BackgroundMost modern citrus cultivars have an interspecific origin. As a foundational step towards deciphering the interspecific genome structures, a reference whole genome sequence was produced by the International Citrus Genome Consortium from a haploid derived from Clementine mandarin. The availability of a saturated genetic map of Clementine was identified as an essential prerequisite to assist the whole genome sequence assembly. Clementine is believed to be a ‘Mediterranean’ mandarin × sweet orange hybrid, and sweet orange likely arose from interspecific hybridizations between mandarin and pummelo gene pools. The primary goals of the present study were to establish a Clementine reference map using codominant markers, and to perform comparative mapping of pummelo, sweet orange, and Clementine.ResultsFive parental genetic maps were established from three segregating populations, which were genotyped with Single Nucleotide Polymorphism (SNP), Simple Sequence Repeats (SSR) and Insertion-Deletion (Indel) markers. An initial medium density reference map (961 markers for 1084.1 cM) of the Clementine was established by combining male and female Clementine segregation data. This Clementine map was compared with two pummelo maps and a sweet orange map. The linear order of markers was highly conserved in the different species. However, significant differences in map size were observed, which suggests a variation in the recombination rates. Skewed segregations were much higher in the male than female Clementine mapping data. The mapping data confirmed that Clementine arose from hybridization between ‘Mediterranean’ mandarin and sweet orange. The results identified nine recombination break points for the sweet orange gamete that contributed to the Clementine genome.ConclusionsA reference genetic map of citrus, used to facilitate the chromosome assembly of the first citrus reference genome sequence, was established. The high conservation of marker order observed at the interspecific level should allow reasonable inferences of most citrus genome sequences by mapping next-generation sequencing (NGS) data in the reference genome sequence. The genome of the haploid Clementine used to establish the citrus reference genome sequence appears to have been inherited primarily from the ‘Mediterranean’ mandarin. The high frequency of skewed allelic segregations in the male Clementine data underline the probable extent of deviation from Mendelian segregation for characters controlled by heterozygous loci in male parents.

The genetic structure of 2n gametes and, particularly, the parental heterozygosity restitution at each locus depends on the meiotic process by which they originated, with firstdivision restitution and second-division restitution (SDR) being the two major mechanisms. The origin of 2n gametes in citrus is still controversial, although sexual polyploidisation is widely used for triploid seedless cultivar development. In this study, we report the analysis of 2n gametes of mandarin cv 'Fortune' by genotyping 171 triploid hybrids with 35 simple sequence repeat markers. The microsatellite DNA allele counting-peak ratios method for alleledosage evaluation proved highly efficient in segregating triploid progenies and allowed half-tetrad analysis (HTA) by inferring the 2n gamete allelic configuration. All 2n gametes arose from the female genitor. The observed maternal heterozygosity restitution varied between 10 and 82%, depending on the locus, thus SDR appears to be the mechanism underlying 2n gamete production in mandarin cv 'Fortune'. A new method to locate the centromere, based on the best fit between observed heterozygosity restitution within a linkage group and theoretical functions under either partial or no chiasmata interference hypotheses was successfully applied to linkage group II. The maximum value of heterozygosity restitution and the pattern of restitution along this linkage group would suggest there is partial chiasma interference. The implications of such a restitution mechanism for citrus breeding are discussed.

Tetraploid clementine displays mainly tetrasomic inheritance. Genetic structures of 2n SDR and 2 × gametes from DD clementine are complementary and will guides triploids citrus breeding strategies. Triploid breeding is developed worldwide to create new seedless cultivars. Citrus triploid hybrids can be recovered from 2x × 2x sexual hybridizations as a consequence of the formation of unreduced gametes (2n), or from 4x × 2x interploid hybridizations in which tetraploid parents used are most often doubled-diploid (DD). Here we have analyzed the inheritance in doubled-diploid clementine and compared the genetic structures of gametes of DD clementine with SDR unreduced gametes of diploid clementine. Parental heterozygosity restitution (PHR) with DD parents depends on the rate of preferential chromosome pairing and thus the proportion of disomic versus tetrasomic segregations. Doubled-diploid clementine largely exhibited tetrasomic segregation. However, three linkage groups had intermediate segregation and one had a tendency for disomy. Significant doubled reduction rates (DR) rates were observed in six of the nine LGs. Differences of PHR between 2n SDR and 2x DD gametes were highest in the centromeric region and progressively decreased toward the distal regions where they were not significant. Over all markers, PHR was lower (two-thirds) in SDR 2n gametes than in DD-derived diploid gametes. The two strategies appear complementary in terms of genotypic variability. Interploid 4x × 2x hybridization is potentially more efficient for developing new cultivars that are phenotypically closer to the diploid parent of the DD than sexual hybridization through SDR 2n gametes. Conversely, 2x × 2x triploidisation has the potential to produce novel products with characteristics for market segmentation strategies.

Seedlessness is one of the most important characteristics for mandarins for the fresh-fruit market and mandarin triploid hybrids have this trait. Citrus triploid plants can be recovered by 2x x 2x sexual hybridisations as a consequence of the formation of unreduced gametes at low frequency. Triploid embryos are found in small seeds that do not germinate under greenhouse conditions. Extensive breeding programs based on this type of hybridisation require very effective methodologies for embryo rescue and ploidy evaluation. In this work, we describe an effective methodology to recover triploid hybrids from 2x x 2x hybridisations based on in vitro embryo rescue and ploidy level determination by means of flow cytometry. The influence of parents and environmental conditions on obtaining triploid hybrids has been analysed. The strongest effect was associated with the genotype of the female parent while a strong interaction was found between the male parent genotype and environmental conditions. The effect of the female parent genotype on the length of the juvenile phase was also demonstrated by observing a large number of progenies over the last 10 years. The methodology described here has enabled us to obtain over 4,000 triploid hybrids so far, of which 13 have been protected in the European Union and two are being extensively planted by citrus growers to establish new commercial plots. These triploid hybrids have been analysed with simple sequence repeats markers to differentiate all the new triploid varieties and their parents, and thus molecular identification will help defend plant breeders' rights.

Fifteen SNP markers were successfully designed that produced clear allele signals that were in agreement with previous genotyping results at the diploid level. The analysis of DNA mixes between two haploid lines (Clementine and pummelo) at 13 different ratios revealed a very high correlation (average = 0·9796; s.d. = 0·0094) between the allele ratio and two parameters [θ angle = tan(-1) (y/x) and y' = y/(x + y)] derived from the two normalized allele signals (x and y) provided by KASPar. Separated cluster analysis and analysis of variance (ANOVA) from mixed DNA simulating triploid and tetraploid hybrids provided 99·71 % correct allelic configuration. Moreover, triploid populations arising from 2n gametes and interploid crosses were easily genotyped and provided useful genetic information. This work demonstrates that the KASPar SNP genotyping technique is an efficient way to assign heterozygous allelic configurations within polyploid populations. This method is accurate, simple and cost-effective. Moreover, it may be useful for quantitative studies, such as relative allele-specific expression analysis and bulk segregant analysis.