Rapid development of molecular markers by next-generation sequencing linked to a gene conferring phomopsis stem blight disease resistance for marker-assisted selection in lupin (Lupinus angustifolius L.) breeding

Selection for phomopsis stem blight disease (PSB) resistance is one of the key objectives in lupin (Lupinus angustifolius L.) breeding programs. A cross was made between cultivar Tanjil (resistant to PSB) and Unicrop (susceptible). The progeny was advanced into F₈ recombinant inbred lines (RILs). The RIL population was phenotyped for PSB disease resistance. Twenty plants from the RIL population representing disease resistance and susceptibility was subjected to next-generation sequencing (NGS)-based restriction site-associated DNA sequencing on the NGS platform Solexa HiSeq2000, which generated 7,241 single nucleotide polymorphisms (SNPs). Thirty-three SNP markers showed the correlation between the marker genotypes and the PSB disease phenotype on the 20 representative plants, which were considered as candidate markers linked to a putative R gene for PSB resistance. Seven candidate markers were converted into sequence-specific PCR markers, which were designated as PhtjM1, PhtjM2, PhtjM3, PhtjM4, PhtjM5, PhtjM6 and PhtjM7. Linkage analysis of the disease phenotyping data and marker genotyping data on a F₈ population containing 187 RILs confirmed that all the seven converted markers were associated with the putative R gene within the genetic distance of 2.1 CentiMorgan (cM). One of the PCR markers, PhtjM3, co-segregated with the R gene. The seven established PCR markers were tested in the 26 historical and current commercial cultivars released in Australia. The numbers of “false positives” (showing the resistance marker allele band but lack of the putative R gene) for each of the seven PCR markers ranged from nil to eight. Markers PhtjM4 and PhtjM7 are recommended in marker-assisted selection for PSB resistance in the Australian national lupin breeding program due to its wide applicability on breeding germplasm and close linkage to the putative R gene. The results demonstrated that application of NGS technology is a rapid and cost-effective approach in development of markers for molecular plant breeding.     

A strategy to develop molecular markers applicable to a wide range of crosses for marker assisted selection in plant breeding: a case study on anthracnose disease resistance in lupin (Lupinus angustifolius L.)

A key challenge in marker-assisted selection (MAS) for molecular plant breeding is to develop markers linked to genes of interest which are applicable to multiple breeding populations. In this study representative F₂ plants from a cross Mandalup (resistant to anthracnose disease) × Quilinock (susceptible) of Lupinus angustifolius were used in DNA fingerprinting by Microsatellite-anchored Fragment Length Polymorphism (MFLP). Nine candidate MFLP markers linked to anthracnose resistance were identified, then ‘validated’ on 17 commercial cultivars. The number of “false positives” (showing resistant-allele band but lack of the R gene) for each of the nine candidate MFLP markers on the 17 cultivars ranged from 1 to 9. The candidate marker with least number of false positive was selected, sequenced, and was converted into a co-dominant, sequence-specific, simple PCR based marker suitable for routine implementation. Testing on 180 F₂ plants confirmed that the converted marker was linked to the R gene at 5.1 centiMorgan. The banding pattern of the converted marker was consistent with the disease phenotype on 23 out of the 24 cultivars. This marker, designated “AnManM1”, is now being used for MAS in the Australian lupin breeding program. We conclude that generation of multiple candidate markers, followed by a validation step to select the best marker before conversion to an implementable form is an efficient strategy to ensure wide applicability for MAS.     

A molecular marker linked to the mollis gene conferring soft-seediness for marker-assisted selection applicable to a wide range of crosses in lupin (Lupinus angustifolius L.) breeding

To broaden the gene pool of domesticated commercial cultivars of narrow-leafed lupin (Lupinus angustifolius L.), wild accessions are used as parents in crossing in lupin breeding. Among the progenies from wild × domesticated (W × D) crosses, the soft-seediness gene mollis is the most difficult domestication gene to be selected by conventional breeding methods, where molecular marker-assisted selection (MAS) is highly desirable. MAS in plant breeding requires markers to be cost-effective and high-throughput, and be applicable to a wide range of crosses in a breeding program. In this study, representative plants from an F8 recombinant inbred line (RIL) population derived from a W × D cross, together with four cultivars and four wild types, were used in DNA fingerprinting by microsatellite-anchored fragment length polymorphisms (MFLP). Two co-dominant MFLP polymorphisms were identified as candidate markers linked to the mollis gene, and one of the candidate markers was selected and converted into a co-dominant, sequence-specific PCR marker. This marker, designated MoLi, showed a perfect match with phenotypes of seed coat permeability on a segregating population consisting of 115 F8 RILs, confirming the close genetic linkage to the mollis gene. Validation tests showed that the banding pattern of marker MoLi is consistent with all the 25 historical and current commercial cultivars released in Australia, and is consistent with mollis genotypes in 119 of the 125 accessions in the Australian L. angustifolius core collection. Marker MoLi provides a cost-effective way to select the mollis gene in a wide range of W × D crosses in lupin breeding.     

Application of in silico bulked segregant analysis for rapid development of markers linked to Bean common mosaic virus resistance in common bean

Background

Common bean was one of the first crops that benefited from the development and utilization of molecular marker-assisted selection (MAS) for major disease resistance genes. Efficiency of MAS for breeding common bean is still hampered, however, due to the dominance, linkage phase, and loose linkage of previously developed markers. Here we applied in silico bulked segregant analysis (BSA) to the BeanCAP diversity panel, composed of over 500 lines and genotyped with the BARCBEAN_3 6K SNP BeadChip, to develop codominant and tightly linked markers to the I gene controlling resistance to Bean common mosaic virus (BCMV).

Results

We physically mapped the genomic region underlying the I gene. This locus, in the distal arm of chromosome Pv02, contains seven putative NBS-LRR-type disease resistance genes. Two contrasting bulks, containing BCMV host differentials and ten BeanCAP lines with known disease reaction to BCMV, were subjected to in silico BSA for targeting the I gene and flanking sequences. Two distinct haplotypes, containing a cluster of six single nucleotide polymorphisms (SNP), were associated with resistance or susceptibility to BCMV. One-hundred and twenty-two lines, including 115 of the BeanCAP panel, were screened for BCMV resistance in the greenhouse, and all of the resistant or susceptible plants displayed distinct SNP haplotypes as those found in the two bulks. The resistant/susceptible haplotypes were validated in 98 recombinant inbred lines segregating for BCMV resistance. The closest SNP (~25-32 kb) to the distal NBS-LRR gene model for the I gene locus was targeted for conversion to codominant KASP (Kompetitive Allele Specific PCR) and CAPS (Cleaved Amplified Polymorphic Sequence) markers. Both marker systems accurately predicted the disease reaction to BCMV conferred by the I gene in all screened lines of this study.

Conclusions

We demonstrated the utility of the in silico BSA approach using genetically diverse germplasm, genotyped with a high-density SNP chip array, to discover SNP variation at a specific targeted genomic region. In common bean, many disease resistance genes are mapped and their physical genomic position can now be determined, thus the application of this approach will facilitate further development of codominant and tightly linked markers for use in MAS.

Electronic supplementary material

The online version of this article (doi:10.1186/1471-2164-15-903) contains supplementary material, which is available to authorized users.     

Identification and potential use of RAPD markers linked to Yam mosaic virus resistance in white yam (Dioscorea rotundatd)

Resistance to Yam mosaic virus (YMV) in tetraploid white yam (Dioscorea rotundatd) is inherited differentially as a dominant and recessive character. Elite D. rotundata breeding lines with durable resistance to YMV can be developed by pyramiding major dominant and recessive genes using marker‐assisted selection (MAS). The tetraploid breeding line, TDr 89/01444, is a source of dominant genetic resistance to yam mosaic disease. Bulked segregant analysis was used to search for random amplified polymorphic DNA (RAPD) markers linked to YMV resistance in F₁ progeny derived from a cross between TDr 89/01444 and the susceptible female parent, TDr 87/00571. The F₁ progeny segregated 1:1 (resistantsusceptible) when inoculated with a Nigerian isolate of YMV, confirming that resistance to YMV in TDr 89/01444 was dominantly inherited. A single locus that contributes to YMV resistance in TDr 89/01444 was identified and tentatively named Ymv‐1. Two RAPD markers closely linked in coupling phase with Ymv‐1 were identified, both of which were mapped on the same linkage group: OPW18₈₅₀ (3.0 centiMorgans [cM]) and OPX15₈₅₀ (2.0 cM). Both markers successfully identified Ymv‐1 in resistant genotypes among 12 D. rotundata varieties and in resistant F1 individuals from the cross TDr 93–1 × TDr 877 00211, indicating their potential for use in marker‐assisted selection. OPW18₈₅₀ and OPX15₈₅₀ are the first DNA markers for YMV resistance and represent a starting point in the use of molecular markers to assist breeding for resistance to YMV.     

DNA markers tightly linked to a gall midge resistance gene (Gm2) are potentially useful for marker-aided selection in rice breeding

We have developed a polymerase chain reaction (PCR)-based assay that could effectively reduce the time period required to screen and select for Gall Midgeresistant rice lines under field conditions. The primers for the assay were designed on the basis of sequence information of two phenotype specific random amplified polymorphic DNA fragments which were found to be tightly linked to Gall Midge biotype-1 resistance gene (Gm2). The two RAPD fragments, F8₁₇₀₀ in the susceptible parent ARC6650 and F10₆₀₀ in the resistant parent Phalguna, were identified after screening 5450 loci using 520 random primers on genomic DNAs of ARC6650 and Phalguna. These primers, when used in a multiplexed PCR, amplified specifically a 1.7-kb and 0.6-kb fragment in the susceptible and resistant parents, respectively. When this assay was performed on genomic DNAs of 44 recombinant inbred lines derived from ARC6650 x Phalguna and 5 lines derived from other crosses where one of the parents was Phalguna, ARC6650 or their derivatives, the primers amplified a 1.7-kb fragment in all of the susceptible lines or a 0.6-kb fragment in all of the resistant ones. These markers can be of potential use in the marker-aided selection of Gall Midge biotype-1 resistant phenotypes. As screening for resistance can now be conducted independent of the availability of insects, the breeding of resistant varieties can be hastened.     

Development and utility of cleaved amplified polymorphic sequences (CAPS) and restriction fragment length polymorphisms (RFLPs) linked to the Fom-2 fusarium wilt resistance gene in melon (Cucumis melo L.)

Fusarium wilt, caused by Fusarium oxysporum Schlecht f. sp. melonis Snyder & Hans, is a worldwide soil-borne disease of melon (Cucumis melo L.). Resistance to races 0 and 1 of Fusarium wilt is conditioned by the dominant gene Fom-2. To facilitate marker-assisted backcrossing with selection for Fusarium wilt resistance, we developed cleaved amplified polymorphic sequences (CAPS) and restriction fragment length polymorphisms (RFLP) markers by converting RAPD markers E07 (a 1.25-kb band) and G17 (a 1.05-kb band), respectively. The RAPD-PCR polymorphic fragments from the susceptible line ’Vedrantais’ were cloned and sequenced in order to construct primers that would amplify only the target fragment. The derived primers, E07SCAR-1/E07SCAR-2 from E07 and G17SCAR-1/G17SCAR-2 from G17, yielded a single 1.25-kb fragment (designated SCE07) and a 1.05-kb fragment (designated SCG17) (the same as RAPD markers E07 and G17), respectively, from both resistant and susceptible melon lines, thus demonstrating locus-specific associated primers. Potential CAPS markers were first revealed by comparing sequence data between fragments amplified from resistant (PI 161375) and susceptible (’Vedrantais’) lines and were then confirmed by electrophoresis of restriction endonuclease digestion products. Twelve restriction endonucleases were evaluated for their potential use as CAPS markers within the SCE07 fragment. Three (BclI, MspI, and BssSI) yielded ideal CAPS markers and were subsequently subjected to extensive testing using an additional 88 diverse melon cultigens, 93 and 119 F₂ individuals from crosses of ’Vedrantais’ x PI 161375 and ’Ananas Yokneam’×MR-1 respectively, and 17 families from a backcross BC₁S₁ population derived from the breeding line ’MD8654’ as a resistance source. BclI- and MspI-CAPS are susceptible-linked markers, whereas the BssSI-CAPS is a resistant-linked marker. The CAPS markers that resulted from double digestion by BclI and BssSI are co-dominant. Results from BclI- and MspI-CAPS showed over 90% accuracy in the melon cultigens, and nearly 100% accuracy in the F₂ individuals and BC₁S₁ families tested. This is the first report of PCR-based CAPS markers linked to resistance/susceptibility for Fusarium wilt in melon. The RFLP markers resulting from probing with a clone-derived 1.05-kb SCG17 PCR fragment showed 85% correct matches to the disease phenotype. Both the CAPS and RFLP markers were co-dominant, easier to score, and more accurate and consistent in predicting the melon phenotype than the RAPD markers from which they were derived. Received: 28 July 1998 / Accepted: 7 December 1998     

Stability over genetic backgrounds, generations and years of quantitative trait locus (QTLs) for organoleptic quality in tomato

The efficiency of marker-assisted backcross for the introgression of a quantitative trait locus (QTL) from a donor line into a recipient line depends on the stability of QTL expression. QTLs for six quality traits in tomato (fruit weight, firmness, locule number, soluble solid content, sugar content and titratable acidity) were studied in order to investigate their individual effect and their stability over years, generations and genetic backgrounds. Five chromosome regions carrying fruit quality QTLs were transferred following a marker-assisted backcross scheme from a cherry tomato line into three modern lines with larger fruits. Three sets of genotypes corresponding to three generations were compared: (1) an RIL population, which contained 50% of each parental genome, (2) three BC3S1 populations which segregated simultaneously for the five regions of interest but were almost fully homozygous for the recipient genome on the eight chromosomes carrying no QTL and (3) three sets of QTL-NILs (BC3S3 lines) which differed from the recipient line only in one of the five regions. QTL detection was performed in each generation, in each genetic background and during 2 successive years for QTL-NILs. About half of the QTLs detected in QTL-NILs were detected in both years. Eight of the ten QTLs detected in RILs were recovered in the QTL-NILs with the genetic background used for the initial QTL mapping experiment, with the exception of two QTLs for fruit firmness. Several new QTLs were detected. In the two other genetic backgrounds, the number of QTLs in common with the RILs was lower, but several new QTLs were also detected in advanced generations.     

Identification of random amplified polymorphic DNA (RAPD) markers linked to the v locus in barley, Hordeum vulgare L.

 Recombinant backcross lines of barley were produced from a cross between Kanto Nakate Gold (KNG; two-rowed) and Azumamugi (AZ; six-rowed) after backcrosses of F₁ plants with AZ as the recurrent parent. Each of these lines had an introgressed segment from chromosome 2 of KNG. Two recombinant backcross lines, L1 and M3-13, were used for an initial screening of polymorphism. After screening a total of 888 oligonucleotides as arbitrary primers, we identified eight random amplified polymorphic DNAs (RAPDs) between backcross lines and AZ. Among the RAPD fragments, CMNA-38₇₀₀ was linked to the v locus with a recombination frequency of zero, while OPJ-09₈₅₀ and OPP-02₇₀₀ were linked to the v locus at a map distance of 1.4 cM. Thus, the three RAPD markers were clustered around the v locus since the lengths of introgressed chromosomal segments in the L1 and M3-13 lines were no less than 38 cM. The other five RAPD fragments that we identified were not linked to the v locus. Received: 14 January 1997 / Accepted: 14 February 1997