Diploid/polyploid syntenic shuttle mapping and haplotype-specific chromosome walking

toward a rust resistance gene (Bru1) in highly polyploid sugarcane (2n~12x~115)

Loïc Le Cunff*,1, Olivier Garsmeur*,1, Louis Marie Raboin†, Jérome Pauquet*,§§, Hugues

Telismart†, Atthiappan Selvi*,‡‡, Laurent Grivet*,††, Romain Philippe*, Dilara Begum‡,§,

Monique Deu*, Laurent Costet†, Rod Wing‡,**, Jean Christophe Glaszmann* and Angélique

D’Hont*,2

*. CIRAD, UMR DAP, 34398 Montpellier Cedex 5, France.

†. CIRAD, UMR PVBMT, Pôle de protection des plantes, 97410, Saint-Pierre, Réunion,

France.

‡. Clemson University Genomics Institute, Clemson, South Carolina 29634-5727

§. Present address: Epicentre, Madison, Wisconsin 53713

§§. Present address: Biogemma Mondonville, Domaine de Sandreau 31700 Mondoville,

France

**. Present address: AGI, University of Arizona, Plant Sciences Department, P.O. Tucson,

Arizona 85721-0036

††. Present address: Syngenta Seeds S.A.S., F-31790 Saint-Sauveur, France

‡‡. Sugarcane Breeding Institute, Coimbatore, India

1. These two authors contributed equally to this work.

2. Corresponding author

1

Genetics: Published Articles Ahead of Print, published on August 30, 2008 as 10.1534/genetics.108.091355

Running title: Map-based cloning in a highly polyploid genome

Keywords: Polyploid, map-based cloning, synteny, haplotype specific chromosome walking,

sugarcane, Poacea.

Corresponding author: Angelique D’Hont

CIRAD, UMR DAP, TA A-96/03, Avenue Agropolis, 34398 Montpellier Cedex 5, France.

Tel: +33-4-67615927, Fax: +33-4-67615605

E-mail: dhont@cirad.fr

2

ABSTRACT

The genome of modern sugarcane cultivars is highly polyploid (~12x), aneuploid, of

interspecific origin and contains 10 Gb of DNA. Its size and complexity represent a major

challenge for the isolation of agronomically important genes. Here we report on the first

attempt to isolate a gene from sugarcane by map-based cloning, targeting a durable major rust

resistance gene (Bru1). We describe the genomic strategies that we have developed to

overcome constraints associated with high polyploidy in the successive steps of map-based

cloning approaches, including diploid/polyploid syntenic shuttle mapping with two model

diploid species (sorghum and rice) and haplotype-specific chromosome walking. Their

applications allowed us: i) to develop a high-resolution map including markers at 0.28 and

0.14 cM on both sides and 13 markers cosegregating with Bru1, ii) to develop a physical map

of the target haplotype that still includes two gaps at this stage due to the discovery of an

insertion specific to this haplotype. These approaches will pave the way for the development

of future map-based cloning approaches for sugarcane and other complex polyploid species.

3

INTRODUCTION

Sugarcane (Saccharum spp.) is an important tropical grass crop that accounts for 70%

of raw sugar produced worldwide. It is able to partition carbon to sucrose in the stem, a

vegetative organ, in contrast with other cultivated grasses that usually accumulate their

reserve products within seeds. This unique feature was selected by man who first used its soft

watery culm for chewing. Sugarcane is a C4 photosynthetic plant which, combined with its

perennial nature, has made it one of the most productive cultivated plants. Recently, it has

gained increased attention because it represents an important source of renewable biofuel.

However, sugarcane probably has the most complex of all crop genomes studied to date,

mainly due to its very high degree of polyploidy (~12x) and interspecific origin (D'Hont,

2005). It thus represents a major challenge for genetic studies (Grivet and Arruda, 2002,

D’Hont et al, 2008).

Modern sugarcane cultivars are derived from the combination of the polyploid species

S. officinarum, the domesticated sugar-producing species with 2n=8x=80, and S. spontaneum,

a vigorous wild species with 2n=5x=40 to 2n=16x=128 and many aneuploid forms

(Sreenivasan and Ahloowalia, 1987; D’Hont et al., 1998). Both species are thought to have an

autopolyploid origin (Sreenivasan and Ahloowalia, 1987; Grivet et al., 1996). Prompted by

disease outbreaks, breeders combined both genomes a century ago. The hybrids were

backcrossed to S. officinarum in order to recover the thick sugar-containing stalks of this

species. This process was accelerated through the selection of hybrids derived from 2n

transmission of S. officinarum chromosomes (Bremer, 1961). Modern cultivars are derived

from these interspecific crosses. They are highly polyploid (~12x) and aneuploid, with around

120 chromosomes (review by Sreenivasan and Ahloowalia, 1987). Molecular cytogenetics

(D'Hont et al., 1996; Piperidis and D’Hont, 2001; Cuadrado et al., 2004) and genetic mapping

4

studies (Grivet et al., 1996; Hoarau et al., 2001) showed that modern cultivars typically

display 70 to 80% of chromosomes entirely derived from S. officinarum, 10 to 20% from S.

spontaneum and a few chromosomes derived from interspecific recombinations (Figure 1,

D’Hont et al, 2005).

Brown rust of sugarcane is a fungal disease caused by Puccinia melanocephala, and is

present in most sugarcane growing areas. We previously identified a major gene (Bru1)

conferring resistance to brown rust in the modern cultivar R570 (Daugrois et al., 1996). This

was the first well characterized Mendelian trait described in the complex genomic context of

sugarcane. This source of resistance is of particular interest since it is durable. Indeed, Bru1

resistance breakdown has never been observed despite intensive cultivation of R570 for over

20 years in various regions of the world. Moreover, tests under controlled conditions

demonstrated that this gene provides resistance against diverse rust isolates collected in Africa

and America (Asnaghi et al., 2001). This gene is currently the focus of a map-based cloning

approach.

Map-based cloning is becoming increasingly efficient in model crops such as rice (Sun

et al., 2004, Ueda et al. 2005, Xu et al. 2006) thanks to their simple diploid structure, their

small genome size and tremendous molecular resources. However, it is still a major challenge

in more complex crops with a large genome such as bread wheat (34 Gb/2C) or sugarcane (10

Gb/2C, D’Hont et al 2005). Two mechanisms are responsible for the increase in genome size

and thus complexity, i.e. an increase in monoploid genome size related mainly to transposable

element amplification (Bennetzen, 2005, Piegu et al., 2006), and duplications of the whole

genome, i.e. polyploidization. The large sugarcane genome size is mainly due to a very high

degree of polyploidy (~12x), with the size of its monoploid genome (basic set of

chromosomes (Figure 1), ~900 Mb; D’Hont and Glaszmann, 2001, D’Hont et al, 2008) being

similar to that of sorghum (760 Mb) and only 2-fold the rice genome (390 Mb). As compared

5

to bread wheat, the sugarcane genome is around 3-fold smaller, but its redundancy level is

much higher, with 12 homo(eo)logous highly heterozygous haplotypes on average at each

locus as compared to three sets of highly homozygous pairs of haplotypes in bread wheat

(2n=6x=42, AABBDD). This very high level of genetic redundancy in sugarcane makes it

difficult to monitor specific loci in genetic and physical mapping approaches. In addition, in

modern sugarcane cultivars, chromosomes mainly form bivalents at meiosis (Price, 1963;

Burner and Legendre, 1994), but pairing among the homo(eo)logs appears close to random

with only occasional preferential pairing (Grivet et al., 1996; Hoarau et al., 2001; Jannoo et

al., 2004).

In bread wheat, map-based cloning approaches have been successfully developed

especially through the use of diploid donors (or close diploid relatives) of bread-wheat

subgenomes and the abundant genomic resources from this well studied plant (Keller et al.,

2005). In sugarcane, there are no close diploid relatives, only polyploids are known in the

Saccharum genus. In addition, due to its genetic complexity, this species has received very

little research investment despite its economic importance, and molecular resources have just

recently begun to be developed (Grivet and Arruda, 2002).

Despite this challenging complexity, we implemented a map based cloning approach to

isolate the rust resistance gene Bru1. Bru1 was originally linked to a single marker (CDSR29)

on a map built using selfed cv R570 progeny (Grivet et al., 1996, Daugrois et al., 1996).

Asnaghi et al. (2000) refined the genetic map around Bru1 on the basis of existing rice, maize

and sorghum genetic maps. This approach revealed that the targeted region is orthologous to

one end of sorghum consensus linkage group 4 (LG4), the end of the short arm of rice

chromosome 2 and part of maize LG 4 and LG5. It also enabled localization of Bru1 at the

end of one cosegregation group of the R570 homology group VII. However, it did not enable

marker saturation of the region due to the distal position of the gene and the poor density of

6

markers in the distal orthologous map area of rice and sorghum at that stage. Later, Asnaghi et

al. (2004) identified AFLP markers flanking Bru1 at 2 cM on both sides using a BSA

approach.

In this paper, we describe: i) strategies that we developed to overcome various

constraints associated with high polyploidy in the successive steps of a map-based cloning

approaches, including diploid/polyploid syntenic shuttle mapping with two model diploid

species, sorghum and rice and haplotype-specific chromosome walking, and ii) their

successful application in progressing towards isolation of the rust resistance gene Bru1 in

sugarcane.

MATERIALS AND METHODS

Sorghum genetic mapping

A population of 110 RILs, derived from the Sorghum bicolor ssp bicolor intra-specific

cross IS2807 x 379, was genotyped by RFLP, according to the protocol of Dufour et al.

(1996). The markers were then integrated to the composite map of Boivin et al. (1999) using

Mapmaker 3.0 software (Lander et al. 1987).

Sugarcane genetic mapping

Three sub-populations obtained by selfing the sugarcane cultivar R570 were used.

R570 is a typical modern sugarcane cultivar obtained by the Centre d’Essai de Recherche et

de Formation (CERF) in Réunion. A first population of 88 individuals was used originally to

demonstrate the existence of a major resistance gene to brown rust in R570 (Daugrois et al.,

1996), and to construct an RFLP map (Grivet et al., 1996, Asnaghi et al., 2000). A second

population of 695 individuals was developed to initiate high-resolution mapping of the gene

7

Bru1 (Asnaghi et al., 2004). A total of 26 locally-recombinant individuals that displayed

recombination in a 10 cM region surrounding Bru1 were identified in these two populations

and described by Asnaghi et al. (2004). A third population of 1600 individuals was developed

later to increase the resolution of the genetic map around Bru1.

Rust resistance was scored according to the presence/absence of sporulations.

Individual bearing sporulating pustules were classified as susceptible, or otherwise as

resistant. These field evaluations were performed in Réunion using natural infections as

described in Asnaghi et al 2004 and Tai et al 1981.

RFLP probes were first tested on DNA bulks of five resistant individuals and five

susceptible individuals (5 X 2 µg = 10 µg/lane) digested with HindIII, DraI, SstI or EcorV.

When a polymorphic marker was identified between bulks, a subset of 28 individuals

(resistant and susceptible, 10 µg/lane) was used to validate cosegregation of the marker with

Bru1. Then the marker was analyzed on the 26 locally-recombinant individuals from the

above populations and mapped using Morgan’s mapping function. Genomic DNA extractions

and RFLP were performed as described in Grivet et al. (1994).

AFLP

Four ALFP markers (actctt, aaccac6, actctg9R and acgctt17) surrounding Bru1

(Asnaghi et al., 2004) were used to identify new “local recombinants”. The AFLP procedure

was performed as described by Asnaghi et al. (2004).

AFLP markers cloning

AFLP bands were cut from the polyacrylamide gel and transferred into 10 µl of sterile

water overnight to allow the DNA to diffuse out of the gel. PCR amplifications were

performed in an MJ Research PTC 100 Thermal Cycler in 20 µl reaction mixtures containing

8

1 µl of the AFLP fragment, 50 mM KCl, 10 mM TRIS-HCl (pH 8.3), each primer at 0.2 µM

(Invitrogen AFLP primers which revealed the AFLP markers: EcoRI-aac/MseI-cac for

aaccac6 and EcoRI-att/MseI-cag for attcag), 125 µM of each dNTP (already mixed with the

Invitogen MseI primer) and 1 U of Taq polymerase. The samples were denatured at 94° for 5

min and subjected to 35 cycles at 94° for 30 s, 52° for 45 s, and 72° for 1 min, followed by an

extension step for 8 min at 72°. The PCR products were cloned in a pGEM-T vector

(Promega) and then transformed with DH5 alpha thermo-competent cells. The cloned

fragments were used as RFLP probes on the R570 mapping population to confirm their

cosegregation with the corresponding AFLP marker.

BAC libraries

Four BAC libraries were used in this study, including three constructed at CUGI

(http://www.genome.clemson.edu):

The sorghum BAC library (SB_BBc) constructed with bicolor Btx623 genotype

contains 110592 clones with an average insert size of 120 kb, covering 17-fold the haploid

genome.

The rice BAC library (OSJNBa) constructed with Nipponbare genotype contains a

total of 36864 clones with an average insert size of 130 kb, covering 10-fold the haploid

genome.

The sugarcane BAC library (CUGI, SHCRBa) constructed with R570 cultivar contains

103296 clones with an average insert size of 130 kb representing 1.2 total genome equivalents

(Tomkins et al., 1999, Grivet and Arruda, 2002).

The sorghum and rice BAC libraries have been partially ordered by fingerprinting at

CUGI with HindIII using FPCV6 software (Soderlund et al., 2000) and a cut-off of 1e-13.

BAC contigs are available on-line. (http://www.genome.arizona.edu).

9

We constructed a second sugarcane BAC library (CIR) during this study following the

protocol described by CUGI (http://www.genome.clemson.edu/protocols). This library was

built with the DNA of four selfed R570 individuals carrying two copies of Bru1. A total of

110592 clones were produced with an average insert size of 130 kb. This library was

organized in pools of six BACs and thus fit on one nylon membrane. Screenings of these

BAC libraries were performed by hybridization on high-density-filters using a standard

protocol (http://www.genome.clemson.edu/protocols). For the CIR library, when a pool was

identified, a new screening step was carried out to identify the positive BAC in the positive

pool.

BAC-ends isolation

Isolation of the BAC-ends (terminal sequences) was performed by two different

technique. The first one is based on an adapter-anchor PCR method described by Devic et al.

(1997) and adapted for BAC-ends isolation by Bourgeois E. (personnal communication.).

This technique relies on the use of a blunt-end restriction enzyme, a specific adapter and two

steps of nested PCR amplification using primers specific to the adapter and the BAC vector.

Twenty-five ng of BAC DNA was digested separately with four different enzymes (DraI,

EcorV, StuI and HpaI). The adapter was prepared by annealing the complementarity

oligonucleotides, Adema1 (5’-CACTGAATCTTGCTGACTAGGTCTGGGGAGGT-3’) and

Adema2 (5’-P-ACCTCCCCAGAC-NH2-3’). PCR1 was performed using a specific adaptor

primer, MA1 (5’-CTGAATCTTGCTGACT-3’) and two specific BAC vector border primers,

Lac283 (5’- ACGACGTTGTAAAACGACGGCCAGTGAAT-3’) to amplify the left BAC

terminal sequence or Lac439 (5’-AGCTATGACCATGATTACGCCAAGCTATT-3’) to

amplify the right BAC terminal sequence. PCR1 amplifications were performed in a MJR

thermo-cycler with the following conditions: 3 min at 94° followed by 14 cycles of 94° for

10

30s, 65° to 58° for 45s (the first cycle was 65°, and subsequent cycles were reduced by 0.5°)

then 20 cycles of 94° for 30s, 58° for 45 s, and 72° for 2min and 30 s. PCR2 was performed

using a nested specific adaptor primer, MA2 (5’-ATCTTGCTGACTAGGT-3’) and two

nested BAC vector primers, FBAC for the left BAC terminal sequence (5’-

AGTCGACCTGCAGGCATG-3’) or RBAC for the right BAC terminal sequence (5’-

CGCCAAGCTATTTAGGTGA-3’). PCR2 amplification was performed with 1/50 dilution of

PCR1 product using the same conditions as PCR1, except for the extension time at 72° which

was 1 min and 30 s. The second technique is based on direct sequencing of the BAC-

extremities using specific BAC vector border primers, F-BAC and R-BAC. Analysis of the

sequences obtained allows the definition of specific primer pairs which were used to amplify

the BAC-ends by PCR.

PCR products which obtained by the two different technique were finally loaded on a

1% low melting point agarose gel and after staining with ethidium bromide, the amplification

products were cut directly from the gel to be used as probes.

BAC subclones

BAC DNA (200 ng) was digested with two restriction enzymes (HindIII and EcorV).

DNA fragments generated were cloned into pBluescript2SK+ vector. The subclone sizes were

determined on a 1% agarose gel after PCR amplification with M13 universal primers and

clones with the same sizes were considered as identical. To eliminate subclones

corresponding to repeat sequences or bacterial DNA, a Southern blot was made and

hybridized with R570 total DNA and E. coli DNA. The remaining BAC subclones were

isolated from a 1% low melting point agarose gel to be used as probes.

BAC fingerprinting filters

11

Digested BAC DNA (500ng) were separated in a 0.7% agarose gel at 70 V for 24 h

and transferred on nylon filters.

RESULTS

- Diploid/polyploid syntenic shuttle high-resolution mapping using sorghum as a model

species

To refine the location of the sorghum region orthologous to the R570 target region, we

used our sorghum mapping population and, in a second step, information from other sorghum

genetic maps. The two closest AFLP markers (aaccac6 and attcag) distally flanking Bru1

(Figure 2, Step 1), from Asnaghi et al. (2004), were cloned and analyzed by RFLP on our

sorghum mapping population. The aaccac6 locus was mapped on LG4 (previously LG D,

Boivin et al., 1999) at 9.4 cM proximally from CDSR29. The attcag probe revealed no

polymorphism in our sorghum mapping population. However, its hybridization on the

sorghum BAC library (Sb_BBc) identified several BACs including BAC 67N07. One end of

this BAC, F67N07D, was mapped by RFLP on LG 4 at 8.8 cM proximally from CDSR29.

These results allowed us to delimit the sorghum region orthologous to the target sugarcane

region to an 8.8 cM interval between CDSR29 and F67N07D (Figure 2). They also revealed

a local inversion on the R570 genetic map as compared to the sorghum map. Indeed, although

LG VII of the R570 sugarcane genetic map is syntenic and mostly colinear to sorghum LG4

(Asnaghi et al., 2000), the two AFLP markers located distally from CDSR29 in CG VII.1 of

R570 were mapped proximally from CDSR29 on sorghum LG4.

We aligned our sorghum LG4 (previously LG D, Boivin et al., 1999, and unpublished

data) with LG4 (previously LGF) of the high density RFLP map of Bowers et al. (2003). Two

common RFLP loci (CDSR29 and M738) allowed us to delimit the target region on the map

12

of Bowers et al. (2003) (Figure 2, Step 1). Nine RFLP loci located in the target sorghum

region over the two maps at that stage were analyzed on the R570 progeny (Figure 2, Step 1).

Four RFLP probes (pSB1445, cMWG652, pSB0084 and pPAP05H10) revealed a

polymorphic marker between the resistant and susceptible bulks. The four markers were

mapped in the target area with the two closest to Bru1 being revealed by probes pSB0084 and

pPAP05H10 and mapped at 0.6 cM proximally and 0.3 cM distally from Bru1, respectively

(Figure 2, Step 1). These results reduced the target region in R570 to a 0.9 cM interval.

These two loci are separated by 3.1 cM on the sorghum maps of Bowers et al. (2003). The

position of the mapped markers confirmed the inversion between sugarcane R570 CGVII.1

and sorghum and was in agreement with colinearity inside the inversion.

To further refine the genetic map of the Bru1 region, we built an orthologous sorghum

BAC physical map and then derived new probes from this physical map (Figure 2, Step 2),.

Four loci (pSB0084, pSB1565, pSB0927 and pPAP05H10) located in the 3.1 cM orthologous

target region in sorghum were used to screen a partially contiged sorghum BAC library

(Sb_BBc, http://www.genome.arizona.edu/genome/sorghum.html). The identified BACs were

searched for on the on-line sorghum physical map

(http://www.stardaddy.uga.edu/fpc/bicolor/WebAGCoL/WebFPC) to find the corresponding

contig. Together, these four probes allowed us to identify four contigs (named at that stage

Ctg1123, Ctg458, Ctg1363 and Ctg1455). To test if the four identified contigs overlapped: i)

we used the “contig–end best match” option, available on FPCV6, on CUGI BAC fingerprint

data, ii) we compared the HindIII fingerprints of some key BACs from the sorghum four

contigs, and iii) we sub-cloned BACs-ends from the contig extremities and hybridized them

on the sorghum BAC library. Altogether, the results obtained demonstrated that the four

contigs overlapped and thus could be merged into a unique contig of 91 BACs representing a

region of around 350-400 kb (Figure 2, Step 2).

13

Fifty-seven BAC-ends spread over the sorghum contig were cloned and analyzed.

Among them, 14% did not hybridize well with sugarcane DNA and around 56% appeared to

correspond to repeated sequences in sugarcane. Among the remaining 30% (17 clones), four

revealed a polymorphic marker between resistant and susceptible bulks from the self-progeny

of R570. They were all mapped in the target area in R570. Two markers, revealed by the

BAC-ends R24P17eV and F57P2eV, mapped at 0.3 cM proximally to Bru1; one marker,

revealed by the BAC-end F123K21D, cosegregated with Bru1 and one marker, revealed by

the BAC-end R195K15H, mapped at 0.3 cM distally to Bru1 (Figure 2, Step 2). The

sorghum region orthologous to the sugarcane target region (between F57P2eV and

R195K15H) was reduced to approximately 225 kb. Again, the markers position confirmed the

inversion between sorghum and R570 CG VII.1 and was in agreement with colinearity inside

the inversion.

Diploid/polyploid syntenic shuttle high-resolution mapping using rice as a model species

To identify the rice region orthologous to the R570 target region, probes revealing a

marker surrounding Bru1 were tested for hybridization with rice DNA and used to screen the

OSJNBa rice BAC library (Figure 2, Step 3).. The rice probe C673 proximal to Bru1

hybridized to a BAC contig that, according to the on-line rice physical map

(www.genome.arizona.edu), corresponded to rice chromosome 2. The probe pPAP05H10

distal to Bru1 identified BAC clones belonging to the same contig. Their positions revealed

that the orthologous target region in rice correspond to around 600 kb on the short arm of

chromosome 2 (Figure 2, Step 3). According to the orientation of the rice physical map, the

target region in the R570 sugarcane cultivar is inversed as compared to rice. This inversion,

that we also noticed between the target region in R570 and sorghum, must have arisen in

sugarcane after it diverged from sorghum.

14

The sequence of five BAC clones covering most of the rice region orthologous to the

target sugarcane region was available at this stage (AP005304, AP005311, AP004885,

AP004078 and AP004121). We compared these sequences with the SUCEST sugarcane EST

database (238000 sugarcane ESTs assembled in 43000 clusters, Vettore et al. 2003). Ninety

EST clusters displaying at least 80% homology on at least 100 bp with the rice BAC

sequences were identified. cDNA corresponding to 34 of the sugarcane EST clusters were

used as probes on bulks of DNA from susceptible and resistant R570 progeny, and seven

revealed a polymorphic marker. They were all mapped within the sugarcane target area, with

the closest to Bru1 being cBR32 and cBR33 that cosegregated with Bru1 and cBR23 and

cBR37 that surrounded Bru1 at 0.3 cM on both sides (Figure 2, Step 3). These latter two

belonged to BAC AP004885 and were separated by around 100 kb. The marker positions

confirmed the inversion between rice and R570 CG VII.1 and were in agreement with

colinearity inside the inversion. BLAST of the seven mapped EST clusters did not reveal

significant homology to any cloned resistance gene or resistant gene analogue (RGA).

High resolution genetic mapping in sugarcane

A new population of 1600 individuals was developed to increase the resolution of the

genetic map around Bru1. The 400 susceptible individuals identified in this population were

analyzed with four AFLP markers surrounding the gene (actctt, aaccac6, actctg9R, acgctt17

mapped at 3.5 and 2.5 cM distally and at 2.2 and 3.1 cM proximally from Bru1, respectively)

to identify individuals locally-recombinant in the target area. We used only susceptible

individuals because the susceptible phenotype is easier to establish with certainty and since

susceptible individuals represent a quarter of the population but encompass half of the

detectable recombinant individuals in the region. Among the identified locally-recombinant

individuals, 35 displayed recombination between AFLP markers aaccac6 and actctg9R, in

15

accordance with the previously observed genetic distance between these markers (4.7 cM).

The nine markers closest to Bru1 were analysed on these individuals. Two new

recombinations were identified in the 0.6 cM target area but they did not allow separation of

previous cosegregating markers. The new high-resolution map of the target region included

three markers mapped at 0.28 cM on both sides of Bru1 (F57P2eV, cBR37 and R24P17eV

proximally; pPAP05H10, R195K15H and cBR23 distally) and three markers cosegregating

with Bru1 (cBR32, cBR33 and F123K21D) (Figure 2).

Physical mapping of the target locus in sugarcane

We screened the whole R570 BAC library with nine probes mapped in a 0.70 cM

region around Bru1 (between loci pPAP05H10 and PSB0084). We also used fourteen probes

that were not mapped but corresponded to six sorghum BAC-ends from the sorghum

orthologous BAC contig, and eight sugarcane cDNAs displaying homology to the orthologous

rice sequence. To build the physical map, we used BACs that were detected by at least two

probes, including one probe that revealed a marker linked to Bru1 or one probe associated in

another BAC to a probe that revealed a marker linked to Bru1. We were thus able to build a

physical map of 32 BACs covering the Bru1 region (Figure 3). The physical map, at that

stage, contained one gap (between loci cBR32 and R195K15H). To complete this map, 10

BAC-ends from some of the 32 sugarcane BAC clones were isolated and analyzed on bulks of

DNA of susceptible and resistant R570 progeny. One BAC-end (R15N23S) revealed a

polymorphic marker between the resistant and susceptible bulks and was mapped at 0.14 cM

on the distal side of Bru1. This allowed us to reduce the target area between probe cBR37

(0.28 cM proximally) and R15N23S (0.14 cM distally) and to eliminate the gap from the

target area. The BAC-end R15N23S was used to screen the R570 BAC library and allowed us

to identify an additional BAC (164H22) that partially covered the target area.

16

Finally, at this stage, the entire target region (between probe/locus cBR37 and

R15N23) was contained in three R570 BAC clones (15N23, 25N07 and 253G12 presenting a

size of 120, 90 and 135 kb, respectively) and seven R570 BAC clones contained part of the

target area (Figure 3).

Identification of the target haplotype

As sugarcane is polyploid and heterozygous, the obtained physical map did not

correspond to a unique BAC contig but to several hom(oe)ologous BAC contigs

corresponding to hom(oe)ologous chromosomes. To differentiate the haplotypes, we

compared the restriction profiles of most of the BAC clones of the studied region, including

all those belonging to the target region (between probes/loci cBR37 and R15N23S). Four

hom(oe)ologous haplotypes (a to d) were identified for BACs located on the proximal side of

the target area. Five hom(oe)ologous haplotypes (α to ε) were identified for BACs located on

the distal side of the target area. Seven hom(oe)ologous haplotypes (1 to 7) were identified

among the 10 BAC clones covering the target area (Figure 3).

To identify the haplotype bearing Bru1 (target haplotype), we had to determine which

BAC clones were bearing the markers (alleles) linked to Bru1. We thus analyzed the BAC

clones by RFLP with the probe/enzyme combinations that revealed the markers linked to

Bru1 and compared their RFLP profiles with that of R570 (Figure 4). The RFLP patterns of

BAC clones corresponding to the haplotype bearing Bru1 should include the bands/markers

linked to Bru1. Using this procedure, we were able to identify only one BAC clone

corresponding to the target haplotype, BAC 164H22. This BAC clone hybridized only with

marker R15N23S and thus only partially covered the target (Figure 3).

Haplotype-specific chromosome walking using a Bru1-enriched BAC library

17

In order to complete the physical map of the target haplotype, a new BAC library of

110 592 clones with an average insert size of 130 kb was built. We used a mix of DNA from

four resistant selfed R570 progeny that contained two copies of Bru1 with the aim of

increasing the proportion of the target haplotype in this library. These four individuals were

selected according to the band intensity observed on the RFLP profiles for four markers

surrounding Bru1 (F57P2eV, F123K21D, pPAP05H10 and R195K15H), the band intensity

reflecting to the number of marker doses. The new library, that covers 2.8-fold the target

haplotype and 1.4-fold the total genome, was screened with five probes (cBR37, F123K21D,

cBR33, cBR32 and R15N23) mapped in a 0.42 cM region around Bru1. Only BACs

belonging to the target haplotype, identified as explained above, were selected. Two new

BACs were recovered, i.e. BAC CIR9O20 (90 kb) that hybridized with probes cBR37,

F123K21D, cBR33, and BAC CIR12E03 (125 kb) that hybridized with probe cBR32. We

designed primers at the extremities of these two BACs and tested them against each other and

BAC 164H22. This revealed that two gaps still remained in the physical map, i.e. between

BAC CIR9O20 and CIR12E03 and between CIR12E03 and 164H22 (Figure 3).

BAC-ends and BAC-subclones from the three BACs from the target haplotype

(CIR9O20, CIR12E03 and 164H22) were isolated and a chromosome walk was undertaken in

order to fill the two gaps remaining on the target haplotype. Due to polyploidy, only mapped

probes can be used for this chromosome walking step to ensure that this walking is actually

on the target chromosome. We thus first had to map the sub-clones and select the ones

revealing a marker linked to Bru1. Nine such sub-clones (R-CIR9O20, R-CIR12E03,

CIR9O20-D10, CIR9O20-F4, CIR12E03-A5, CIR12E03-B6, CIR12E03-D7, CIR12E03-D10,

and CIR12E03-H6) were obtained, they all cosegregated with Bru1. They were used to screen

the BAC libraries and revealed three new BAC clones from the target haplotype (CIR43P06,

22M06 (85 kb) and CIR12H16). Analysis of these BAC clones showed that CIR12H16 was

18

included in BAC CIR12E03 and that CIR43P06 and 22M06 partially overlapped with BAC

CIR12E03 but not with CIR9O20 or 164H22 (Figure 3). BAC-end sequences of CIR43P06

and 22M06 corresponded to repeated sequences. Both BAC-ends of CIR12H16 and one sub-

clone of 22M06 (22M06-H5) were mapped and cosegregated with Bru1. However, they did

not reveal any new BAC of the target haplotype when used to screen the BAC libraries.

Finally, after all these chromosome walking steps, the physical map of the target

haplotype contained six BACs but two gaps remained. The genetic map included 15 markers

that cosegregated with Bru1 (Figure 3).

All sub-clones that revealed a marker linked to Bru1 were sequenced. Interestingly,

subclone CIR9O20-F4 showed homologies with the X and XI domains of Rpg1 (Brueggeman

et al., 2002), a barley rust resistance protein that corresponds to a Serine/Threonine Kinase

and subclone CIR12E03-A5 displayed homology with the VIII, IX and X domains of a rice

S/T kinase (9639.t0149 on TIGR database) and Rpg1.

Identification of an insertion—impact on recombination

During the BAC library screening steps with the BAC subclones, we noticed that nine

of the subclones (R-CIR9O20, R-CIR12E03, CIR12E03-D10, R-CIR12H16, CIR12E03-A5,

22M06-H5, CIR12E03-B6, CIR12E03-H6 and CIR12E03-D7), that revealed a marker

cosegregating with Bru1 and thus belong to the target haplotype, did not hybridize with the

other hom(oe)ologous BAC clones (Figure 3). This highlighted the presence of an insertion

specific to the target haplotype. This inserted segment corresponds to part of BAC CIR9O20

and BAC CIR12E03 and thus comprises one of the two gaps remaining on the target

haplotype contig.

To assess the impact of this insertion on recombination, we attempted to map a few

probes/loci surrounding the insertion on the other hom(oe)ologous cosegregation groups. Five

19

probes were analysed, including three probes mapped distally from Bru1 (cBR8, cBR20 and

R15N23S), one mapped proximally from Bru1 (cBR56) and one that genetically cosegregated

with Bru1 (F123K21D) but physically mapped proximally from the insertion (Figure 5). For

the mapping, we used a subset of 112 individuals from the selfed R570 progeny. Four

markers (revealed by probes cBR8, cBR20, R15N23S and F123K21D) were mapped on CG

VII-b, two markers (revealed by probes cBR20 and F123K21D) were mapped on CG VII-c

and two markers (revealed by probes cBR8 and cBR56) were mapped on VII-d. The genetic

distance between a given pair of markers varied between the hom(oe)ologous cosegregation

group. For all the intervals comprising the insertion, the distance was lower for the target

cosegregation group (CG VII-1) as compared to the other hom(oe)ologous cosegregation

group (Figure 5). This showed that the insertion induced a reduction of recombination on the

target haplotype.

DISCUSSION

In this paper, we report on the first map-based cloning attempt in highly polyploid

sugarcane.

Diploid/polyploid syntenic shuttle mapping facilitated the development of a high density

genetic map around Bru1

Modern sugarcane cultivars have a large genome with an estimated complete genetic

map size of 17000 cM (Hoarau et al., 2001) and up to 12 alleles on average that can coexist at

a given locus. This implies that only single-dose markers (ie markers present on only one of

the hom(oe)ologous haplotypes) can be used for high-resolution mapping. Our strategy, to

help rapidly saturate a given target area, was to use bulk segregant analysis while also

20

benefiting from the good syntenic relationships between sugarcane and two model diploid

Poaceae species, i.e. sorghum and rice (Grivet 1994, Dufour et al. 1997, Glaszmann 1997,

Guimares et al. 1997, Ming et al. 1998, Jannoo et al 2007), that are estimated to have

diverged from sugarcane around 8 and 50 Mya, respectively (Jannoo et al., 2007, Wolfe et al.,

1989). Genetic maps, physical maps and/or genome sequence data from these two species

were tapped to select loci potentially present in the target area. The derived RFLP probes

were analyzed on bulks of DNA from resistant and susceptible selfed R570 progeny. This

allowed us in one step to select single-dose markers that were linked to Bru1 and thus to

eliminate multiple dose markers and markers not linked to Bru1.

We fully tapped the various genomic resources of the three species in order to develop a

high-resolution genetic map around Bru1. The origin and type of probes tested during these

steps gave contrasting results. Around 44% of the probes derived from sorghum genetic maps

revealed a marker linked to Bru1, as compared to only 7% of the BAC-ends tested from the

sorghum orthologous physical map and to 20% of the sugarcane cDNAs selected for their

homology to the rice orthologous sequence. This was due to the fact that the probes derived

from the sorghum genetic map were already selected as corresponding to single or low copy

loci in sorghum. Most of them were from Gramineae species other than sorghum, and were

thus already selected for cross-hybridization with other species. Sugarcane cDNAs selected

for their homology to the rice orthologous sequence also generally corresponded to conserved

single or low copy loci, but conservation of synteny between rice and sugarcane is lower than

between sorghum and sugarcane (Glaszmann et al., 1997). By contrast, we estimated that

around 56% of the sorghum BAC-ends tested corresponded to high or moderate repeated

sequences in sugarcane. This percentage is approximate because, due to polyploidy, the RFLP

profiles of single copy loci in sugarcane are typically multiple-banded and cannot be clearly

21

separated from moderately repeated sequences. In addition, 14% of the sorghum BAC-ends

did not hybridize well with sugarcane DNA.

The combined use of synteny relationships between sugarcane, sorghum and rice and

BSA was very successful since it allowed us to reduce the size of the genetic interval

containing the Bru1 gene from 4 cM to 0.56 cM and to identify nine markers within this

genetic interval. These nine markers allowed us to initiate the sugarcane physical map. These

results illustrate the value of exploiting model diploid species for mapping complex genomes.

The efficiency of this study, in contrast to that of Asnaghi et al. (2000), could mainly be

explained by the fact that we were able to refine the location of the orthologous sorghum and

rice region and access newly available molecular resources, in particular high density

sorghum RFLP genetic maps, a partially ordered sorghum BAC library and the rice genome

sequence. The availability of the complete sorghum genome sequence

(http://www.phytozome.net/sorghum) should greatly accelerate the high-resolution mapping

steps in future studies of this type in sugarcane.

Haplotype-specific chromosome walking

To develop a physical map, again since sugarcane is polyploid and highly heterozygous,

we had to develop specific methodologies. To identify among hom(oe)ologous BAC clones

the ones corresponding to the haplotype bearing Bru1, we had to determine which ones

encompass the markers (alleles) linked to Bru1. This meant that to develop the physical map,

and for further steps of chromosome walking toward the gene, only probes revealing a

markers linked to Bru1 could be used. Therefore physical and genetic mapping had to be

tightly associated throughout physical mapping process.

22

Another difficulty, due to the large size of the sugarcane genome, was that to obtain a

BAC library with reasonable genome coverage, a very large number of BAC clones is

required. For example, although the R570 BAC library (Tomkins et al., 1999) that we first

used includes as many as 103296 BAC clones with an average size of 130 kb, this

corresponded to a coverage of only 1.3-fold the total genome, and thus to a probability of only

73.9% of finding any particular DNA segment. By comparison, the sorghum library used in

this study includes a similar number of BAC clones (110592), with a similar average size but,

since sorghum is diploid and homozygous, this represents a coverage of 17-fold the genome

and a 99.9% probability of finding a given segment. One way to overcome this problem is to

build a much larger library, which then becomes very tedious to manipulate. We thus

developed an alternative strategy that involved building a library with a genotype comprising

more copies of the target haplotype. This material allowed us to specifically double the

proportion of the target haplotype in the new BAC library (as compared to the R750 library).

Genotypes with even more copies of the target haplotype could be used in order to continue

increasing the proportion of the target haplotype. For example, it could be possible to again

self the selfed R570 progeny and select individuals with four copies of the target haplotype.

To accelerate screening of the new library, BAC clones were picked and pooled by six in each

well. The new BAC library represents 1.4 genome equivalents but statistically contains 2.8-

fold the target area. Added to the 1.3 genome coverage of the R570 BAC library of Tomkins

et al. (1999), we now have BAC libraries covering 4-fold (4x) the target area overall, with a

98.5% probability of finding the target segment.

Using these strategies, we were able to build a physical map that encompasses seven

hom(oe)ologous BAC contigs, including three hom(oe)ologous BAC clones that entirely

cover the target region. In addition, we reduced the target interval to 0.42 cM and identified

12 additional markers cosegregating with Bru1.

23

Structure of the target haplotype

We showed that part of the target haplotype corresponded to an inserted chromosome

segment with no homology to the homo(eo)logous haplotypes. This type of discovery has

already complicated several map-based cloning projects in diploid species (Stirling et al.,

2001, Barker et al., 2005). This is due to the fact that these insertions generally induce severe

repression of recombination (Neu et al., 2002, Barker et al., 2005 Wei et al., 1999, Stirling et

al., 2001), as we observed in the Bru1 region. These insertions may result from genomic

rearrangement or, as frequently observed for resistance genes, to introgression of resistance

from an alien source. Hoarau et al. (2001) suggested that the chromosome carrying Bru1

originated from S. officinarum, a species that represents the main component of modern

cultivar genomes. We identified, in the inserted segment, two subclones (12E3-D07 and

22M06-H05) that revealed an atypical sugarcane RFLP profile with only one band (that

cosegregating with Bru1). This suggested that the inserted chromosome segment might

originate from S. spontaneum, the minority part of the modern sugarcane cultivar genome.

This would be in line with the fact that the interspecific crosses from which all modern

cultivars are derived were performed 100 years ago to overcome disease outbreaks in S.

officinarum by utilizing resistance from the wild S. spontaneum species. Recently, Raboin et

al. (2006) identified a second major brown rust resistance gene that is also suspected to

originate from S. spontaneum and mapped on a linkage group non-orthologous to R570

linkage bearing Bru1 (Raboin et al., 2006).

Two BAC sub-clones from the target haplotype showed homology with the S/T kinase

barley rust resistance gene Rpg1. This suggests the possible presence of an S/T kinase cluster

in the target region. These clones are being further investigated as they represent potential

candidate genes for Bru1. In sorghum, rust resistance loci have been localised (Tao et al.,

24

1998), including a major rust resistance QTL, genetically associated with a homolog of the

maize rust resistance gene Rp1-D (McIntyre et al. 2004). In addition, McIntyre et al. (2005)

showed that some RGAs can be found in a syntenic position in sugarcane and sorghum.

According to available databases and reports, no resistance genes or QTLs have been

identified in the Bru1 orthologous rice and sorghum region. Note however, that the colinearity

between sorghum, rice and sugarcane in the target region has been upset by the insertion

identified in the target haplotype.

We are currently sequencing ten sugarcane BAC clones representing the different

haplotypes that are hom(oe)ologous or cover the Bru1 haplotype. These sequences will be

mined to complete the cloning of Bru1, especially: i) to characterize the two S/T kinase

candidate sequences identified in this study and identify additional candidates genes, ii) to

localize the sorghum region orthologous to the sugarcane target haplotype-inserted-

chromosome segments, and iii) to develop new probes to complete the physical map of the

target haplotype.

In addition, we are currently characterizing a population of sugarcane cultivars with

different rust resistance phenotypes with the markers cosegregating with Bru1 to try

identifying recombinant haplotypes to reduce the size of the target region. Linkage

disequilibrium was evaluated in sugarcane cultivars and varies from 0 to 30cM with a sharp

decrease after 5 cM (Jannoo et al. 1999, Raboin et al 2008).

25

ACKNOWLEDGEMENTS

We thank the International Consortium for Sugarcane Biotechnology (ICSB) for

financial support for this work. We also thank P. Arruda for providing access to the SUCEST

sugarcane EST and cDNA resources, and N. Yahiaoui and Rob Miller for critical reading of

the manuscript.

26

LITERATURE CITED

Asnaghi, C., D. Roques, S. Ruffel, C. Kaye, J.Y. Hoarau, H. Télismart, J.C. Girard, L.M.

Raboin, A.M. Risterucci, L. Grivet, A. D’Hont, 2004 Targeted mapping of a sugarcane rust

resistance gene Bru1 using bulked segregant analysis and AFLP markers. Theor. Appl.

Genet., 108, 759-764.

Asnaghi, C., A. D'Hont, J.C. Glaszmann, P. Rott, 2001 Resistance of sugarcane cultivar R570

to Puccinia melanocephala isolates from different geographic locations. Plant. Dis., 85, 282-

286.

Asnaghi, C., F. Paulet, C. Kaye, L. Grivet, J.C. Glaszmann, A. D’Hont, 2000 Application of

synteny across the Poaceae to determine the map location of a rust resistance gene of

sugarcane. Theor. Appl. Genet., 10, 962-969.

Barker, C.L., T. Donald, J. Pauquet, M.B. Ratnaparkhe, A. Bouquet, A.F. Adam-Blondon,

M.R. Thomas, I. Dry, 2005 Genetic and physical mapping of the grapevine powdery mildew

resistance gene, Run1, using a bacterial artificial chromosome library. Theor. Appl. Genet.,

111, 370-7.

Bennetzen, J.L. 2005 Transposable elements, gene creation and genome rearrangement in

flowering plants. Curr. Opin. Genet. Dev., 15, 621-7.

Boivin, K., M. Deu, J.F. Rami, G. Trouche, P. Hamon, 1999 Towards a saturated sorghum

map using RFLP and AFLP markers. Theor. Appl. Genet., 98, 320-328.

27

Bowers, J.E., C. Abbey, S. Anderson, C. Chang, X. Draye, A.H. Hoppe, R. Jessup, C. Lemke,

J. Lennington, Z. Li, Y.R. Lin, S.C. Liu, L. Luo, B.S. Marler, R. Ming, S.E. Mitchell, D.

Qiang, K. Reischmann, S.R. Schulze, D.N. Skinner, Y.W. Wang, S. Kresovich, K.F. Schertz,

A.H. Paterson, 2003 A high-density genetic recombination map of sequence-tagged sites for

sorghum, as a framework for comparative structural and evolutionary genomics of tropical

grains and grasses. Genetics, 165, 367-86.

Bremer, G. 1961. Problems in breeding and cytology of sugarcane. I. A short history of

sugarcane breeding - the original forms of Saccharum. Euphytica, 10, 59-78.

Brueggeman, R., N. Rostoks, D. Kudrna, A. Kilian, F. Han, J. Chen, A. Druka, B. Steffenson,

A. Kleinhofs, 2002 The barley stem rust-resistance gene Rpg1 is a novel disease-resistance

gene with homology to receptor kinases. Proc. Natl. Acad. Sci., U S A, 99, 9328-33.

Burner, D.M. and B.L. Legendre, 1994 Cytogenetic and fertility characteristics of elite

sugarcane clones. Sugar Cane, 1, 6-10.

Cuadrado, A., R. Acevedo, S. Moreno Diaz de la Espina, N. Jouve, C. de la Torre, 2004

Genome remodelling in three modern S. officinarum X S. spontaneum sugarcane cultivars. J.

Exp. Bot., 55, 847-854.

Daugrois, J.H., L. Grivet, D. Roques, J.Y Hoarau, H. Lombard, J.C Glaszmann, A. D’Hont,

1996 A putative major gene for rust resistance linked with an RFLP marker in sugarcane

cultivar R570. Theor. Appl. Genet., 92, 1059–1064.

28

Devic, M., S. Albert, M. Delseny, T.J. Roscoe, 1997 Efficient PCR walking on plant genomic

DNA. Plant Physiol. Biochem., 35, 331–339.

D'Hont A., Souza G.M., Menossi M., Vincentz M, Marie-Anne Van-Sluys A-M., Glaszmann J.C., and Ulian E. Sugarcane: a Major Source of Sweetness, Alcohol, and Bio-energy. In: Moore, PH and Ming R (eds.) Tropical Crop Plant Genomics. Springer, N.Y. (2008).

D’Hont, A. 2005 Unraveling the genome structure of polyploids using FISH and GISH;

examples of sugarcane and banana. Cytogenet. Genome Res., 109, 27-33. Review.

D’Hont, A., J.C. Glaszmann, 2001 Sugarcane genome analysis with molecular markers: a first

decade of research. Proc. Int. Soc. Technol., 24, 556-559.

D’Hont, A., D. Ison, K. Alix, C. Roux, J.C. Glaszmann, 1998 Determination of basic

chromosome numbers in the genus Saccharum by physical mapping of ribosomal RNA genes.

Genome, 41, 221-225.

D’Hont, A., L. Grivet, P. Feldmann, S. Rao, N. Berding, J.C. Glaszmann, 1996

Characterisation of the double genome structure of modern sugarcane cultivars Saccharum

spp by molecular cytogenetics. Mol. Gen. Genet., 250, 405–413.

Dufour, P., M. Deu, L. Grivet, A. D'Hont, F. Paulet, A. Bouet, C. Lanaud, J.C. Glaszmann, P.

Hamon, 1997 Construction of a composite sorghum genome map and comparison with

sugarcane, a related complex polyploidy. Theor. Appl. Genet., 94, 409-418.

29

Dufour, P., L. Grivet, A. D’Hont, M. Deu, G. Trouche, J.C. Glaszmann, P. Hamon, 1996

Comparative genetic mapping between duplicated segments on maize chromosomes 3 and 8

and homeologous regions in sorghum and sugarcane. Theor. Appl. Genet., 92, 1024-1030.

Glaszmann, J.C., P. Dufour, L. Grivet, A. D’Hont, M. Deu, F. Paulet, P. Hamon, 1997

Comparative genome analysis between several tropical grasses. Euphytica, 96, 13-21.

Grivet, L., P. Arruda, 2002 Sugarcane genomics: depicting the complex genome of an

important tropical crop. Curr. Opin. Plant Biol., 5, 122-7.

Grivet, L., A. D’Hont, D. Roques, P. Feldmann, C. Lanaud, J.C. Glaszmann, 1996 RFLP

mapping in cultivated sugarcane Saccharum spp: genome organization in a highly polyploid

and aneuploid interspecific hybrid. Genetics, 142, 987-1000.

Grivet, L., A. D’Hont, P. Dufour, P. Hamon, D. Roques, J.C. Glaszmann, 1994 Comparative

genome mapping of sugarcane with other species within the Andropogoneae tribe. Heredity,

73, 500-508.

Guimaraes, C.T., G.R. Sillis, B.W. Sobral, 1997 Comparative mapping of Andropogoneae :

Saccharum L. sugarcane ans its relation to sorghum and maize. Proc. Natl. Acad. Sci. USA,

94, 14261-14266.

Hoarau, J.Y., B. Offmann, A. D’Hont, A.M. Risterucci, D. Roques, J.C. Glaszmann, L.

Grivet, 2001 Genetic dissection of a modern sugarcane cultivar Saccharum spp I Genome

mapping with AFLP markers. Theor. Appl. Genet., 103, 84–97.

30

Jannoo, N., L. Grivet, N. Chantret, O. Garsmeur, J.C. Glaszmann, P. Arruda, A. D’Hont, 2007

Orthologous comparison in a gene-rich region among grasses reveals stability in the

sugarcane polyploid genome. Plant J., 50, 574-85.

Jannoo, N., L. Grivet, J. David, A. D’Hont, J.C. Glaszmann, 2004 Differential chromosome

pairing affinities at meiosis in polyploidy sugarcane revealed by molecular markers. Heredity,

93, 460-467.

Jannoo, N., L. Grivet, L., A. Dookun, A., D’Hont, J.C. Glaszmann, 1999 Linkage

disequilibrium among modern sugarcane cultivars. Theor. Appl. Genet., 99, 1053-1060.

Keller, B., C. Feuillet, N. Yahiaoui, 2005 Map-based isolation of disease resistance genes

from bread wheat: cloning in a supersize genome. Genet. Res., 85, 93-100. Review

Lander, E.S., P. Green, J. Abrahamson, A., Barlow, M.J. Daly, S.E. Linclon, L. Newburg,

1987 MAPMAKER, an interactive computer package for constructing primary genetic

linkage maps of experimental and natural populations. Genomics, 1, 174-181.

McIntyre, C.L., S.M. Hermann, R.E. Casu, D. Knight, D., J. Drenth, Y. Tao, S.M. Brumbley,

I.D. Godwin, S. Williams, G.R. Smith, J.M. Manners 2004 Homologues of the maize rust

resistance gene Rp1-D are genetically associated with a major rust resistance QTL in

sorghum. Theor. Appl. Genet., 109, 875-883.

31

Ming, R., S.C Liu, Y.R. Lin, J. Da Silva, W. Wilson, D. Braga, A. Van Deynze, T.F.

Wenslaff, K.K. Wu, P.H. Moore, W. Burnquist, M.E. Sorrells, J.E. Irvine, A.H. Paterson,

1998 Detailed alignment of Saccharum and Sorghum chromosomes: comparative

organization of closely related diploid and polyploid genomes. Genetics, 150, 1663–1682.

Neu, C., N. Stein, B. Keller, 2002 Genetic mapping of the Lr20-Pm1 resistance locus reveals

suppressed recombination on chromosome arm 7AL in hexaploid wheat. Genome, 45, 737-44.

Piegu, B., R. Guyot, N. Picault, A. Roulin, A. Saniyal, H. Kim, K. Collura, D.S. Brar, S.

Jackson, R.A. Wing, O. Panaud, 2006 Doubling genome size without polyploidization:

dynamics of retrotransposition-driven genomic expansions in Oryza australiensis, a wild

relative of rice. Genome Res., 16, 1262-9.

Piperidis, G., A. D’Hont 2001 Chromosome composition analysis of various Saccharum

interspecific hybrids by genomic in situ hybridization GISH Proc. Int. Soc. Technol., 24, 565-

566.

Price, S. 1963 Cytogenetics of modern sugar canes Economic Botany, 17, 97–105.

Raboin, L.M., K.M. Oliveira, L. Le Cunff, H. Telismart, D. Roques, M. Butterfield, J.Y.

Hoarau, A. D'Hont, 2006 Genetic mapping in sugarcane, a high polyploid, using bi-parental

progeny: identification of a gene controlling stalk colour and a new rust resistance gene.

Theor. Appl. Genet., 112, 1382-1391.

32

Raboin L.M., J. Pauquet, M. Butterfield, A. D'Hont, J.C. Glaszmann, 2008. Analysis of

genome-wide linkage disequilibrium in the highly polyploid sugarcane. Theor. Appl. Genet.,

116, 5: 701-714.

Stirling, B., G. Newcombe, J. Vrebalov, I. Bosdet, H.D. Bradshaw, 2001 Suppressed

recombination around the MXC3 locus, a major gene for resistance to poplar leaf rust. Theor.

Appl. Genet., 103, 1129–1137.

Soderlund, C., S. Humphray, A. Dunham, L. French, 2000 Contigs built with fingerprints,

markers, and FPC V47. Genome Res., 10, 1772-87.

Sreenivasan, T.V., B.S. Ahloowalia, 1987 Cytogenetics pp. 211-253 in Sugarcane

improvement through breeding edited by Heinz DJ. Elsevier Science, Amsterdam

SUCEST, Sugarcane EST database, http://wwwwatsonfapespbr/sucesthtm.

Sun, X., Y. Cao, Z. Yang, C. Xu, X. Li, S. Wang, Q. Zhang, 2004 Xa26, a gene conferring

resistance to Xanthomonas oryzae pv. oryzae in rice, encodes an LRR receptor kinase-like

protein. Plant J., 37, 517-27.

Tai, P.Y.P., J.D. Miller, J.L. and Dean, 1981 Inheritance of resistance to rust in sugarcane.

Field Crops Res., 4, 261-268.

Tao, Y.Z., D.R. Jordan, R.G. Henzell, C.L. McIntyre, 1998 Identification of genomic regions

for rust resistance in sorghum. Euphytica, 103, 287-292.

33

Tomkins, J.P., Y. Yu, H. Miller-Smith, D.A. Frisch, S.S. Woo, R. Wing, 1999 A bacterial

artificial chromosome library for sugarcane. Theor. Appl. Genet., 99, 419-424.

Ueda, T., T. Sato, J. Hidema, T. Hirouchi, K. Yamamoto, T. Kumagai, M. Yano, 2005 qUVR-

10, a major quantitative trait locus for ultraviolet-B resistance in rice, encodes cyclobutane

pyrimidine dimer photolyase. Genetics, 171, 1941-50.

Vettore, A.L., F.R. da Silva, E.L. Kemper, G.M. Souza, A.M. da Silva, M.I. Ferro, F.

Henrique-Silva, E.A. Giglioti, M.V. Lemos, L.L. Coutinho, M.P. Nobrega, H. Carrer, S.C.

França, M. Bacci Júnior, M.H. Goldman, S.L. Gomes, L.R. Nunes, L.E. Camargo, W.J.

Siqueira, M.A. Van Sluys, O.H. Thiemann, E.E. Kuramae, R.V. Santelli, C.L. Marino, M.L.

Targon, J.A. Ferro, H.C. Silveira, D.C. Marini, E.G. Lemos, C.B. Monteiro-Vitorello, J.H.

Tambor, D.M. Carraro, P.G. Roberto, V.G. Martins, G.H. Goldman, R.C. de Oliveira, D.

Truffi, C.A. Colombo, M. Rossi, P.G. de Araujo, S.A. Sculaccio, A. Angella, M.M. Lima,

V.E. de Rosa Júnior, F. Siviero, V.E. Coscrato, M.A. Machado, L. Grivet, S.M. Di Mauro,

F.G. Nobrega, C.F. Menck, M.D. Braga, G.P. Telles, F.A. Cara, G. Pedrosa, J. Meidanis, P.

Arruda 2003 Analysis and functional annotation of an expressed sequence tag collection for

tropical crop sugarcane. Genome Res, 13, 2725-35

Wang, Z.X., M. Yano, U. Yamanouchi, M. Iwamoto, L. Monna, H. Hayasaka, Y. Katayose,

T. Sasaki, 1999 The Pib gene for rice blast resistance belongs to the nucleotide binding and

leucine-rich repeat class of plant disease resistance genes. Plant J., 19, 55-64.

34

Wei, F., K. Gobelman-Werner, S.M. Morroll, J. Kurth, L. Mao, R. Wing, D. Leister, P.

Schulze-Lefert, R.P. Wise, 1999 The Mla powdery mildew resistance cluster is associated

with three NBS-LRR gene families and suppressed recombination within a 240-kb DNA

interval on chromosome 5S 1HS of barley. Genetics, 153, 1929-48.

Wolfe, K.H., M. Gouy, Y.W. Yang, P.M. Sharp, W.H. Li, 1989 Date of the monocot-dicot

divergence estimated from chloroplast DNA sequence data. Proc. Natl. Acad. Sci. U S A, 86,

6201-5.

Xu, K., X. Xu, T. Fukao, P. Canlas, R. Maghirang-Rodriguez, S. Heuer, A.M. Ismail, J.

Bailey-Serres, P.C. Ronald, D.J. Mackill, 2006 Sub1A is an ethylene-response-factor-like

gene that confers submergence tolerance to rice. Nature, 10, 705-8.

35

FIGURE LEGENDS

Figure 1: Schematic representation of a typical modern sugarcane cultivar genome.

Each bar represents a chromosome; white coloring represents regions originating from S.

officinarum and grey coloring from S. spontaneum. Chromosomes aligned in the same row

are hom(oe)ologous and represent a homology group (HG). Chromosomes assembled in the

vertical rectangle correspond to a monoploid genome (MG) of S. officinarum. The key

characteristics of this genome are the high level of ploidy, the aneuploidy, the bispecific

origin of the chromosomes, the existence of structural differences between chromosomes of

the two origins, and the presence of interspecific chromosome recombinants.

Figure 2: Saturation of the sugarcane target genetic region from various resources.

Step 1: Three markers originally surrounding Bru1 in sugarcane (orange) were mapped on the

sorghum genetic maps (green) of Boivin et al. (1999 and unpublished data) (white arrows) to

identify the corresponding sorghum orthologous region in this map and by comparison in the

map of Bowers et al. (2003). Double green arrows link common markers between these two

sorghum maps. Several sorghum RFLP loci (underlined) from these two sorghum maps were

analyzed on R570 progeny allowing mapping new markers (pointed by green arrows) on the

sugarcane map. Step 2: Four RFLP probes (indicated in italic) from the sorghum genetic map

were used to construct a local sorghum orthologous physical map. Green arrows point to

sorghum BAC contigs obtained through BAC library screening and by consulting the online

sorghum physical map. Double black arrows indicate links between contigs (see text for

details). Most sorghum BAC-ends were analyzed on R570 progeny allowing four BAC-ends

(pointed by green arrows) to be genetically mapped in the sugarcane target region. Step 3:

Rice RFLP loci C673 originally map proximally from Bru1 in sugarcane and the RFLP loci

36

pPAP5H10 loci (derived from the sorghum genetic map) mapped proximally from Bru1

where used to screen a rice BAC library and to identify the corresponding rice orthologous

physical map (from Clemson University, www.genome.arizona.edu, represented in red).

Orange arrows point to the rice BACs identified. Sugarcane cDNA with homology to the rice

orthologous sequence were analyzed on R570 progeny allowing mapping new markers

(pointed by red arrows) on the sugarcane map. Genetic distances are indicated in

centimorgans. For the sugarcane map, distances indicated on right side are based on 312

individuals and those indicated on left side and in brackets are based on 712 individuals.

Markers used to build the sugarcane physical map are indicated in bold.

Figure 3: Physical map of the Bru1 region.

BAC clones are represented by vertical lines: orange for the target haplotype, brown for the

hom(e)ologous haplotypes and green for sorghum. Dotted-lines and white rounds indicate the

localisation of probes used on the sorghum or/and sugarcane physical map or/and on the

genetic map of the Bru1 region in sugarcane cv R570. Boxes assemble BAC clones for the

same haplotype. Probes in green represent those from the sorghum genetic or physical map,

those in red are from the sugarcane cDNA library, and those in orange are BAC-ends or

subclones of sugarcane BACs.

Figure 4: Method for identifying the BAC corresponding to the target haplotype versus

hom(oe)ologous BACs.

A. Schematic representation of hom(oe)ologous chromosome segments (haplotypes) bearing

four allelic RFLP markers with probeA/HindIII (a1 to a4). The target haplotype bearing allele

a1 is represented in black. B: RFLP profiles of R570 and six BACs analysed with probe A

37

(R15N23) in combination with HindIII. Asterics indicate BACs selected with probe A. BAC

164H22 contains marker a1 (pointed by an arrow) and thus belongs to the target haplotype.

Figure 5: Impact of the insertion on recombination

Genetic distances are indicated in centimorgans and are based on 112 individuals. The gray

cases indicate haplotype segments comprising the insertion.

38

recombinants

S. spontaneum

S. officinarum recombinants

S. spontaneum

S. officinarum

Figure 1

MG

HG

Figure 2

Bru

1

SUG

AR

CA

NE

Gen

etic

map

CG

VII

-1 R19

5K17

H

cBR

56

cBR

8

cBR

20

cBR

33cB

R32

cBR

37

pSB

0084

pPA

P05

H10

cBR

23

F12

3K21

DF5

7P2e

V

R24

P17

eV

C67

3

attc

ag

aacc

ac6

pSB

1445

cMW

G65

2aa

ccgt

aagc

ta23

CD

SR29

actc

tg9R

acgc

tt17

0.3

0.3

0.3

0.3

0.3

2.2

(0.2

8)

(0.2

8)

(0.1

4)

(0.1

4)

(0.1

4)

1.6

0.6

0.3

SUG

AR

CA

NE

cDN

A

RIC

E

Chr

2

BA

C s

eque

nces

CD

SR29

cMW

G65

2

SOR

GH

UM

Gen

etic

map

LG

4

PSR

8

2.1

5.6

M73

8

BN

L7.4

3U

MC

111

P206

08

AA

CC

AC

6

F67N

07D

1.1

1.60.6

PRC0

345a

CD

SR29

AES

T240

pSB

1445

pSB0

084

pSB1

565

pPAP

05H

10

M73

8

pSB0

927

pPA

P08G

01

SOR

GH

UM

Gen

etic

map

LG

4 6.2

1.5

1.6 1.5

3.1

2.3

(Boi

vin

et a

l199

9)

(Bow

erse

t al2

003)

AP004078AP004885

AP005311

AP005304

Centromer

R24

P17e

VR

57P2

eVF1

23K

21D

R19

5K17

H

Ctg1123Ctg458

Ctg1363

Ctg1455

Centromer

Centromer

Centromer

SOR

GH

UM

BA

C c

onti

g

Step

1

Step

2

Step

3

AP004121

Bru

1

SUG

AR

CA

NE

Gen

etic

map

CG

VII

-1 R19

5K17

H

cBR

56

cBR

8

cBR

20

cBR

33cB

R32

cBR

37

pSB

0084

pPA

P05

H10

cBR

23

F12

3K21

DF5

7P2e

V

R24

P17

eV

C67

3

attc

ag

aacc

ac6

pSB

1445

cMW

G65

2aa

ccgt

aagc

ta23

CD

SR29

actc

tg9R

acgc

tt17

0.3

0.3

0.3

0.3

0.3

2.2

(0.2

8)

(0.2

8)

(0.1

4)

(0.1

4)

(0.1

4)

1.6

0.6

0.3

SUG

AR

CA

NE

cDN

A

RIC

E

Chr

2

BA

C s

eque

nces

CD

SR29

cMW

G65

2

SOR

GH

UM

Gen

etic

map

LG

4

PSR

8

2.1

5.6

M73

8

BN

L7.4

3U

MC

111

P206

08

AA

CC

AC

6

F67N

07D

1.1

1.60.6

PRC0

345a

CD

SR29

AES

T240

pSB

1445

pSB0

084

pSB1

565

pPAP

05H

10

M73

8

pSB0

927

pPA

P08G

01

SOR

GH

UM

Gen

etic

map

LG

4 6.2

1.5

1.6 1.5

3.1

2.3

(Boi

vin

et a

l199

9)

(Bow

erse

t al2

003)

AP004078AP004885

AP005311

AP005304

Centromer Centromer

R24

P17e

VR

57P2

eVF1

23K

21D

R19

5K17

H

Ctg1123Ctg458

Ctg1363

Ctg1455

Centromer

Centromer

Centromer

SOR

GH

UM

BA

C c

onti

g

Step

1

Step

2

Step

3

AP004121

CIR

12E

3 ~1

25 k

bC

IR9O

20 ~

90 k

b22

M6

~85

kb16

4H22

~11

0 kb

R-CIR9O20

CIR12E3-A5

CIR12E3-D10R-CIR12H16

22M6-H5

R-CIR12E3

CIR12 E3-B6

CIR12 E3-D7CIR12E3-H6

Targ

et h

aplo

type

7

GAP

GAP

15N

23 ~

120

kb

253G

12 ~

135

kb

197G

04 ~

100

kb

53A

11 ~

80 k

b

142J

21 ~

135

kb

135P

16 ~

90 k

b

cBR37

1 2 3 6

4

24P

17 ~

90

kb

149J

14 ~

120

kb

195K

17 ~

95K

b

5

F123K21D

cBR33

R15N23S

F12H16ScBR32

CIR9O20-D10

CIR9O20-F4

123K

21

91A

01

25N

07 47B

17

28G

16

CIR

12H

16

R24P17eV

91A

01

186A

6

237P

21

3E15

183A

14

256D

06

217H

18

130G

17

109O

20

237G

08

82G

08

47H

06

81O

07

147P

08

86O

15

R91A01H

162L

23

PSB0084

R150O5S

F143C06H

F150O05V

183A

14

150O

05

64F0

5

175K

08

50H

06

137O

05

4F22

14D

02

R195K17S

cBR23

pPAP05H10

inse

rtio

n

a b c d

γ

ε

α β δ

CIR

43P

06

0.14

cM

0.14

cM

0.14

cM

0.28

cM

Bru

1

OrthologousSorghum

BAC contig Sugarcane Homoeologous BACs Genetic mapof CG VII-1

Targ

et re

gion

Figure 3

vector

230I

0225

3G12

*13

5P16

*14

4E02

*

R57

0

114F

1416

4H22

*

Hom(oe)ologouschromosomes

Targethaplotype

Bru 1

Locus AProbe A / Hind III a1 a2 a3 a4

a1

a2

a3

a4

Figure 4

A B

cBR8cBR20

pPAP05H10cBR23R195K17S

R15N23S

Fcir12H16ScBR32

cir12EH06cir12EB06

cir12ED07

cir12EA05

Rcir12E03Rcir9O20cir9O20F04cir9O20D10cBR33

cBR37F57P2eVR24P17eVPSB0084cBR47cBR56

R570 CG VII-1

inse

rtion

F123K21D

0.9

0.9

cir12E3D10Rcir12H16

0.9

Bru

1

TABLEAU

--2.7 cM0 cMR15N23S-F123K21D

-7.2 cM3.7 cM1.8 cMcBR20-F123K21D

--4.7 cM1.8 cMcBR8-F123K21D

10.3 cM--2.7 cMcBR8-cBR56

--1 cM1.8 cMcBR20-R15N23S

--2 cM1.8 cMcBR8-R15N23S

CGVII-d

CGVII-c

CGVII-b

CG VII-1Loci intervals

Bru1haplotype

Homo(e)ologoushaplotypes

Figure 5