Sexual and Apomictic Seed Reproduction in Aronia Species with Different Ploidy Levels

Abstract

The genus Aronia Medik., also known as chokeberry, is a group of deciduous shrubs in the Rosaceae family, subtribe Pyrinae. The four commonly accepted species include A. arbutifolia (L.) Pers., red chokeberry; A. melanocarpa (Michx.) Elliott, black chokeberry; A. prunifolia (Marshall) Reheder, purple chokeberry; and A. mitschurinii (A.K. Skvortsov & Maitul). Wild and domesticated Aronia species are found as diploids, triploids, and tetraploids. Genetic improvement of polyploid Aronia genotypes has been limited by suspected apomixis, which may be widespread or distinct to tetraploids. The objectives of this study were to elucidate the reproductive mechanisms of Aronia species and reveal the occurrence of apomixis within the genus and along ploidy lines. Twentynine Aronia accessions [five A. melanocarpa (23), five A. melanocarpa (43), eight A. prunifolia (33), four A. prunifolia (43), six A. arbutifolia (43), and one A. mitschurinii (43)] were used in this study. Intra-accession variability was evaluated by growing out progeny from each open-pollinatedmaternal accession and comparing plant phenotypes, ploidy levels, and amplified fragment length polymorphism (AFLP) marker profiles between the progeny andmaternal accession. Progeny of diploid and tetraploid maternal plants had ploidy levels identical to maternal plants, except for UC009 (A. melanocarpa, 23) which produced a mix of diploids and tetraploids. UC143 and UC149 (A. prunifolia, 33) produced all triploid offspring, whereas all other triploid accessions produced offspring with variable ploidy levels including 23, 33, 43, and 53. Pentaploid Aronia has not been previously reported. Diploid accessions produced significant AFLP genetic variation (0.68–0.78 Jaccard’s similarity coefficient) in progeny, which is indicative of sexual reproduction. Seedlings from tetraploid accessions had very little AFLP genetic variation (0.93–0.98 Jaccard’s similarity coefficient) in comparison with their maternal accession. The very limited genetic variation suggests the occurrence of limited diplosporous apomixis with one round of meiotic division in tetraploid progeny. Triploid accessions appear to reproduce sexually or apomictically, or both, depending on the individual. These results support our understanding of Aronia reproductive mechanisms and will help guide future breeding efforts of polyploid Aronia species. Native to eastern regions in North America, the genus Aronia is a group of deciduous shrubs in the Rosaceae family, subtribe Pyrinae. The Pyrinae subtribe has a base chromosome count of n = 17 (Postman, 2011), and Aronia species are commonly found as diploids (2n = 2x = 34) and tetraploids (2n = 4x = 68) with some occurrence of triploids (Brand, 2010; Hovmalm et al., 2004; Leonard et al., 2013). Polyploidization events in Aronia likely resulted from the fusion of unreduced gametes from the same or different species to produce autopolyploids and allopolyploids, respectively. The three commonly accepted Aronia species include A. arbutifolia (L.) Pers., red chokeberry (tetraploid); A. melanocarpa (Michx.) Elliott, black chokeberry (diploid and tetraploid); and A. prunifolia (Marshall) Reheder, purple chokeberry (triploid and tetraploid). A fourth species of Aronia has been recognized as A. mitschurinii [(A.K. Skvortsov & Maitul) tetraploid], and it is used in commercial fruit production, usually as the cultivars Viking or Nero (Leonard et al., 2013). However, genetic marker and phenotypic data suggest that nearly all A. mitschurinii cultivars, an intergeneric hybrid involvingA. melanocarpa · Sorbus aucuparia L. (Leonard et al., 2013; Skvortsov and Mautulina, 1982), are genetically identical and likely renames of a single genotype (J.D. Mahoney and M.H. Brand, unpublished data). Interest in Aronia is high because their fruits contain high levels of antioxidants and polyphenols (Brand et al., 2017; Wu et al., 2004; Zheng and Wang, 2003), they are valuable as replacements for invasive exotic ornamental plants (Brand, 2010), and they are widely adapted to various geographic regions (Dirr, 2009; McKay, 2001). Aronia flowers are thought to be protogynous and self-compatible (Connolly, 2014). Polyploid Aronia species have been reported to reproduce apomictically, via gametophytic apomixis, resulting in embryos that are identical or nearly identical to maternal plants (Brand, 2010). Hovmalm et al. (2004) reported that diploid A. melanocarpa produced highly heterogeneous offspring and tetraploid plants produced homogeneous offspring, suggesting that polyploid Aronia reproduce apomictically. Gametophytic apomixis occurs when a progenitor cell in the megasporangium forms a megagametophyte (Grossniklaus et al., 2001; Richards, 2003). Gametophytic apomixis is further classified into two categories: diplospory and apospory. In diplosporous apomictic plants, the megagametophyte forms from an unreduced or partially reduced megaspore. When a partially reduced megaspore is involved, meiosis is initiated but fails before completion and cell division continues mitotically (Bicknell and Koltunow, 2004). The result is an unreduced megagametophyte derived from a megaspore in which homologous recombination and one round of segregation may have occurred. Apospory refers to an unreduced megagametophyte arising from nucellar or integument tissue (Koltunow and Grossniklaus, 2003). Talent (2009) mentions that pseudogamous gametophytic apospory is common in the Maloid Rosaceae (Pyrinae), where seed development requires pollination, but the embryo has no paternal inheritance and only the endosperm is fertilized. Both diplospory and apospory have been reported to occur in the same species, including the Pyrinae genera Crataegus (Muniyamma and Phipps, 1979, 1984a, 1984b) and Sorbus (Jankun and Kovanda, 1988). In normal sexual reproduction, genetic uniformity and hybrid vigor are lost after the F1 generation, but with apomixis these traits can be maintained through many generations due to a fixed heterozygosity (Koltunow et al., 1995; Ortiz et al., 2013; Richards 2003). For this reason, seed propagation of apomictic selections is possible and allows growers to achieve high yields while avoiding more expensive vegetative propagation methods (Barcaccia and Albertini 2013). In apomictic temperate fruit crops, it is advantageous to use vegetative propagation from mature phase plants rather than regenerate from juvenile seed. Aronia Received for publication 26 Nov. 2018. Accepted for publication 24 Jan. 2019. This research was partially supported by the U.S. Department of Agriculture Multistate Hatch NC007 Plant Germplasm and Information Management and Utilization. Corresponding author. E-mail: mark.brand@ uconn.edu. 642 HORTSCIENCE VOL. 54(4) APRIL 2019 requires 3 to 5 years before reaching the mature phase for flowering to occur, so use of apomictic seeds delays fruit production. However, Mahoney et al. (2018) report that cotyledons have a greater shoot regeneration rate than mature phase leaf explants; therefore, it may be advantageous to use apomictic seed tissue (i.e., cotyledons) for genetic transformation of Aronia. Although apomixis can be an advantage, it also can present challenges to controlled breeding and genetic exchange. For example, genetic improvement of polyploid Aronia genotypes has been hindered by the occurrence of apomixis during attempted crosses. A number of molecular marker techniques such as random amplified polymorphic DNA (RAPD), intersimple sequence repeat, and cpDNA marker analysis have been popular in identifying apomixis in plants (Arnholdt-Schmitt, 2000; Hovmalm et al., 2004; Ludwig et al., 2013; Robertson et al., 2010; Smolik et al., 2011). AFLP analysis often has been preferred over other molecular methods for its efficiency (Leonard et al., 2013; Lubell et al., 2008; Obae and Brand, 2013). The AFLP technique has proven to be a more cost-effective way of producing a large number of markers (Mueller and Wolfenbarger, 1999). In this study, we use AFLP, in conjunction with ploidy analysis and plant phenotype, to elucidate the reproductive mechanisms of Aronia species and reveal the occurrence of apomixis within the genus and among ploidy levels. Materials and Methods Plant material. Twenty-nine Aronia accessions [five A. melanocarpa (2·), five A. melanocarpa (4·), eight A. prunifolia (3·), six A. prunifolia (4·), four A. arbutifolia (4·), and one A. mitschurinii (4·)] were used in this study as maternal genotypes (Table 1). Plants were maintained at the University of Connecticut Research Farm in Storrs, CT. The maternal accessions were grown in a randomized field planting consisting of 120 Aronia accessions (with three replicates) and representing all four species and various ploidy levels. Intra-accession progeny variability was evaluated by growing out seedlings from their open-pollinated maternal accessions. Each maternal plant had the opportunity to be pollinated by any other accession in the field collection. Seeds were collected from a single maternal accession plant of each by cleaning them from the fruits and air drying them before placement in cool, dark storage (13 ± 2 C, relative humidity 55 ± 5%) until further use. Seeds were cold stratified in moist sand for 90 d in 50-mL conical centrifuge tubes at 5 C. Stratified seeds were sowed in 32-oz. clear plastic salad trays containing potting medium with a ratio of 5:3:1 screened composted pine bark, sphagnum peatmoss, and sand and placed under cool white fluorescent light (80 mmol·m·s) at 24 C. Seedlings were transferred to standard 50-cell plug trays with the same 5:3:1 mix and eventually transferred to 1-gallon pots. Four clonal softwood stem cuttings from A. mitschurinii (4·) were rooted in mid-June and served as a control. Phenotypic observations. Within accessions, maternal plants and 2-year-old progeny were compared for overall plant form or habit, branching structure, leaf shape and size, phyllotaxy, degree