Microsatellite isolation and marker development in carrot - genomic distribution, linkage mapping, genetic diversity analysis and marker transferability across Apiaceae

Abstract

Background: The Apiaceae family includes several vegetable and spice crop species among which carrot is the most economically important member, with ~21 million tons produced yearly worldwide. Despite its importance, molecular resources in this species are relatively underdeveloped. The availability of informative, polymorphic, and robust PCR-based markers, such as microsatellites (or SSRs), will facilitate genetics and breeding of carrot and other Apiaceae, including integration of linkage maps, tagging of phenotypic traits and assisting positional gene cloning. Thus, with the purpose of isolating carrot microsatellites, two different strategies were used; a hybridization-based library enrichment for SSRs, and bioinformatic mining of SSRs in BAC-end sequence and EST sequence databases. This work reports on the development of 300 carrot SSR markers and their characterization at various levels.

Results: Evaluation of microsatellites isolated from both DNA sources in subsets of 7 carrot F2 mapping populations revealed that SSRs from the hybridization-based method were longer, had more repeat units and were more polymorphic than SSRs isolated by sequence search. Overall, 196 SSRs (65.1%) were polymorphic in at least one mapping population, and the percentage of polymophic SSRs across F2 populations ranged from 17.8 to 24.7. Polymorphic markers in one family were evaluated in the entire F2, allowing the genetic mapping of 55 SSRs (38 codominant) onto the carrot reference map. The SSR loci were distributed throughout all 9 carrot linkage groups (LGs), with 2 to 9 SSRs/LG. In addition, SSR evaluations in carrot-related taxa indicated that a significant fraction of the carrot SSRs transfer successfully across Apiaceae, with heterologous amplification success rate decreasing with the target-species evolutionary distance from carrot. SSR diversity evaluated in a collection of 65 D. carota accessions revealed a high level of polymorphism for these selected loci, with an average of 19 alleles/locus and 0.84 expected heterozygosity.

Conclusions: The addition of 55 SSRs to the carrot map, together with marker characterizations in six other mapping populations, will facilitate future comparative mapping studies and integration of carrot maps. The markers developed herein will be a valuable resource for assisting breeding, genetic, diversity, and genomic studies of carrot and other Apiaceae.

Figure 1

Distribution of microsatellites in genomic and transcript sequences of carrot. Frequency distribution (%) of predominant repeat types (A-C) and sequence motifs (D-F), by microsatellites origins GSSRs (A, D), BSSRs (B, E), and EST-derived SSRs (ESSRs) (C, F), inside and outside open reading frames (ORFs).

Figure 2

Relationship between polymorphism and repeat number classes for perfect microsatellites. Polymorphism Index values were calculated based on marker evaluations in 7 carrot F₂ families as described in materials and methods. Numbers in parenthesis above the bars indicate the percentage of polymorphic markers for a particular repeat number class (i.e., the % of markers that were polymorphic in at least one F₂ population). The mean number and range of repeat units for each SSR type are provided in the legend. *All penta and hexanucleotides repeats had less than 5 repeat units and, thus, were only included in the first class. ^† For SSR markers that included more than 1 perfect microsatellite in their amplicon sequence the sum of all repeat units was considered [e.g., (AT)6, (CTT)5 = 11 repeats]. ^☐ Includes all markers with perfect SSRs.

Figure 3

Carrot SSR marker transferability across Apiaceae species. Number and percentage of SSR markers that successfully generated PCR products of nearly expected size (i.e., the expected amplicons length for carrot, Additional File 1 - Table S1) in 23 Apiaceae accessions, including 8 accessions of carrot (1 inbred line, 6 commercial cultivars, and 1 wild carrot) (black bars), 8 non-carrot Daucus species (grey bars), and 7 non-Daucus Apiaceae species (white bars). Names of carrot cultivars and inbred lines, and the common names of non-carrot species are in parentheses.

Figure 4

Genetic linkage maps of the wild carrot parent QAL and cultivated parent B493. Microsatellites mapped in this work are denoted in red. Carotenoid biosynthesis genes [8] and DcMTD markers [9] are denoted in blue and green letters, respectively. Codominant markers present in both QAL and 493 maps are underlined. B493-map positions for significant QTL for root total carotenes (displaying 95% support intervals) in LG5 (former LG2) and LG7 (former LG5), were estimated based on data published previously [7,58], and represented by yellow circles. LGs were ordered and named according to their corresponding physical chromosomes [19].