An integrated physical, genetic and cytogeneticmap around the sunn locus of Medicagotruncatula

E. Schnabel, O. Kulikova, R.V. Penmetsa, T. Bisseling, D.R. Cook, and J. Frugoli

Abstract: The sunn mutation of Medicago truncatula is a single-gene mutation that confers a novel supernodulationphenotype in response to inoculation with Sinorhizobium meliloti. We took advantage of the publicly availablecodominant PCR markers, the high-density genetic map, and a linked cytogenetic map to define the physical and ge-netic region containing sunn. We determined that sunn is located at the bottom of linkage group 4, where a fine-structure genetic map was used to place the locus within a ~400-kb contig of bacterial artificial chromosome (BAC)clones. Genetic analyses of the sunn contig, as well as of a second, closely linked BAC contig designated NUM1, indi-cate that the physical to genetic distance within this chromosome region is in the range of 1000 –1100 kb�cM–1. Theratio of genetic to cytogenetic distance determined across the entire region is 0.3 cM�µm–1. These estimates are in goodagreement with the empirically determined value of ~300 kb�µm–1 measured for the NUM1 contig. The assignment ofsunn to a defined physical interval should provide a basis for sequencing and ultimately cloning the responsible gene.

Key words: FISH, physical to genetic distance, Medicago truncatula, map-based cloning.

Résumé : La mutation sunn chez le Medicago truncatula est une mutation monogénique qui confère un nouveau phé-notype de supernodulation en réponse à l’inoculation avec le Sinorhizobium meliloti. Les auteurs ont tiré avantage demarqueurs PCR codominants, d’une carte génétique à haute densité et d’une carte cytogénétique, tous du domaine pu-blique, afin de déterminer les positions physique et génétique du locus sunn. Les auteurs ont trouvé que le locus sunnest situé sur le bas du groupe de liaison 4 (LG4) où une carte physique fine de la région a été employée afin de locali-ser ce locus au sein d’un contig de clones BAC (« bacterial artificial chromosomes ») mesurant 400 kpb. Une analysegénétique du contig contenant le locus sunn et d’un contig BAC voisin, appelé NUM1, a révélé que le ratio des distan-ces physique/génétique varie entre 1000 et 1100 kb/cM. Le ratio des distances génétique/cytogénétique a été évalué à0,3 cM/µm. Ces estimés sont en bon accord avec le ratio de 300 kb/cM mesuré empiriquement pour le contig NUM1.La localisation du locus sunn au sein d’un intervalle physique bien défini devrait servir de tremplin pour séquencer etcloner le gène responsable.

Mots clés : FISH, ratio des distances physique/génétique, Medicago truncatula, clonage positionnel.

[Traduit par la Rédaction] Schnabel et al. 672

Introduction

The symbioses of legumes and Rhizobium bacteria havelong been of interest, both for practical agricultural applica-tions and for the insight they provide into plant–microbeinteractions. Despite the long history of legume mutantsaffected in nodule development, relatively few of the af-

fected nodulation genes have been cloned and the plant ge-netic pathways for nodulation are only beginning to be char-acterized (for examples see Sagan et al. 1995; Penmetsa andCook 1997; Keen et al. 1999; Catoira et al. 2000; Penmetsaand Cook 2000; Wais et al. 2000; Walker et al. 2000;Kawaguchi et al. 2002). Thus, compared with Arabidopsis,where the impact of cloning single-gene mutations on theunderstanding of plant processes has been huge, plantnodulation mutants have had a relatively minor impact onour understanding of nodulation. All four plant genes in-volved in nodulation that have been cloned, one byinsertional mutagenesis, one by orthology to an Arabidopsismutant, and the other two by positional cloning of EMS mu-tations, are implicated in signal transduction. The nin locusidentified in Lotus japonicus is predicted to encode a tran-scription factor like protein with a proposed role in sensingplant nitrogen status (Schauser et al. 1999). The astray mu-tant in L. japonicus encodes a leucine-zipper protein(Nishimura et al. 2002a). Two additional genes have beenshown to encode putative receptor kinases. The dmi2 geneof Medicago truncatula and its orthologs in Medicago sativaand Lotus japonicus are predicted to encode a receptor

Genome 46: 665–672 (2003) doi: 10.1139/G03-019 © 2003 NRC Canada

665

Received 24 September 2002. Accepted 21 February 2003.Published on the NRC Research Press Web site athttp://genome.nrc.ca on 5 June 2003.

Corresponding Editor: J.B. Bell.

E. Schnabel and J. Frugoli1. Department of Genetics andBiochemistry, Clemson University, Clemson, SC, U.S.A.O. Kulikova and T. Bisseling. Department of Plant Sciences,Laboratory of Molecular Biology, Wageningen University,Wageningen, The Netherlands.D. Cook and R.V. Penmetsa. Department of PlantPathology, University of California at Davis, Davis, CA,U.S.A.

1Corresponding author (e-mail: jfrugol@clemson.edu).

I:\gen\gen4604\G03-019.vpJuly 8, 2003 8:29:02 AM

Color profile: Generic CMYK printer profileComposite Default screen

kinase required for development both of symbiotic nitrogenfixation and mycorrhizal associations (Endre et al. 2002;Stracke et al. 2002), whereas the HAR1/NARK gene identi-fied in L. japonicus (Nishimura et al. 2002b; Krusell et al.2002) and soybean (Searle 2003) encodes a receptor kinaseaffecting nodule number; interestingly, HAR1/NARK is ho-mologous to the CLAVATA1 gene of Arabidopsis thalianaand is predicted to play a role in long-distance signaling ofnodule meristem formation.

Numerous other mutants with phenotypes that affect sym-biosis in various ways await molecular characterization. Thesunn supernodulation mutant of M. truncatula is one suchmutant. Plants homozygous for the sunn mutation produceapproximately 10 fold more nodules than wild-type plants.Despite this increase in nodule number, the nodules retaintheir correct spatial orientation in the root (across from xy-lem poles), and the nodules form within the same develop-

mental zone of the root as on wild-type plants (Penmetsa etal. 2003).

Medicago truncatula is a self-fertile relative of the impor-tant forage crop alfalfa. However, in contrast to alfalfa andseveral other crop legume species, M. truncatula has a rela-tively small (~5 × 108 bp) genome, autogamous and diploidgenetics, and prolific seed production, all of which contrib-ute to its desirability as a molecular genetic system to studyaspects of legume biology. In addition to its biological at-tributes, numerous genetic and genomics resources havebeen developed that facilitate the map-based cloning ofnodulation genes (Cook et al. 1997; Cook 1999; Harrison2000; Frugoli and Harris 2001; Oldroyd and Geurts 2001).Among these resources are a collection of >170 000 ex-pressed sequence tags (ESTs) (Covitz et al. 1998; Györgyeyet al. 2000; Bell et al. 2001) and corresponding bio-informatic databases (Quackenbush et al. 2000; Lamblin et

© 2003 NRC Canada

666 Genome Vol. 46, 2003

Band size (bp)

Marker name (type) Primers (5′→3′) TA (°C) Enzyme A17 A20

18L14L CGTAACATTCTCATTATCGCTGCTAT 55 StyI 412 232+180(CAPS) AAGTAATCCGGTGATTGATTTTTCTCC

ES63B19R* GGTTTCTTCTCTGCAATATTCAATGTT 52 AluI 306+132+7 438+7(CAPS) AGAAGAGCTCGGTAAAAGGTCAGTATC

NUM1 GATGCTGCTCCTGTTGTTGTTTC 55 TaqI 243+48+29 272+48(CAPS) AAAACAAAGTAAAAACAATATCTTTAAAAATC

ES22H10L* GAAATGAATGAATGGAACTT 49 — 515 563†

(LP) AAATTGAACTTACTATGACACAES22H10LB GAAATGAATGAATGGAACTT 51 — — 514(AS-PCR) TTCATGCACCTACTATGACACAES75A2L* CACATATGCTATAAGCTGAAAATAAGTTGG 53 — 514 515

(SNP) TTTTGGTACCATATTGCAGTCTTTGTAAAT 8 T’s at nt335 9 T’s at nt335ES61G8R* ATGCAGGTTTTCCATGTAACAATGA 52 — 299 299

(SNP) AGTGAGCCACCAGACTAGTTTGACAT (nt134 = T) (nt134 = C)MtLAX5* CCCCTCATGGTATTTCACTTG 50 — 564 508

(LP) CATTATTTCCCTGTACCAACCAACES19E14L ATTCTATGATGCAAGGCTCGTG 53 Sau3A 300+190+129+28 300+190+157

(dCAPS) GCACAATTTTCATTTAATGGAGACAAGATDNABP CCCTATGAGCTTGGGTTTGTCT 53 AflII 580+280+260 860+260

(CAPS) CTCTGGCATACGTGTTCAGCES72D11R TATATTATGGTCGGAGTTCG 48 PstI 357+116+35 357+151

(CAPS) AAATTTCTTGGTAGTAGTTAGTAES10P20-1 TTAAAAGTGTGCTCCTCCTCGTAGT 50 — 510 511

(SNP) TAACCTTCGATTCTATATTGGACAC (nt103 = A)ES10P20-2 ACGGATATTCTGTCCAACCCTACCT 54 BstUI 397 531+46

(CAPS) ACTTGGGCCTTCAGTCATGCTCGTAES10P20R* GTTGGCGCAATGTGTTGGTCTCT 52 DraI 542

(CAPS) AAATGTGAGTTTGGTGGCAGTADK43 GGGACTAATTAAAAGAGAGAAAAAGAAAA 55 RsaI 300+50 350

(CAPS) CGCGTGAGCCTCTTGAGCTTGATGCAPX GCTTTGCCAAACACATCCCTC 55 SspI 400+270 670

(CAPS) CAATTACTGGTTATACCTTTTCTCGG

Note: CAPS, cleaved amplified polymorphic sequences; LP, length polymorphism; SNP, single nucleotide polymorphism; dCAPS, derived CAPS.*Sequences of these BAC ends are deposited as GenBank accession Nos. BH836470-BH836473, CC454084-CC454086, AY115845, and AC134242.†Owing to kinetics of the PCR, this band does not amplify in the presence of the A17 product. Determination of A17 and het genotype of the plant

requires additional use of ES22H10LB.

Table 1. Markers used in this analysis.

I:\gen\gen4604\G03-019.vpJuly 8, 2003 8:29:02 AM

Color profile: Generic CMYK printer profileComposite Default screen

al. 2003), as well as coordinated physical mapping andwhole-genome sequencing efforts. As a species,M. truncatula is native to the Mediterranean basin, wheremany polymorphic ecotypes have been collected and curated(Bonin et al. 1996a; Bonin et al. 1996b). Researchers havedeveloped over 500 co-dominant PCR-based genetic mark-ers based on ESTs, bacterial artificial chromosome (BAC)end sequences, and simple sequence repeat (SSR) poly-morphisms that function across a range of characterizedecotypes. The majority of these genetic markers have beenplaced within the context of an emerging physical map, andthe corresponding BAC clones have been used in fluores-

cence in situ hybridization (FISH) experiments to produce adetailed cytogenetic map (Kulikova et al. 2001).

Recent advances in FISH have expanded the use of thistool in plants (for review see de Jong et al. 1999). In thecase of M. truncatula, the availability of an integrated ge-netic and cytogenetic map (Kulikova et al. 2001) permits theuse of FISH analysis of pachytene chromosomes to aid map-based cloning of a gene (e.g., Ané et al. 2002). In this study,we used the combined resources of codominant PCR-basedgenetic markers, an integrated genetic and cytogenetic map,and an emerging physical map to delimit the sunn locus tothe bottom of linkage group 4 and to develop a fine-structure

© 2003 NRC Canada

Schnabel et al. 667

Fig. 2. Physical contigs around NUM1 marker and sunn locus. Numbers below the chromosome line indicate recombination eventsbetween markers in the population, whereas those above indicate distance in cM. Markers are described in Table 1. BAC numbers referto the Mt_ABb library deposited at the Clemson University Genomics Institute. When the exact location of a marker on a BAC is notknown, it is indicated by a bracket with two arrows flanking the region of the BAC where the marker lies. Scale on bottom is distance(kb) determined by fingerprint data (see Materials and methods) with BACs used for FISH analysis (Fig. 3A) indicated in color.

Fig. 1. (A) Gross positioning of sunn on LG 4 as determined by Mapmaker version 2 (see Materials and methods). Distances on theleft side of chromosome are in centimorgans. Markers are from http://www.medicago.org. (B) Fine mapping of sunn in larger popula-tion (see Materials and methods). Distances to the right of the chromosome line are in centimorgans (cM).

I:\gen\gen4604\G03-019.vpJuly 8, 2003 8:29:03 AM

Color profile: Generic CMYK printer profileComposite Default screen

genetic and physical map of the region. We also determinedthe relationship between the physical, genetic, andcytogenetic distances to more precisely characterize thegenomic context of sunn.

Materials and methods

Genetic mappingFor purposes of genetic mapping, individuals homozygous

for the sunn mutation (in the genetic background of geno-type A17) were hand emasculated and crossed with pollenfrom ecotype A20, as described by Penmetsa and Cook(2000). F1 progeny were allowed to self pollinate, resultingin the production of multiple, independent F2 populationsconsisting of thousands of F2 seed. Plants from 8 individualsunn × A20 F2 populations (1771 F2 progeny) were scoredfor the sunn phenotype as described by Penmetsa et al.(2003) and a subset of 287 sunn homozygotes was identi-fied. These plants were used to map the sunn mutation rela-tive to cleaved amplified polymorphic sequence (CAPS)markers (Table 1) that were previously developed againstEST and BAC sequences (e.g., Kulikova et al. 2000; Cook etal., personal communication). The genetic marker informa-tion is available as a *.pdf file at http://www.medicago.org.Briefly, DNA was isolated from F2 progeny using QiagenDNAeasy Plant kits (Qiagen Inc., Valencia, Calif.). PCRfragments were amplified from genomic DNA using primersspecific for each marker, digested with the appropriate re-striction enzyme, and run on agarose gels to revealpolymorphisms. Because each marker behaves in acodominant fashion, all allelic combinations were readilyscored. The resulting genotype assignments were compiledand linkage groups assigned using Mapmaker 2.0 (White-head Institute) with the Haldane mapping function (LOD =3.0, recombination fraction = 0.40.

For initial mapping of the sunn locus, the analysis focusedon a small group of 29 individuals homozygous for the sunnmutation. Once linkage to markers on chromosome 4 wasdetermined, the mapping population was expanded to in-clude all 287 F2 progeny homogygous for the sunn locus.This analysis defined the precise position of sunn relative topreviously developed genetic markers.

Analysis of BAC clones and chromosome walkingTo isolate BAC clones and initiate a chromosome walk to-

wards the sunn locus, PCR products corresponding to thenearest flanking genetic markers were labeled with 32P-dCTP by random priming and used to screen a subset ofhigh-density nylon filters of a 25× BAC library (Mt_ABblibrary publicly available at http://www.genome.clemson.edu/orders/). Hybridization was performed according to theprotocols posted at the Clemson University Genomic Insti-tute Web site (http://www.genome.clemson.edu/protocols/).Positive clones were verified by PCR using primers corre-sponding to the original genetic marker. The resulting BACclones were subjected to restriction enzyme fingerprintingand assembled into a physical contig by manual analysis ofrestriction patterns. The size of individual BAC clones wasestimated by adding fragment sizes determined by migrationrelative to known molecular standards. During the walk, thegenetic map was continually refined by construction of new

genetic markers based on sequence information from theends of individual BAC clones (see markers designated bythe prefix ES in Table 1). Briefly, PCR primers developedfrom BAC end sequence information were used to amplifyand sequence DNA from the parental genotypes A17 andA20. Polymorphisms were converted to markers and used toscore cross-over events within the physical map. For thispurpose, genotype analysis focused exclusively on plantspreviously determined to contain recombination eventsflanking sunn. Amplified BAC end fragments were used ashybridization probes to isolate the next BACs in the walk.BAC end sequences used to generate new markers were de-posited in GenBank under the accession numbers shown inTable 1.

Fluorescence in situ hybridizationThree-color FISH mapping on M. truncatula accession

Jemalong A17 pachytene chromosomes was performed as in

© 2003 NRC Canada

668 Genome Vol. 46, 2003

Fig. 3. Three-color FISH of BAC clones to M. truncatulaJemalong A17 chromosomes at late pachytene. (A) FISH map-ping of the end BACs of the NUM1 contig (BAC 63H18, greensignal; BAC 63B19, red signal) and BAC 182C19 (DK43marker, yellow signal) on the long arm of chromosome 4.(B) FISH mapping of BAC 63H18 (red signal), BAC 63B19(green signal), and BAC 70L14 (DNABP marker, yellow signal).Arrow indicates centromere position of chromosome 4. Scale bar =5 µm.

I:\gen\gen4604\G03-019.vpJuly 8, 2003 8:29:13 AM

Color profile: Generic CMYK printer profileComposite Default screen

Kulikova et al. (2001), using the indicated BACs as probes.BAC DNA was labeled by nick translation either withdigoxigenin-11-dUTP (green fluorescence after detection),biotin-16-dUTP (red fluorescence), or both in a 1:1 ratio(yellow fluorescence after merging red and green signals).Slides were analyzed under an epifluorescence microscope(Zeiss, Oberkochen–Jena, Germany) equipped with a CCDcamera. FISH signals of different probes were captured indi-vidually and merged using the GENUS program (AppliedImaging, New Castle Upon Tyne, U.K.). Measurements ondigitized images were made using the Image ProPlus pro-gram (Media Cybernetics, Silver Springs, Md.). Each mea-surement was done in 7–8 cells.

Results

Initial positioning of sunnAn F2 population derived from a cross between sunn (in

the genetic background of genotype A17) and the polymor-phic ecotype A20 was used to map the sunn mutant relativeto CAPS markers corresponding to known gene sequences.In initial mapping experiments, 29 individuals homozygousfor the sunn mutation were analyzed to establish a linkagegroup affiliation for the locus. This limited analysis was suf-ficient to assign sunn to a gene-rich region of M. truncatulalinkage group (LG) 4 (Fig. 1A).

The small size of the initial mapping population did notpermit precise positioning of sunn relative to several closelylinked genetic markers on the lower arm of LG 4. Therefore,a total of 287 individuals homozygous for the sunn mutationwere analyzed to refine the relationship between sunn andflanking CAPS markers (Table 1). The resulting linkagemap, shown in Fig. 1B, places sunn between NUM1 andmarkers APX and DK43. These later two markers could notbe ordered based on recombinant individuals in the availableF2 population.

Before initiating a chromosome walk towards the sunnlocus, we sought to determine the orientation of the markersAPX and DK43. CAPS markers APX and DK43 were usedas probes to identify M. truncatula BAC library clones andcontigs for each marker were assembled with no overlap be-tween the contigs apparent in restriction enzyme fingerprints(data not shown). Owing to the absence of recombinants inour population between APX and DK43, and as a prelude tochromosome walking, we used three-point FISH analysis todetermine the orientation of the corresponding BAC clones.The results demonstrated an order of NUM1, (sunn), DK43,and APX (H. Choi, D. Kim, T. Uhm, E. Limpens, H. Lim,G. Kiss, P. Kalo, O. Kulikova, T. Bisseling, and D. Cook,unpublished data). Therefore, DK43 and NUM1 were usedas the flanking markers for the walk.

Walking toward sunnBAC end sequences from clones identified by the NUM1

probe were used to design PCR primers. DNA fragmentsamplified with these primers were labeled with [32P]dCTPand used in hybridization experiments to isolate additionalBAC clones from the library. After confirmation by PCRwith the primer sets, the neighboring BAC clones werefingerprinted, and the distal end of the longest BAC was se-quenced and used for the next “step” in the walk. The proce-dure was repeated, and BAC end sequences were used todesign new markers. These new genetic markers, designatedby the prefix ES (Table 1), were designed based on sequenceanalysis of PCR products amplified from the two mappingparents. In our experience, in the region surrounding sunn,three out of four BAC end sequences with non-repetitiveDNA sequences yielded a polymorphism. The rate of poly-morphism (typically single nucleotide polymorphism (SNP))between the A17 and A20 parents has previously been estab-lished to be approximately 1 SNP/150 bp in non-coding,single-copy regions of the genome (H. Choi and D. Cook,unpublished data). The new ES markers were scored on theabove F2 population to identify recombination within thephysical map and thus orient the contig relative to sunn.Using this iterative procedure, a contig of approximately415 kb was constructed around the NUM1 marker, hereafterreferred to as the NUM1 contig (Fig. 2).

FISH analysis of the walkWithin the NUM1 contig, genetic markers ES75A2L and

ES63B19R span 0.35 cM and encompass an estimated350 kb of DNA, suggesting a relatively low rate of recombi-nation within this physical interval (i.e., ~1 Mb DNA/cM).With the intent of estimating physical–genetic relationshipsin this region of chromosome 4, and in particular to estimatethe physical distance to sunn, we conducted FISH analysiswith genetically mapped BAC clones. We used BAC clones63H18 and 63B19 (Fig. 2) from the ends of the NUM1contig as a reference to calibrate physical distance. The mid-points of these clones were estimated from physical mapdata to be 300 kb apart (see Fig. 2). Fluorescently labeledBAC clones were hybridized to pachytene-stage chromo-somes, and the physical distance between the centers of hy-bridization signals was measured in micrometres. Thedistance values shown in Table 2 are the average of measure-ments conducted on a minimum of seven nuclei. A represen-tative hybridization is shown in Fig. 3A. Initially, we soughtto estimate the physical distance between DK43 and theNUM1 contig, as this interval defines the extremities of thechromosome walk. As shown in Fig. 3A, three-point hybrid-ization with the NUM1 contig BAC clones (63H18 and63B19) and the DK43 BAC clone (182C19) was sufficient toestimate the physical distances between the centers of eachhybridization signal. In particular, the distance from 63H18

© 2003 NRC Canada

Schnabel et al. 669

BACs Distance (µm±SD) Recombination events cM cM�µm–1 Bases (kb)

182C19-63H18 9.90±0.07 17 2.96 0.30 unknown70L14-63H18 2.93±0.90 6 1.04 0.35 unknown63H18-63B19 1.11±0.41 2 0.35 0.32 355

Table 2. Cytogenetic and genetic distances between BAC clones (µm).

I:\gen\gen4604\G03-019.vpJuly 8, 2003 8:29:13 AM

Color profile: Generic CMYK printer profileComposite Default screen

© 2003 NRC Canada

670 Genome Vol. 46, 2003

(at one end of the NUM1 contig) to BAC 182C19 is 9.90 ±0.07 µm and covers 3.0 cM (17 recombination events)(Fig. 3A; Table 2), yielding a genetic to cytogenetic distanceratio of 0.3 cM�µm–1. Because the centers of BAC clones63H18 and 63B19 are separated by approximately 300 kb,the relationship between cytogenetically defined and physi-cal map based distance estimates is approximately300 kb�µm–1. By extrapolation, we estimate that the sunn in-terval (i.e., from ES75A2L to 182C19) corresponds to ap-proximately 3 Mb of DNA.

During the course of the chromosome walk describedabove, EST marker DNABP was identified and mapped nearNUM1 (H. Choi and D. Cook, unpublished data). When ana-lyzed on the expanded F2 population (described above),DNABP mapped to the genetic interval between DK43 andNUM1, and no recombinants were identified between sunnand DNABP. Based on hybridization of 32P-labeled DNABPprobe to high-density BAC library filters (Nam et al. 1999),we identified clone 70L14, which was analyzed by three-point FISH along with clones 63B19 and 63H18 (Figs. 2 and3B). As shown in Table 2, the cytogenetic distance between70L14 and 63H18 was determined to be 2.93 ± 0.90 µm.The resulting genetic to cytogenetic distance ratio in therange of 0.3–0.5 cM�µm–1 is consistent with the values de-termined for 63H18 to 182C19 (Fig. 2; Table 2). If the esti-mate of 300 kb�µm–1 obtained for the NUM1 contig is alsoconserved throughout this region, then the distance between70L14 and 63H18 would be approximately 900 kb.

With the intent of further defining the sunn region, weused a BAC from the Mt_ABb library corresponding to70L14 (clone 31B9) as the starting point from which a 400-kb contig was constructed. The effort was aided by an ongo-ing physical mapping project for M. truncatula (D.J. Kim, J.Gish, and D. Cook, unpublished data). BAC end sequencingand genetic marker development from flanking BAC clones(i.e., genetic markers ES10P20R and ES19E14L, Table 1)established that the sunn locus resides within this geneticand (or) physical interval (Fig. 2). However, none of the ge-netic markers between ES10P20R and ES19E14L (i.e.,ESP10P20-2, ES10P20-1, ES72D11R, and DNABP) re-vealed recombination events with the sunn contig, and thusthe current mapping population is not adequate to furthernarrow the sunn locus within the ~400-kb BAC contig.

Discussion

Abundant PCR-based genetic markers and a good molecu-lar genetic system like that present in M. truncatula allowedrapid placement of the sunn mutation on the bottom of LG 4based on a relatively small mapping population (Fig. 1A),while the combination of genetic, physical and cytogeneticmapping resources allowed a more precise placement of thelocus on the genetic and physical map (Figs. 1B and 2).

By pairing the FISH analysis in Fig. 3A and Table 2 withthe data from our physical BAC walk, we determined valuesfor both physical distance in basepairs and cytogenetic dis-tance in micrometres between selected genetic markers. Thegenetic interval spanning markers 63B19 to DK43 coversapproximately 11 µm and an estimated 3 Mb of DNA. Be-tween these genetic markers, three separate measurements ofgenetic to cytogenetic distance yielded similar average val-

ues (i.e., 0.3–0.35 cM�µm–1), suggesting relatively similarrates of recombination throughout the region (assumingequivalent degrees of chromatin condensation). Kulikova etal. (2001) reported a genetic to cytogenetic distance ratio of1.1 cM�µm–1 on LG 4, based on a mapping population of 93individuals. Similarly, recent work combining FISH and ge-netic mapping on other chromosomes in M. truncatula hasdetermined genetic to cytogenetic distances of 0.8 cM�µm–1

for the top of LG 2, 1.8 cM�µm–1 for the center of LG 5, and1.6 cM�µm–1 for the center of LG 8 (Ané et al. 2002). Thevalue of 0.3 cM�µm–1 reported here for the sunn region ofLG 4 is well below this range. Although this could poten-tially be due to differences in the degree of chromatin con-densation between the regions analyzed (Fransz et al. 2000),our data indicate that the explanation is more likely to be alower rate of recombination. We calculated rates of 1000and 1100 kb�cM–1 for the NUM1 and sunn contigs, respec-tively, as compared with rates of 200–350 kb�cM–1 in theabove mentioned regions of LG 2, LG 5, and LG 8 (Ané etal. 2002).

Although much work has been done relating physical togenetic distances in multiple species, studies that examinethe relationship of genetic to cytogenetic distance are lesscommon (de Jong et al. 1999). Using extended-fiber FISHon Arabidopsis, a method that typically produces higher esti-mates of the ratio between genetic and cytogenetic distance,Jackson et al. (1998) obtained values of 1.7 cM per 10.74 ±0.95 µm, effectively 5.7–6.9 cM�µm–1. Analyses of the ratiobetween physical and cytogenetic distances in plants indicatea wide range in chromatin condensation, from 230 to2240 kb�µm–1 in Arabidopsis (Fransz et al. 2000), a 50-foldvariation in tomato (Zhong et al. 1999) and a similar rangein grasses (King et al. 2002). However, none of these authorsreported the correlation between these values and geneticdistance (i.e., recombination frequency).

In addition to the integrated map of M. truncatula(Kulikova et al. 2001), a similar map has been published foranother model legume, L. japonicus (Pedrosa et al. 2002),and for chromosome 10 of rice (Oryza sativa) (Cheng et al.2001). In contrast to our study of the sunn region inM. truncatula, Pedrosa and colleagues did not relate geneticdistance to cytogentic distance, whereas Cheng and col-leagues expressed cytogenetic distances as fractional lengthsof the chromosome.

The combination of genetic, cytogenetic, and physicalmapping techniques has allowed us to define an estimated 3-Mb region around the sunn locus of M. truncatula. The re-sults suggest a relatively constant ratio of genetic tocytogenetic distance between three pairs of genetic markersthat together span the region, but with values that are signifi-cantly lower than those obtained for other regions of theMedicago genome (i.e., Kulikova et al. 2001; Ané et al.2002). This analysis allowed us to construct a ~400 kbcontig, with recombination events occurring within 50 kb ofeach end, delimiting the sunn mutation to a region of ~300–400 kb. Sequencing of this area and cloning of the mutantlocus is currently underway.

Acknowledgements

We would like to thank Dong-Jin Kim and Hongkyu Choi

I:\gen\gen4604\G03-019.vpJuly 8, 2003 8:29:13 AM

Color profile: Generic CMYK printer profileComposite Default screen

© 2003 NRC Canada

Schnabel et al. 671

for sharing molecular markers and physical map data, andBruce Roe for the public sequencing effort (http://www.genome.ou.edu/medicago.html), both of which aided the as-sembly of the sunn contig. We thank Brian Stephens forhelp with DNA preparations of the mapping population. Par-tial support for this work was provided by USDA GrantNo. 2002-35319-12529 and CREES SC-1700150 to J.F., andby the National Science Foundation IBN-9808197 to D.C.T.B. and O.K. are supported by the European Union FifthFramework Program, QLG-CT2000-30676.

References

Ané, J.-M., Lévy, J., Thoquet, P., Kulikova, O., de Billy, F.,Penmetsa, V., Kim, D.-J., Debellé, F., Rosenberg, C., Cook, D.R., Bisseling, T., Huguet, T., and Dénarié J. 2002. Genetic andcytogenetic mapping of DMI1, DMI2, and DMI3 genes ofMedicago truncatula involved in Nod factor transduction,nodulation, and mycorrhization. Mol. Plant–Microbe Interact.15: 1108–1118.

Bell, C.J., Dixon, R.A., Farmer, A.D., Flores, R., Inman, J., Gonza-les, R.A., Harrison, M.J., Paiva, N.L., Scott, A.D., Weller, J.W.,and May, G.D. 2001. The Medicago genome initiative: a modellegume database. Nucleic Acids Res. 29: 114–117.

Bonin, I., Huguet, T., Gheratdi, M., Prosperi, J.M., and Olivieri, I.1996a. High level of polymorphism and spatial structure in aselfing plant species Medicago truncatula Gaertn. using RAPDsmarkers. Am. J. Bot. 83: 843–855.

Bonin, I., Prosperi, J.-M., and Olivieri, I. 1996b. Genetic markersand quantitative genetic variation in Medicago truncatula(Leguminosae): a comparative analysis of population structure.Genetics, 143: 1795–1805.

Catoira, R., Galera, C., de Billy, F., Penmetsa,V., Journet, E.P.,Maillet, F., Rosenberg, C., Cook, D., Gough, C., and Dénarié, J.2000. Four genes of Medicago truncatula controlling compo-nents of a Nod factor transduction pathway. Plant Cell, 12:1647–1665.

Cheng, Z., Presting, G.G., Buell, C.R., Wing, R.A., and Jiang, J.2001. High-resolution pachytene chromosome mapping of bac-terial artificial chromosomes anchored by genetic markers re-veals the centromere location and the distribution of geneticrecombination along chromosome 10 of rice. Genetics, 157:1749–1757.

Cook, D. 1999. Medicago truncatula — a model in the making!Curr. Opin. Plant Biol. 2: 301–304.

Cook, D., VandenBosch, K., Huguet, T., and deBruijn, F.J. 1997.Model legumes get the nod. Plant Cell, 9: 275–281.

Covitz, P.A., Smith, L.S., and Long, S.R. 1998. Expressed se-quence tags from a root-hair-enriched Medicago truncatulacDNA library. Plant Physiol. 117: 1325–1332.

de Jong, J.H., Fransz, P., and Zabel, P. 1999. High resolution FISH inplants-techniques and applications. Trends Plant Sci. 4: 258–263.

Endre, G., Kereszt, A., Kevei, Z., Mihacea, S., Kaló, P., and Kiss,G.B. 2002. A receptor kinase gene regulating symbiotic noduledevelopment. Nature (London), 417: 962–966.

Fransz, P.F., Armstrong, S., de Jong, J.H., Parnell, L.D., vanDrunen, C., Dean, C., Zabel, P., Bisseling, T., and Jones, G.H.2000. Integrated cytogenetic map of chromosome arm 4S ofA. thaliana: structural organization of heterochromatic knob andcentromere region. Cell, 100: 367–376.

Frugoli, J., and Harris, J. 2001. Medicago truncatula on the move.Plant Cell, 13: 458–463.

Györgyey, J., Vaubert, D., Jiménez-Zurdo, J.I., Charon, C.,Troussard, T., Kondorosi, A., and Kondorosi, E. 2000. Analysis

of Medicago truncatula nodule-Expressed Sequence Tags. Mol.Plant–Microbe Interact. 13: 62–71.

Harrison, M.J. 2000. Molecular genetics of model legumes. TrendsPlant Sci. 5: 414–415.

Jackson, S.A., Wang, M.L., Goodman, H.M., and Jiang, J. 1998.Application of fiber-FISH in physical mapping of Arabidopsisthaliana. Genome, 41: 566–572.

Kawaguchi, M., Imaizumu-Anraku, H., Koiwa, H., Niwa, S., Ikuta,A., Syono, K., and Akao, S. 2002. Root, roothair, and symbioticmutants of the model legume Lotus japonicus. Mol. Plant–Mi-crobe Interact. 15: 17–26.

Keen, N., Gelvin, S., and Long, S. 1999. Summary of IS-MPMImeeting, July 1999, Amsterdam. Mol. Plant–Microbe Inter. 12:835–838.

King, J., Armstead, I.P., Donnison, I.S., Thomas, H.M., Jones,R.N., Kearsey, M.J., Roberts, L.A., Thomas, A., Morgan, W.G.,and King, I.P. 2002. Physical and genetic mapping in the grassesLolium perenne and Festuca pratensis. Genetics, 161: 315–324.

Krusell, L., Madsen, L.H., Sato, S., Aubert, G., Genua, A.,Szczyglowski, K., Duc, G., Kaneko, T., Tabata, S., de Bruijn, F.,Pajuelo, E., Sandal, N., and Stougaard, J. 2002. Shoot control ofroot development and nodulation is mediated by a receptor-likekinase. Nature (London), 420: 422–426.

Kulikova, O., Gualtieri, G., Geurts, R., Kim, D.-J., Cook, D.,Huguet, T., de Jong, J.H., Fransz, P.F., and Bisseling, T. 2001.Integration of the FISH pachytene and genetic maps ofMedicago truncatula. Plant J. 27: 49–58.

Lamblin, A.J., Crow, J.A., Johnson, J.E., Silverstein, K.A.T.,Kunau, T.M., Kilian, A., Benz, D., Stromvik, M., Endre, G.,VandenBosch, K.A., Cook, D.R., Young, N.D., and Retzel, E.F.2003. MtDB: a database for personalized data mining of themodel legume Medicago truncatula transcriptome. NucleicAcids Res. 31: 196–201.

Nam, Y.W., Penmesta, R.V., Endre, G., Uribe, P., Kim, D.-J., andCook, D.R. 1999. Construction of a bacterial artificial chromo-some library of Medicago truncatula and identification of clonescontaining ethylene response genes. Theor. Appl. Genet. 98:638–646.

Nishimura, R., Ohmori, M., Fujita, H., and Kawaguchi, M. 2002a.A Lotus basic leucine zipper protein with a RING-finger motifnegatively regulates the developmental program of nodulation.Proc. Natl. Acad. Sci. U.S.A. 99: 15 206 – 15 210.

Nishimura, R., Hayashi, M., Wu, G.J., Kouchi, H., Imaizumi-Anraku, H., Murakami, Y., Kawasaki, S., Akao, S., Ohmori, M.,Nagasawa, M., Harada, K., and Kawaguchi, M. 2002b. HAR1mediates systemic regulation of symbiotic organ development.Nature (London), 420: 426–429.

Oldroyd, G., and Geurts, R. 2001. Medicago truncatula, goingwhere no plant has gone before. Trends Plant Sci. 6: 552–554.

Pedrosa, A., Sandal, N., Stougaard, J., Schweizer, D., andBachmair, A. 2002. Chromosome map of the model legume Lo-tus japonicus. Genetics, 161: 1661–1672.

Penmetsa, R.V., and Cook, D.R. 1997. A legume ethylene-insensitive mutant hyperinfected by its rhizobial symbiont. Sci-ence (Washington, D.C.), 275: 527–530.

Penmetsa, R.V., and Cook, D.R. 2000. Isolation and characteriza-tion of diverse developmental mutants in Medicago truncatula.Plant Physiol. 123: 1387–1397.

Penmetsa, R.V., Frugoli, J.A., Smith, L., Long, S., and Cook, D.R.2003. Genetic evidence for dual pathway control of nodule num-ber in Medicago truncatula. Plant Physiol. 131: 998–1008.

Quackenbush, J., Liang, F., Holt, I., Pertea, G., and Upton, J. 2000.The TIGR gene indices: reconstruction and representation of ex-pressed gene sequences. Nucleic Acids Res. 28: 141–145.

I:\gen\gen4604\G03-019.vpJuly 8, 2003 8:29:13 AM

Color profile: Generic CMYK printer profileComposite Default screen

© 2003 NRC Canada

672 Genome Vol. 46, 2003

Sagan, M., Morandi, D., Tarenghi, E., and Duc, G. 1995. Selectionof nodulation and mycorrhizal mutants in the model plantMedicago truncatula (Gaertn.) after γ-ray mutagenesis. PlantSci. 111: 63–71.

Schauser, L., Roussis, A., Stiller, J., and Stougaard, J. 1999. Aplant regulator controlling development of symbiotic root nod-ules. Nature (London), 402: 191–195.

Stracke, S., Kistner, C., Yoshida, S., Mulder, L., Sato, S., Kaneko,T., Tabata, S., Sandal, N., Stougaard, J., Szczyglowski, K., andParniske, M. 2002. A plant receptor-like kinase required forboth bacterial and fungal symbiosis. Nature (London), 417:959–962.

Wais, R.J., Galera, C., Oldroyd, G., Catoira, R., Penmetsa, R.V.,Cook, D., Gough, C., D.J., and Long, S.R. 2000. Genetic analy-

sis of calcium spiking responses in nodulation mutants ofMedicago truncatula. Proc. Natl. Acad. Sci. U.S.A. 97: 13 407 –13 412.

Walker, S.A., Viprey, V., and Downie, J.A. 2000. Dissection ofnodulation signaling using pea mutants defective for calciumspiking induced by Nod factors and chitin oligomers. Proc. Natl.Acad. Sci. U.S.A. 97: 13 413 – 13 418.

Zhong, X.-B., Bodeau, J., Fransz, P.F., Williamson, V.M., vanKammen, A., de Jong, J.H., and Zabel, P. 1999. FISH to meioticpachytene chromosomes of tomato locates the root-knot nema-tode resistance gene Mi-1 and the acid phosphatase gene Aps-1near the junction of euchromatin and pericentromeric hetero-chromatin of chromosome arms 6S and 6L, respectively. Theor.Appl. Genet. 98: 365–370.

I:\gen\gen4604\G03-019.vpJuly 8, 2003 8:29:13 AM

Color profile: Generic CMYK printer profileComposite Default screen