An EST-SSR-based linkage map for Persea americana Mill. (avocado)

Tree Genetics & Genomes (2009) 5:553–560 DOI 10.1007/s11295-009-0208-y ORIGINAL PAPER An EST-SSR-based linkage map for Persea americana Mill. (avocado) James W. Borrone & J. Steven Brown & Cecile L. Tondo & Margarita Mauro-Herrera & David N. Kuhn & Helen A. Violi & Robert T. Sautter & Raymond J. Schnell Received: 21 August 2008 / Revised: 10 December 2008 / Accepted: 29 March 2009 / Published online: 6 May 2009 # US Government 2009 Abstract Recent enhancement of the pool of known molecular markers for avocado has allowed the construction of the first moderately dense genetic map for this species. Over 300 SSR markers have been characterized and 163 of these were used to construct a map from the reciprocal cross of two Florida cultivars 'Simmonds' and 'Tonnage'. One hundred thirty-five primer pairs amplified 163 usable loci with 20 primer pairs amplifying more than one locus. 'Tonnage' was heterozygous for 152 (93%) loci, whereas 'Simmonds' was heterozygous for 64 (39%). Null alleles were identified at several loci. Linkage maps were produced for both reciprocal crosses and combined to generate a composite linkage map for the F1 population of 715 individuals. The composite map contains 12 linkage groups. Linkage groups ranged in size from 157.3 cM Communicated by: F. Gmitter Electronic supplementary material The online version of this article (doi:10.1007/s11295-009-0208-y) contains supplementary material, which is available to authorized users. J. W. Borrone : J. S. Brown : C. L. Tondo : M. Mauro-Herrera : D. N. Kuhn : H. A. Violi : R. T. Sautter : R. J. Schnell (*) United States Department of Agriculture, Agricultural Research Service, Subtropical Horticulture Research Station, National Germplasm Repository, 13601 Old Cutler Road, Miami, FL 33158, USA e-mail: [email protected] J. S. Brown e-mail: [email protected] D. N. Kuhn e-mail: [email protected] Present Address: J. W. Borrone : M. Mauro-Herrera Department of Botany, Oklahoma State University-Stillwater, 104 Life Sciences East, Stillwater, OK 74078-3013, USA (LG2) to 2.4 cM (LG12) and the number of loci mapped per group ranged from 29 (LG1) to two (LG12). The total map length was 1,087.4 cM. Only seven markers were observed to have segregation distortion (α≤0.05) across both sub-composite (reciprocal) maps. Phenotypic data from traits of horticultural interest are currently being collected on this population with the ultimate goal of identifying useful quantitative trait loci and the development of a marker-assisted selection program. Keywords Avocado . Reciprocal crosses . Florida cultivars Introduction Avocado (Persea americana Mill.) is an evergreen tree native to Mesoamerica. P. americana is a highly heterozygous species consisting of several botanical varieties, or subspecies, adapted to geographically distinct environments. This species includes three largely uncultivated, wild botanical varieties [var. floccosa Mez., var. steyermarkii Allen, and var. nubigena (Williams) Kopp] and three horticulturally important botanical varieties, also known as "races" [var. americana Mill. "West Indian", var. guatemalensis Williams "Guatemalan", and var. drymifolia (Schlecht. and Cham.) Blake "Mexican"] (Scora and Bergh 1990). Sexual barriers do not exist between the botanical varieties and hybrids are common (Nakasone and Paull 1999). Guatemalan–Mexican hybrids, e.g.‘Hass’, ‘Bacon’, and ‘Fuerte’, are cultivated in the dry Mediterranean-like environments of Mexico, California, Israel, Chile, South Africa, and the upland areas of Asia. West Indian or West Indian–Guatemalan hybrids, e.g. ‘Simmonds’, ‘Tonnage’, ‘Choquette’, ‘Monroe’, and ‘Tower 2’, are cultivated in the more humid, tropical-like environments of Florida, Hawaii, Southeast Asia, Guatemala, southern Mexico, and the 554 Caribbean. Mexico is the primary producer of avocados in the world, centered in the state of Michoacán and 'Hass' is the predominant cultivar grown worldwide (Ashworth et al. 2007). In the United States, avocado was introduced into Florida in 1833, California in 1848, and Hawaii in 1855. Southern California produces 90% of the total avocado crop of the United States, with Florida and Hawaii accounting for 9% and 1%, respectively (USDA 2008). The genetic control of important production traits in avocado, such as alternate bearing, need for cross-pollination, oil content, flavor characteristics, and Phytophthora root rot (PRR) tolerance are not well-understood. Lavi et al. (1991) investigated genetic variation among progeny from five cultivars using eight phenotypic traits. Additive genetic variance was found to be low for fruit weight and fruit density, medium for tree size, inflorescence length and seed size, and large for softening time, harvest duration, and flowering intensity. The inheritance of skin color, flowering group, and anise scent was estimated using 14 parents and 1,688 total progeny (Lavi et al. 1993). In each case, the traits were found to be coded by several loci with several alleles at each locus. Lavi et al. (1993) suggested that the phenotype shifts from one phase to the other when some threshold value is reached. Ploetz et al. (2002) screened 2,355 openpollinated progeny from 51 accessions of the USDA avocado germplasm collection located in Miami, Florida for tolerance to PRR. Parents that transmit tolerance to their progeny were identified and the broad sense heritability of the tolerance reaction was estimated to be 0.45. Improvement of avocado cultivars has principally occurred through chance selection of open-pollinated seedlings, mainly by growers. Breeding avocado, like most fruit trees, is difficult because of large plant size, long evaluation periods, outcrossing behavior, and the fact that large scale, controlled pollinations are difficult (Lammerts 1942; Degani et al. 2003). The haploid genome size of avocado has been estimated to be 8.83×108 bp (Arumuganathan and Earle 1991). Avocado contains 24 chromosomes with bivalent pairing at meiosis indicating that n=12 (Darlington and Wylie 1945). Polyploids, though described from naturally occurring P. americana seedlings, are rare (<1%; Garcia 1975). Avocado has been proposed to be the result of an ancient polyploid event (Chanderbali et al. 2008). However, evidence from analyses of germplasm collections and seedling populations using simple sequence repeat (SSR) markers have demonstrated diploid inheritance for most markers although some primer pairs have amplified more than one locus (Schnell et al. 2003; Borrone et al. 2007, 2008). The first and only existing linkage map of avocado was produced by Sharon et al. (1997) from 50 progeny of a cross between 'Pinkerton' and 'Ettinger' using 50 SSR markers, 17 random amplified polymorphic DNA markers, and 23 minisatellite DNA fingerprint (DFP) markers. Tree Genetics & Genomes (2009) 5:553–560 Twelve linkage groups with 34 mapped loci covering 352.6 cM were identified. Seven linkage groups contained two markers, two linkage groups contained three markers, one linkage group contained four markers, and two linkage groups contained five markers each. A larger population and an increased number of genetic markers were needed to produce a linkage map useful for quantitative trait loci (QTL) discovery. The recent development of over 300 SSR markers (Sharon et al. 1997; Ashworth et al. 2004; Borrone et al 2007; Borrone unpublished) enables the development of a saturated genetic linkage map and the potential identification of QTLs controlling horticultural traits of interest in avocado. Mapping populations have been developed in California (Ashworth et al. 2007; Chen et al. 2007) and in Spain (Viruel et al. 2007) by producing fullsib families similar in size to those used by Sharon et al. (1997) and focusing upon Guatemalan–Mexican hybrids. Recently, a large population of seedlings from a commercial grove inter-planted with two cultivars of opposite flowering types, ‘Tonnage’ (Type B) and ‘Simmonds’ (Type A), in approximately equal numbers was screened to determine the out-crossing rate in avocado under south Florida conditions (Borrone et al. 2008). Eight fully informative SSR markers identified 870 seedlings as progeny of a reciprocal cross between ‘Tonnage’ and ‘Simmonds’. Using these seedlings together with previously reported SSR markers and SSR markers developed from expressed sequence tags (EST-SSR), we have developed the first large mapping population and produced a linkage map for QTL discovery in avocado, focusing upon West Indian–Guatemalan hybrids. Materials and methods Plant material DNA isolation and identification of the open-pollinated progeny of a reciprocal cross between P. americana var. guatemalensis×americana cv. Tonnage and P. americana var. americana cv. Simmonds is described in Borrone et al. (2008). Linkage maps were constructed from 715 F1 individuals: 456 'Tonnage'×'Simmonds' progeny and 259 'Simmonds'×'Tonnage' progeny. SSR markers The parents, 'Tonnage' and 'Simmonds', and 96 of the 'Tonnage'×'Simmonds' progeny were screened with 261 SSR markers to determine segregation patterns. All primers used are listed in the Molecular Ecology Resources database (http://tomato.bio.trinity.edu) and are listed on the Subtropical Horticulture Research Station (SHRS) Plant Tree Genetics & Genomes (2009) 5:553–560 555 Sciences web site (http://www.shrstropicalcrops.org/static/ Appendix.htm). Their corresponding loci and alleles found in ‘Simmonds’ and ‘Tonnage’ are given in the supporting documentation (Appendix A). SSR markers developed from publicly available EST sequences (Borrone et al. 2007) were designated "SHRSPa" followed by a three-digit number. Numerous SSR markers were also developed and used that were not reported in Borrone et al. (2007). SSR markers previously developed by Sharon et al. (1997) were also used in this map, designated by the prefix “AV” followed by alpha-numeric characters, and aided in identifying corresponding linkage groups. Several primer pairs, including those previously reported by Sharon et al. (1997), produced additional amplification products differing in size from those previously reported. SSR primer pairs producing multiple marker loci were named using the primer name followed by ".1", ".2", etc. (e.g. SHRSPa100.1, SHRSPa100.2, etc.). Gene descriptions for EST-SSRs (Appendix A) mapped into the population were obtained by BLASTX analysis using BLAST2GO (Conesa et al. 2005). Primer pairs producing a heterozygous marker genotype in at least one parental genome were retained for map construction. Some primer pairs produced complex electrophoretic patterns, in which allelic assignments could not be made with confidence in terms of individual loci, and these markers were not used. Null alleles were identified for several markers, as reported previously in avocado (Sharon et al. 1997; Ashworth et al. 2004; Borrone et al. 2007, 2008) and these markers were retained. In some instances, an amplification product was present in one parent and entirely absent in the other, but clearly segregated in the progeny in a 1:1 ratio. These were scored as dominant loci and are designated with an “a” preceding the locus name (Appendix A). PCRs were conducted in 96-well plates as described in Borrone et al. (2008) with the following exceptions: PCRs were prepared both by hand and by using a MultiPROBE II PLUS HT EX Robot Liquid Handling System (PerkinElmer, Life Sciences, Downers Grove, IL, USA) and the postamplification ethanol precipitation step described was omitted. Instead, the amplifications were diluted up to 20 μL with sterile H2O and from 1 to 2 μL of the amplification reaction were added directly to a mixture of 19.8 μL of H2O and 0.2 μL of ROX-labeled GeneScan™-400HD molecular size standard (Applied Biosystems, Foster City, CA, USA). Capillary electrophoresis was performed on the ABI Prism® 3730 Genetic Analyzer using Performance Optimized Polymer 7 (POP 7, Applied Biosystems, USA) and analyzed using GeneMapper® v4.0 (Applied Biosystems). Linkage groups were based on independent LOD scores ranging from 4.0 to 10.0 using the regression algorithm with a maximum possible recombination frequency of <0.5 and Kosambi's mapping function. Sub-composite maps were first produced for both reciprocal crosses ('Tonnage'×'Simmonds' and 'Simmonds'×'Tonnage') and maps were also produced for both parental clones, carefully eliminating markers with high contributions to overall chi-square values, very high segregation distortion, and excessive missing data. A composite linkage map, referred to hereafter as the Florida F1 map, was generated for the Florida F1 population by combining reciprocal cross data (sub-composite maps), linkage group by linkage group, and also carefully eliminating markers with high contributions to overall chi-square values. Linkage groups were numbered in descending order according to the number of markers originally grouped together. Markers in common between this map and the map of Sharon et al. (1997) were used to identify corresponding linkage groups. Linkage analyses At a minimum LOD score of 4.0, 12 linkage groups (LG) representing the haploid set of 12 chromosomes of P. americana were generated from 163 markers. Descriptive data for the Florida F1 map are given in Table 1. Linkage Linkage analysis was done using JoinMap® 4.0 (Van Ooijen 2006) for a "CP" (cross-pollinated) population. Results and discussion SSR amplification The final map (Fig. 1, Table 1) was constructed from 163 markers generated by 135 primer pairs, 112 designed from EST-SSRs, and 23 SSR primers previously developed by Sharon et al. (1997). Twenty primer pairs amplified more than one locus, with 15 amplifying two loci, four amplifying three loci, and one amplifying six loci. 'Tonnage' was heterozygous for 93% of the total markers used in its sub-composite map, whereas 'Simmonds' was heterozygous for 39% of the total markers in its subcomposite map, comparable with the previous estimate for heterozygosity for each cultivar using 12 SSR markers (Borrone et al. 2008). The Florida F1 Map contained 16 fully informative markers, four polymorphic alleles between the two parents, and 103 partially informative markers. One hundred nineteen markers were scored on both subcomposite maps. An additional 43 markers were scored only on the 'Tonnage'×'Simmonds' progeny and one additional marker was scored only on the 'Simmonds'×'Tonnage' progeny (Appendix A). Table 2 contains the final number of markers mapped for each sub-composite map after marker quality control. Linkage analysis 556 Tree Genetics & Genomes (2009) 5:553–560 Fig. 1 Florida F1 composite map from the reciprocal cross between 'Tonnage' and 'Simmonds' avocado cultivars based on 163 SSR markers and 715 F1 individuals. Marker positions, given in cM from the Kosambi mapping function, and names are written on the left and right sides of each LG, respectively. Symbols denote markers with statistically significant segregation distortion beginning with two symbols denoting α≤0.05 and continuing to four symbols<0.005. Markers distorted in both sub-composite maps are denoted by asterisks (*), those in ‘Tonnage’בSimmonds’ are denoted by a cross (†), and those in ‘Simmonds’בTonnage’ are denoted by the double cross (‡) Table 1 Composite linkage map results from a combination of the reciprocal cross data Map LG 1 2 3 4 5 6 7 8 9 10 11 12 Total/average Florida F1 composite map, made from data from both sub-composite maps No. loci Length (cM) Average cM/locus 29 144.2 5.0 21 157.3 7.5 17 136.9 8.1 15 90.3 6.0 17 117.9 6.9 13 92.8 7.1 15 117.0 7.8 11 92.6 8.4 12 44.8 3.7 8 71.6 9.0 3 19.8 6.6 2 2.4 1.2 163 1,087.6 6.4 Map LG 1 2 3 4 No. loci Length (cM) cM/locus No. loci Length (cM) cM/locus 29 143.9 5.0 27 146.5 5.4 21 170.1 8.1 18 154.7 8.6 16 138.7 8.7 16 138.6 8.7 15 95.7 6.4 15 96.3 6.4 No. loci Length (cM) cM/locus 16 144.8 9.1 11 113.6 10.3 3 38.5 12.8 4 35 8.8 No. loci Length (cM) cM/locus 20 141.1 7.1 1a 11 69.1 6.3 18 141.1 7.8 16 132.2 8.3 11 115.5 10.5 13 79.6 6.1 8 92.6 11.6 14 124.6 8.9 0 – – 11 115.4 10.5 4 41.3 10.3 13 77.5 6.0 Tonnage×Simmonds 456 individuals Tonnage maternal Simmonds paternal Simmonds×Tonnage 259 individuals Simmonds maternal Tonnage paternal No. loci Length (cM) cM/locus No. loci Length (cM) cM/locus 1b 2 16.5 8.3 5 5a 12 117.3 9.8 5b 3 17.6 5.9 15 116.5 7.8 5c 2 29.3 14.7 5a 5b 3 19.9 6.6 5a 10 84.4 8.4 3 41.9 14.0 5b 2 14.8 7.4 6 103.6 17.3 10 106.3 10.6 7 8 9 10 11 12 Totala 13 93.7 7.2 12 92.1 7.7 15 119.6 8.0 13 121.1 9.3 11 95.3 8.7 11 95.4 8.7 12 46.6 3.9 11 47.0 4.3 8 71.3 8.9 8 70.9 8.9 3 23.7 7.9 3 23.8 7.9 2 3.1 1.6 2 3.1 1.6 162 1,119.0 7.5 151 1,106.0 7.1 7 100.8 14.4 5 68.6 13.7 0 – – 6 48.4 8.1 2 46.9 23.5 0 – – 0 – – 60 658.4 12.1 9 89.2 9.9 12 99.1 8.3 8 74.6 9.3 8 39.8 5 6 68.9 11.5 3 13.0 4.3 2 1.2 0.6 120 953.4 7.4 6 92.3 15.4 9 85.6 9.5 3 42.6 14.2 11 99.5 9.0 0 – – 8 74.6 9.3 5 44.3 8.9 8 37.1 4.6 0 – – 6 68.8 11.5 0 – – 3 13.0 4.3 0 – – 2 1.2 0.6 45 502.3 11.5 113 944.7 7.7 6 Tree Genetics & Genomes (2009) 5:553–560 Table 2 Mapping results Mapping results of the reciprocal cross (sub-composite maps) and parental maps a For LGs that contain parts a and b in bold, these are smaller-sized fragments of the entire LG and both were used to calculate the overall length of the linkage map. The exception is LG5a that represents the entire length of the LG in the ‘Tonnage’בSimmonds’ map. LG5a (not LG5b or LG5c) was used to calculate the overall length of the linkage map 557 558 groups ranged in size, longest to shortest, from 157.3 cM (LG2) to 2.4 cM (LG12), and the number of markers mapped per group ranged from 29 (LG1) to two (LG12). The total length of the Florida F1 map was 1,087.4 cM. This Florida F1 map is three times the size of the map reported by Sharon et al. (1997) and contains five times as many markers. Comparison of markers shared between the Florida F1 map and the F1 map of Sharon et al. (1997) allowed the identification of several analogous linkage groups; an example is given in Fig. 2 for LG2. Ten linkage groups of the 12 described by Sharon et al. (1997) contained SSR markers. Nine of the ten corresponded to seven linkage groups in the Florida F1 map (Appendix A). Distances between SSR markers within LGs reported by Sharon et al. (1997) and distances between these same markers placed in the sub-composite maps and the Florida F1 map corresponded well, which is remarkable given the disparity in the population sizes and the numbers of markers between the two maps. The lengths of the linkage groups found for the Florida F1 map correspond well with the physical description of the chromosomes reported by Garcia (1975). Garcia (1975) describes the chromosomes as being very small at metaphase with an asymmetric size distribution, ranging from 2.3 to 6.1 μm, and defined four groups based upon similarity of size. Three chromosomes were the largest, two others the same size but smaller, five yet smaller and indistinguishable from one another except that two of these chromosomes were highly heterochromatic, and two chromosomes were extremely small as compared with the others. A similar grouping of sizes is observed in the Florida F1 linkage map. In general, the ‘Tonnage’בSimmonds’ linkage groups were longer than the ‘Simmonds’בTonnage’ linkage groups. This can be Fig. 2 Comparison of marker order for LG 2 among the Florida F1 (composite) map, the parental maps for the 'Tonnage'×'Simmonds' cross, and location of LG 10 and 12 from Sharon et al. (1997). Lines between the LGs connect the common markers by their position Tree Genetics & Genomes (2009) 5:553–560 attributed to the larger number of markers mapped in the ‘Tonnage’בSimmonds’ sub-composite map versus its reciprocal (162 vs. 120), and to the larger number of individuals in this cross (456), slightly less than double that of the ‘Simmonds’בTonnage’ cross (Table 2). Colinearity was well-conserved between the reciprocal crosses with a few exceptions, for example, when two markers mapped closely in an isolated region. This was also reflected in the relatively small numbers of pairs of markers with significant heterogeneity of recombination frequency at the 5% level or lower between the sub-composite maps. LG1 had 18 total pairs out of all possible pairwise combinations; LG2, 11 pairs; LG6, 11 pairs and there were only very small numbers of pairs in other linkage groups. Most of the heterogeneity of recombination frequency between the sub-composite maps was due to the differential polymorphism between ‘Simmonds’ and ‘Tonnage’, and hence, the numbers of markers placed upon each of the subcomposite maps. Five linkage groups (LGs 3, 4, 8, 11, and 12) were composed almost entirely from markers polymorphic in the ‘Tonnage’ parent. The formation of linkage groups for all maps was straightforward except for a small number of markers, due to the high homozygosity of ‘Simmonds’. On LG11, out of three total markers, only a single marker, AVAG11, was informative in both parents while the other two markers segregated only in ‘Tonnage’ (SHRSPa009 and SHRSPa203). LG 5 formed three linkage groups in the 'Tonnage'×'Simmonds' map, and two linkage groups in the 'Simmonds'×'Tonnage' map (Table 2, Appendix A). This was due to the differential polymorphism exhibited between the parents, causing insufficient linkage among groups of markers only in the ‘Tonnage’בSimmonds’ sub-composite Tree Genetics & Genomes (2009) 5:553–560 map. The fragments observed in the ‘Simmonds’ × ‘Tonnage’ map were due to the lower number of markers mapped in this sub-composite map (Appendix A). The parts of this linkage group formed a single linkage group in the composite map with little heterogeneity of recombination frequency between markers on the sub-composite maps. The Florida F1 (composite) map contained a total of 50 markers exhibiting segregation distortion (α≤0.05), counting markers appearing in both maps only once. Only seven markers were significantly distorted in both maps (Fig. 1). The ‘Tonnage’בSimmonds’ map contained 21 significantly distorted markers that were not distorted in its reciprocal cross that mostly clustered on LG5, LG6, and LG11. The ‘Simmonds’בTonnage’ map contained 22 significantly distorted markers not distorted in the reciprocal cross, many of which were found in three clusters on LG4, LG8, and LG9. The only clusters of distorted markers common to both maps occurred on LG4 and LG5. Colinearity among markers was maintained in linkage groups when common markers exhibited segregation distortion in only one of the two sub-composite maps. For example, on LG11 segregation of all three markers was heavily distorted on the ‘Tonnage’בSimmonds’ map, yet no distortion was observed for any of the three markers in the ‘Simmonds’בTonnage’ map. The estimated length of LG11 in the ‘Tonnage’בSimmonds’ map, 23.7 cM, was more than double that of LG11 (13 cM) in the ‘Simmonds’× ‘Tonnage’ map (Table 2). By contrast, all distorted markers on LG9 were mapped in both crosses but were only distorted in the 'Simmonds'×'Tonnage' progeny. The four markers without segregation distortion were only mapped from the 'Tonnage'×'Simmonds' cross (Appendix A). There are several reasons to expect differences in segregation distortion in the reciprocal maps. It has been demonstrated that successful penetration of the ovary by pollen is time and temperature-dependent and varies among pollen donors (Sedgley and Grant 1983). Abscission of flowers and immature fruits occurs in avocado and selective abscission of immature fruit is reported to have a genetic basis (Degani et al. 1986, Degani et al. 1997). Two different types of fruit set are described in avocado, Type 1 and Type 2. 'Simmonds' typically sets a large crop with subsequent fruit drop (Type 1) while 'Tonnage' has a smaller initial fruit set with very little drop (Type 2) (Davenport 1982; M. Bass, personal communication). The marked differences observed in segregation distortion between the reciprocal crosses are perhaps related to pollen compatibility or fruit set differences. The two parental cultivars 'Simmonds' and 'Tonnage' differ for many phenotypic traits useful in humid lowland environments like south Florida. 'Simmonds' is of the West Indian race and believed to be a seedling of 'Pollock'. It was selected in south Florida and first propagated commercially 559 in 1921 and is still a major commercial cultivar some 85 years later. 'Simmonds' has a light green, oblong-oval to pear-shaped and large fruit with medium size seed, a low oil content (3–6%), imparts tolerance to PRR in its progeny, has the A flowering type, has high yielding ability, and is an early season (June–July) cultivar (Campbell and Malo 1978; Ploetz et al. 2002). West Indian selections were the only important commercial cultivars in Florida until the 1920s when competition from Cuba depressed the market for Florida avocados. A number of Guatemalan–West Indian hybrids had been selected which ripened in the fall and winter extending the season and plantings shifted to include these hybrids. One of these was the cultivar 'Tonnage' which is now considered a minor cultivar in south Florida. 'Tonnage', a seedling of ‘Taylor’, was first propagated commercially in 1930. It has dark green, pearshaped fruit with medium seed size, moderate oil content (8–15%), does not impart tolerance to PRR in its progeny, is a B flowering type, and is a late season (August– September ) cultivar (Ploetz et al. 2002). The number of SSR markers now available for avocado has allowed the development of this first moderately dense map composed of 163 loci. Linkage maps developed using molecular markers, such as the one reported here, enable the detection and use of QTLs affecting traits of economic importance. The F1 population, along with 20 clonal plants each of 'Simmonds' and 'Tonnage' were planted in the field at the SHRS in April 2007. Phenotypic data collection began in 2008 and will continue for the next five years. Additional markers, both SSR and single nucleotide polymorphism, will be placed on the linkage map as they are developed. Our goal is to identify QTLs for the aforementioned horticultural traits and use these in a marker-assisted selection (MAS) program similar to the successful approach we have taken with cacao (Schnell et al. 2007). References Arumuganathan K, Earle ED (1991) Nuclear DNA content of some important plant species. Plant Mol Biol Rep 9:208–218 Ashworth VETM, Kobayashi MC, De La Cruz M, Clegg MT (2004) Microsatellite markers in avocado (Persea americana Mill.): development of dinucleotide and trinucleotide markers. Scientia Hort 101:255–267 Ashworth VETM, Chen H, Clegg MT (2007) 17 Avocado. In: Kole C (ed) Genome mapping and molecular breeding in plants, fruits and nuts, vol 4. Springer, Berlin, pp 325–329 Borrone JW, Schnell RJ, Violi HA, Ploetz RC (2007) Seventy microsatellite markers from Persea americana Miller (avocado) expressed sequence tags. Mol Ecol Notes 7(3):439–444 Borrone JW, Olano CT, Kuhn DN, Brown JS, Schnell RJ, Violi HA (2008) Outcrossing in Florida avocados as measured using microsatellite markers. J Amer Soc Hort Sci 133:255–261 560 Campbell CW, Malo SE (1978) Review of methods for measuring avocado maturity in Florida. Proc Amer Soc Hort Sci Trop Reg 22:16–26 Chanderbali AS, Albert VA, Ashworth VETM, Clegg MT, Litz RE, Soltis DE, Soltis PS (2008) Persea americana (avocado): bring ancient flowers to fruit in the genomics era. BioEssays 30:386–396 Chen H, Ashworth VETM, Xu S, Clegg MT (2007) Quantitative analysis of growth rate in avocado. J Amer Soc Hort Sci 132 (5):691–696 Conesa A, Götz S, García-Gómez JM, Terol J, Talón M, Robles M (2005) Blast2GO: a universal tool for annotation, visualization and analysis in functional genomics research. Bioinformatics 21 (18):3674–3676 Darlington CD, Wylie AP (1945) Chromosome atlas of flowering plants. George Allen & Unwin LTD, London, p 16 Davenport TL (1982) Avocado growth and development. Proc Florida State Hort Soc 95:92–99 Degani C, Goldring A, Gazit S, Lavi U (1986) Genetic selection during the abscission of avocado fruitlets. HortScience 21(5):1187–1188 Degani C, El-Batsri R, Gazit S (1997) Outcrossing rate, yield, and selective fruit abscission in ‘Ettinger’ and ‘Ardith’ avocado plots. J Amer Soc Hort Sci 122(6):813–817 Degani C, Lahav E, El-Batsri R, Gazit S (2003) Caging single avocado trees with a beehive does not guarantee exclusive formation of selfed progeny. HortScience 38(7):1433–1434 Garcia AV (1975) Cytogenetic studies in the genus Persea (Lauraceae). I. Karyology of seven species. Can J Genet Cytol 17(2):173–180 Lammerts WE (1942) Progress report on avocado breeding. Calif Avocado Soc 1942 Yrbk 27:36–41 Lavi U, Lahav E, Genizi A, Degani C, Gazit S, Hillel J (1991) Quantitative genetic analysis of traits in avocado cultivars. Plant Breeding 106:149–160 Tree Genetics & Genomes (2009) 5:553–560 Lavi U, Lahav E, Degani C, Gazit S (1993) Genetics of skin color, flowering group, and anise scent in avocado. J Heredity 84 (1):82–84 Nakasone HY, Paull RE (1999) Tropical fruit, ch. 4. CABI, New York, pp 76–102 Ploetz R, Schnell RJ, Haynes J (2002) Variable response of openpollinated seedlings progeny of avocado to Phytophthora root rot. Phytoparasitica 30(3):262–268 Schnell RJ, Brown JS, Olano CT, Power EJ, Krol CA, Kuhn DN, Motamayor JC (2003) Evaluation of avocado germplasm using microsatellite markers. J Amer Soc Hort Sci 128(6):881–889 Schnell RJ, Kuhn DN, Brown JS, Olano CT, Phillips-Mora W, Amores FW, Motamayor JC (2007) Development of a marker assisted selection program for cacao. Phytopathology 97 (12):1664–1669 Scora RW, Bergh B (1990) The origins and taxonomy of avocado (Persea americana) Mill. Lauraceae. Acta Hort 275:387–394 Sedgley M, Grant WJR (1983) Effect of low temperatures during flowering on floral cycle and pollen tube growth in nine avocado cultivars. Scientia Hort 187:207–213 Sharon D, Cregan PB, Mhameed S, Kusharska K, Hillel J, Lahav E, Lavi U (1997) An integrated genetic linkage map of avocado. Theor Appl Genet 95:911–921 USDA FAS Market News (2008) Avocado Update. http://www.fas. usda.gov/htp/horticulture/Avocados/AvocadoArticle2008.pdf. Cited 1 August 2008 Van Ooijen JW (2006) JoinMap® 4, software for the calculation of genetic linkage maps in experimental populations. Kyazma BV, Wageningen, Netherlands Viruel MA, Gross-Germann E, Barceló A (2007) Desarrollo de un mapa genético con marcadores SSRs y AFLPs en aguacate. VI World Avocado Congress, Viña del Mar, Chile, Nov 12–16 2007