT-PCR, we performedsequencing analysis ofLfng from spleen mRNA forJsr/Jsr,+N/+N , and C3H/Hemice. The sequence ofLfngin Jsr/Jsrand+N/+N mice agreed completely with thatin the database derived from the C3H/He embryo (Accession Number AF015768).In addition it was reported recently that the sequence of theLfngcoding region ofthe adult mouse (Accession Number AK004642) corresponds to that of the em-bryo. Therefore, the cause ofJsrat the molecular level did not arise in the codingregion ofLfng, although there is still another possibility that a mutation of theintron or promoter region leads to a splicing defect, specifically in its embryonicLfngmRNA, or a decrease in its expression, respectively. Generally, it is quite im-probable that theJsrmutation occurs in theLfnggene, because theLfngknockoutmice are reported to be recessive (Evrardet al., 1998; Zhang and Gridley, 1998)but theJsr mutation is dominant, which would produce an abnormal protein thatdisturbs the normal process of somitogenesis.

After making a high resolution map (Fig. 1) for the positional cloning ofJsr, a BAC contig nearJsr was constructed (Fig. 6). The whole length of theBAC contig was about 380 kb. To decide the position ofJsr in the contig, geneticmapping of both the T7 and sp6 ends of the BAC382 clone was attempted. Both

P1: ZCK



ends of BAC382 were successfully mapped, and the indication was thatJsr mustbe included in BAC382 and located near the sp6 side. BAC489 included both theT7 and sp6 ends of BAC382 (Fig. 3). However, the molecular length of BAC489(240 kb) was shorter than that of BAC382 (250 kb). Therefore, BAC489 wasthought to have a deleted region and not to be suitable for positional cloning ofJsr. Furthermore, from the results of B1 IRS-PCR, the overlap of these BAC cloneswas consistent with the construction of the BAC contig. Thus, BAC382 has a highpossibility to include theJsrgene, but the genetic size of the insert is still too largeto analyze the nucleotide sequence directly. It will be necessary to obtain smallfragments containingJsrsubdivided by sonication or restriction enzyme digestion,which can make the mapping and sequencing study of theJsr causal gene easier.

REFERENCES

Aulehla, A., and Johnson, R. L. (1999). Dynamic expression of lunatic fringe suggests a link betweennotch signaling and an autonomous cellular oscillator driving somite segmentation.Dev. Biol.207:49.

Barrantes, I. B., Elia, A. J., Wunsch, K., De Angelis, M. H., Mak, T. W., Rossant, J., Conlon, R. A.,Gossler, A., and de la Pompa, J. L. (1999). Interaction between Notch signalling and Lunaticfringe during somite boundary formation in the mouse.Curr. Biol. 9:470.

Cohen, B., Bashirullah, A., Dagnino, L., Campbell, C., Fisher, W. W., Leow, C. C., Whiting, E., Ryan,D., Zinyk, D., Boulianne, G., Hui, C., Gallie, B., Phillips, R. A., Lipshitz, H. D., and Egan, S. E.(1997). Fringe boundaries coincide with Notch-dependent patterning centres in mammals andalter Notch-dependent development inDrosophila. Nat. Genet. 16:283.

Dale, K. J., and Pourquie, O. (2000). A clock-work somite.Bioessays22:72.Dietrich, W., Katz, H., Lincoln, S. E., Shin, H. S., Friedman, J., Dracopoli, N. C., and Lander, E. S.

(1992). A genetic map of the mouse suitable for typing intraspecific crosses.Genetics131:423.Evrard, Y. A., Lun, Y., Aulehla, A., Gan, L., and Johnson, R. L. (1998).Lunatic fringeis an essential

mediator of somite segmentation and patterning.Nature394:377.Hrabe, de A. M., McIntyre, J., II, and Gossler, A. (1997). Maintenance of somite borders in mice

requires theDeltahomologueDllI. Nature 386:717.Johnston, S. H., Rauskolb, C., Wilson, R., Prabhakaran, B., Irvine, K. D., and Vogt, T. F. (1997). A

family of mammalian Fringe genes implicated in boundary determination and the Notch pathway.Development124:2245.

Miyoshi, H., Kon, Y., Seo, K. Y., Jin, H. K., Hasegawa, A., and Watanabe, T. (1999). Jumbled spineand ribs (Jsr): A new mutation on mouse chromosome 5.Mamm. Genome10:213.

Moloney, D. J., Panin, V. M., Johnston, S. H., Chen, J., Shao, L., Wilson, R., Wang, Y., Stanley, P.,Irvine, K. D., Haltiwanger, R. S., and Vogt, T. F. (2000). Fringe is a glycosyltransferase thatmodifies Notch.Nature406:369.

Nock, C., Gauss, C., Schalkwyk, L. C., Klose, J., Lehrach, H., and Himmelbauer, H. (1999). Technol-ogy development at the interface of proteome research and genomics: Mapping nonpolymorphicproteins on the physical map of mouse chromosomes.Electrophoresis20:1027.

Rawls, A., Wilson-Rawls, J., and Olson, E. N. (2000). Genetic regulation of somite formation.Curr.Top. Dev. Biol.47:131.

Zhang, N., and Gildley, T. (1998). Defects in somite formation inlunatic fringe-deficient mice.Nature394:374.

Documents

Genetic Analysis of Jumbled Spine and Ribs (Jsr) Mutation Affecting the Vertebral Development in Mice