ORIGINAL ARTICLES
Development of DNA Markers for Interspecific and Intersubgeneric Hybrids Involved in Tree Peony Cultivar Groups
2022 Volume 91 Issue 3 Pages 382-387
Details
2022 Volume 91 Issue 3 Pages 382-387
Tree peony (Paeonia spp.) is a highly ornamental shrub originally from China. Its cultivars are classified into four groups: Chinese, Japanese, French, and American, based on morphological characteristics and geographic location. The French and American cultivars were developed by crossing the wild species P. lutea Delavay ex Franch. and P. delavayi Franch. with Japanese and Chinese cultivars. In this study, we evaluated phenetic relationships among major cultivar groups in the genus Paeonia L., using a dendrogram reconstructed from band pattern data for random amplified polymorphic DNA (RAPD) markers developed in a previous study. We also developed two sequence tagged site (STS) markers using RAPD markers specific to the French and American tree peony cultivar groups and an intersubgeneric hybrid. The former is an LuDeB marker able to detect the genetic relationship with P. lutea and P. delavayi, and the latter is an HPB marker able to identify the cpDNA of herbaceous peony in intersubgeneric hybrid cultivars between herbaceous and tree peony. The possession of bands obtained in this study also supported the breeding history of cultivar development in the genus Paeonia.
Tree peony (Paeonia spp.) is a popular ornamental flowering shrub originally from China (Hosoki et al., 1991). There are about 1,500 cultivars in the world (Zhang et al., 2012), and the cultivars are classified based on morphology and geographical origin into four groups: Chinese, Japanese, French, and American (Liu, 2003; Wang et al., 2014). In Japan, tree peony has been cultivated since the Heian period (794–1185) at the latest (Hosoki et al., 1991). Japanese and Chinese cultivar groups have been developed independently in each country based on intraspecific variations of P. suffruticosa Andr. Japanese and Chinese cultivars, which have pink, red, purple, and white flowers, were introduced to Europe and North America in the 18th and 19th centuries, respectively (Hao et al., 2013). French and American cultivars were bred by crossing Japanese and Chinese cultivars with the wild species P. lutea or P. delavayi, which have genes for the yellow color (Hosoki et al., 1991). Almost all French and American cultivars produce chalcone, a yellow pigment trait originating from P. lutea or P. delavayi (Hosoki et al., 1991). Chalcone production results in yellow to reddish-orange flowers that are not found among Japanese and Chinese cultivars (Hosoki et al., 1991). Thus, French and American cultivars are popular for their unique flower colors (Hao et al., 2013) and, in addition, develop small flowers with drooping stalks. The Daphnis hybrid cultivar group was bred by backcrossing Japanese cultivars with American cultivars (Cheng, 2007), which produces an upright flower in a variety of colors.
Tree peony is classified in section Moutan DC. in the genus Paeonia L. (Paeoniaceae). In addition to this group of woody species, two herbaceous sections in the genus are recognized, namely section Paeonia and section Onaepia Lindl. (Stern, 1943; Zhao et al., 2008). Crosses between herbaceous and tree peonies usually yield almost no viable seeds; nevertheless, more than 140 hybrids between herbaceous and tree peony cultivars have been developed by the Itoh crossing method, which obtains intersubgeneric hybrids between herbaceous peony as a seed parent, and progeny of the P. lutea (Lutea hybrid) as a pollen parent (Yang et al., 2020; Sun et al., 2021). Recently, Hong (2021) divided Paeonia into subgenus Moutan for woody plants and subgenus Paeonia for herbaceous plants. Thus, the cross between P. lactiflora Pall (herbaceous peony) and tree peony was revised as an intersubgeneric cross. In intersubgeneric hybrids between herbaceous and tree peony cultivars, known as Itoh hybrids, the flower and leaf types are similar to those of tree peony cultivars, whereas the growth characteristics and flowering periods are similar to herbaceous peony cultivars. Most intersubgeneric hybrids show strong adaptability and produce abundant flowers with rich flower colors (Ma et al., 2012; Yang et al., 2020). The introduction of yellow pigment from P. lutea and P. delavayi and the breeding of intersubgeneric hybrids increased the range of flower colors and variety of peony cultivars. Such hybrids are highly influential in cultivar development in the genus Paeonia.
It is important to clarify the phylogenetic relationships and breeding process of each cultivar group for selection of parents to use in future breeding. We identified more than 350 peony cultivars using 48 random amplified polymorphic DNA (RAPD) markers developed in a previous study (Mochida et al., 2020). In the previous study, RAPD bands specific for the French and American cultivar groups and intersubgeneric hybrids were determined by comparison of band patterns between cultivar groups. We also identified bands specific for the intersubgeneric hybrid ‘Oriental Gold’.
In this study, a dendrogram was constructed using RAPD data from tree peony cultivar groups, the wild species P. lutea and P. delavayi, herbaceous peony cultivars, and intersubgeneric hybrids between herbaceous and tree peony cultivars. The phenetic distance and relationships of each cultivar group were evaluated. In addition, DNA markers were developed to determine bands specific to the French and American tree peony cultivar groups and intersubgeneric hybrids. Using the markers, we investigated the possession of bands specific to the wild species (P. lutea and P. delavayi) and to herbaceous peonies.
In addition to the 353 cultivars used in the previous study (Mochida et al., 2020), two wild species (P. lutea and P. delavayi), four herbaceous peony cultivars, and 12 intersubgeneric hybrids of herbaceous and tree peony cultivars were collected from the tree peony resources collection of the Botanic Gardens of Toyama, Shimane prefectural flower garden (Shimane-hananosato), and the Plant Breeding Laboratory of the Faculty of Life and Environmental Sciences of Shimane University (Table 1). For the two wild species, DNA was extracted from the leaves using the method as described in Kobayashi et al. (1998). For the other cultivars, either DNA was extracted from winter buds using a modified CTAB method (Mochida et al., 2020) or DNA extracted in the previous study was used.
Wild species and cultivars of genus Paeonia used in this study and their band patterns in DNA markers.
Phenetic relationships among the two wild species and 56 cultivars, comprising herbaceous peonies, intersubgeneric hybrids between herbaceous and tree peony cultivars, and Japanese, Chinese, American, and French cultivars, were analyzed using the 48 RAPD markers developed previously (Mochida et al., 2020). Data for the newly collected cultivars were combined with the data for cultivars from the previous report. To construct a dendrogram, a data matrix for the presence or absence of bands (scored as 1 and 0, respectively) for each marker was generated. The squared Euclidean distance between each species and cultivar was calculated, and a hierarchical clustering analysis by Ward’s method was performed using the Black-Box statistical analysis package <http://aoki2.si.gunma-u.ac.jp/BlackBox/BlackBox.html>.
STS primers using cultivar groups and cultivar-specific RAPD bandsWe attempted to make STS markers based on the two RAPD markers specific to the cultivar or the cultivar groups obtained in the previous report (Mochida et al., 2020). In the previous analysis, the band generated by the RAPD marker OPC14c_1100 was detected specifically in the French and American cultivar groups, Daphnis hybrids, and ‘Oriental Gold’. The marker OPA11_1500 represented a 1500 bp band detected in almost all peony cultivars, but no band was detected specifically in ‘Oriental Gold’, which is an intersubgeneric hybrid between herbaceous and tree peony cultivars. For the marker OPA11_1500, we aimed to create STS primers based on the nucleotide sequence of the band detected in Japanese cultivars. The OPC14c_1100 band of the American cultivar ‘Banquet’ and the OPA11_1500 band of the Japanese cultivar ‘Kayūsen’ were each excised from an agarose gel and purified using the LaboPass Gel Extraction and Purification Kit (Hokkaido System Science Co., Ltd., Sapporo, Japan). The DNA was cloned using the Mighty TA-cloning Kit (Takara Bio Inc., Shiga, Japan). The plasmid DNA was purified from Escherichia coli with the confirmed inserts using the Xprep Plasmid DNA Mini Kit (PhileKorea, Inc., Daejeon, Republic of Korea). The DNA sequence was outsourced to Hokkaido System Science <http://www.hssnet.co.jp>. Based on the obtained DNA sequences, the STS primers were named LuDeB (Lutea and Delavayi Band) to detect P. lutea and P. delavayi bands, and named HPB (Herbaceous Peony Band) to detect herbaceous peonies or tree peonies bands, and designed using Primer3 Plus <http://primer3plus.com/> (Table 2). A homology search was performed with the BLAST tool of the National Center of Biotechnology Information (NCBI; <https://www.ncbi.nlm.nih.gov/>) using the obtained DNA sequences as queries.
Sequences of STS primers developed in this study.
Two wild species and 78 cultivars were analyzed using the developed DNA primers. The PCR amplifications were performed in 10 μL reaction mixtures containing 50 ng gDNA, 1× reaction buffer, 2 mM MgCl2, 0.2 mM dNTPs, 0.25 U of TaKaRa Ex Taq (Takara Bio), and 1 μM of each STS primer. The reaction conditions for each amplification were as follows: preheating at 94°C for 3 min, 35 cycles of denaturation at 94°C for 1 min, annealing at 61°C for 1 min, extension at 72°C for 2 min, and final extension at 72°C for 4 min. The reactions were performed on a TP600 thermal cycler (Takara Bio). The fragments were separated by electrophoresis in 2% agarose gel and 1× TAE buffer supplemented with 0.25 μg·mL−1 ethidium bromide at 100 V for 50 min (LuDeB) or 35 min (HPB), and photographed under ultraviolet light.
Based on the band patterns generated using the 48 RAPD markers, a dendrogram for the two wild species and 56 cultivars was constructed (Fig. 1). Using 48 RAPD markers, it was possible to distinguish among all accessions, including newly added wild species, intersubgeneric hybrids and herbaceous peonies, except for two sets of bud mutations. The dendrogram comprised two large clusters. Cluster I consisted of cultivars of P. suffruticosa raised in Japan and China. Cluster II included the wild species, French and American tree peony cultivars, herbaceous peony cultivars, intersubgeneric hybrids between herbaceous and tree peony cultivars, and Daphnis hybrids.
Result of cluster analysis based on the presence or absence of the 48 RAPD markers and two STS primers using Paeonia samples. + or − indicate presence or absence of LuDeB and HPB marker band, respectively.
Using STS primer LuDeB originated from RAPD marker OPC14c_1100, a distinct band of 263 bp was detected in the wild species P. lutea and P. delavayi, French and American cultivars, and cultivars genetically related to P. lutea and P. delavayi. No band of 263 bp was detected among the Japanese and Chinese cultivars, and herbaceous peony cultivars (Table 1; Fig. 2). The presence or absence of the band with the LuDeB marker was consistent with the analysis result of the RAPD marker OPC14c_1100.
Results of the PCR amplifications using the developed STS primers: (A) LuDeB, (B) HPB. M: size marker, 1: P. lutea, 2: P. delavayi, 3: Kakōden, 4: Flame, 5: Suigetsu, 6: Kinkaku, 7: Kinkō, 8: Kintei, 9: High Noon, 10: Vesuvian, 11: Mystery, 12: Ōkan, 13: Toribute, 14: Shakudō-no-kagayaki, 15: Oriental Gold, 16: Bartzella, 17: Kopper Kettle, 18: Hanakisoi, 19: Hōju, 20: Kijōden, 21: Gyokurōshun, 22: Hiō, 23: Jurō; The black arrow indicates estimated band size.
Using STS primer HPB originated from RAPD marker OPA11_1500, a distinct band of 502 bp was detected in the wild species, French, American, Japanese, and Chinese cultivars, and Daphnis hybrids. In the four herbaceous peony cultivars and 13 intersubgeneric hybrids, a distinct band of 395 bp was detected, which was 100 bp shorter than the assumed band size (Table 1; Fig. 2). The presence or absence of the tree peonies specific 502 bp band detected in HPB marker analysis was consistent with the results of the RAPD marker OPA11_1500.
All the wild species and cultivars tested were classified into four groups: two groups based on the presence or absence of the LuDeB band, and two groups based on the band size generated by the HPB primer (Table 1). All cultivars of intersubgeneric hybrids were characterized by the LuDeB band and the 395 bp HPB band. All species, French and American cultivars, and Daphnis hybrids analyzed in this study showed the presence of the LuDeB band and the 502 bp HPB band. All herbaceous peony cultivars tested in this study lacked the LuDeB band and had the 395 bp HPB band. All Japanese and Chinese cultivars analyzed lacked the LuDeB band and had the 502 bp HPB band.
Homology search of LuDeB and HPB markersNo highly homologous sequence was detected in a homology search of the NCBI database using the nucleotide sequence of the band for the RAPD marker OPC14c_1100, which was used to develop the STS primer LuDeB. However, the nucleotide sequence of the band for the RAPD marker OPA11_1500, which was used to develop the STS primer HPB, was highly homologous (98%; 860/874) with chloroplast DNA nucleotide sequences of tree peony (P. suffruticosa chloroplast MH793271).
Since newly added wild species, intersubgeneric hybrids and herbaceous peonies have been identified, RAPD markers have been developed that are useful in the genus Paeonia (Fig. 1). Based on the presence or absence of the band for the DNA marker LuDeB, P. lutea, P. delavayi, French and American cultivars, and Lutea and Delavayi hybrids roughly formed one group, and herbaceous peony, Japanese, and Chinese cultivars were roughly aggregated into a separate group. These two groups were further divided into two groups according to the band size amplified by the DNA marker HPB, and thus the accessions were categorized into four groups. The HPB marker bands discriminated between herbaceous and tree peonies in a group of herbaceous peonies and intersubgeneric hybrids, and a group of wild species and tree peony cultivars that were not involved in the development of herbaceous peony cultivars. The Japanese and Chinese cultivar groups were categorized into one group in the classification based on DNA markers, consistent with the cluster analysis of the RAPD marker dataset, suggesting that these cultivar groups are genetically distantly related to the other cultivar groups. That the Japanese and Chinese cultivar groups derived from P. suffruticosa are genetically distantly related to foreign cultivar groups, including P. lutea and P. delavayi, is consistent with the morphological classification reported by Hamada et al. (1989). It is also consistent with the genetic classifications reported by Hosoki et al. (1997) and Zhang et al. (2012). In addition, herbaceous peonies and intersubgeneric hybrids, which are considered to be genetically distantly related to the peonies, were also classified into Cluster II. The reason for this was thought to be that the 48 RAPD markers were specialized in identifying tree peony cultivars, and no common bands in tree peonies or herbaceous peonies were selected.
Paeonia lutea and P. delavayi are used as parents in tree peony hybridization, and have contributed to the breeding of yellow-flowered cultivars and the development of unique floral colors, such as orange and chestnut, among peonies. Using the STS primer LuDeB, a common band was detected in the wild species P. lutea and P. delavayi, French and American cultivars for which either species was a parent, and Lutea and Delavayi hybrids. The band amplified by the STS primer LuDeB was also detected in the intersubgeneric hybrid ‘Oriental Gold’, which was raised from a cross between a herbaceous peony (as the seed parent) and the French cultivar ‘Kinkō’ as the pollen parent. The LuDeB band was specifically detected in P. lutea, P. delavayi, and their hybrid progeny, and was not detected among Japanese and Chinese cultivars in which P. lutea and P. delavayi were not involved in their pedigree (Hao et al., 2013). Therefore, it was concluded that the STS primer LuDeB was capable of detecting gene transfer from the wild species P. lutea and P. delavayi. Thus, the DNA marker LuDeB was useful to identify the genetic contribution of P. lutea and P. delavayi to yellow-flowered cultivar development.
Intersubgeneric hybrids are of high utility because they combine characteristics of both herbaceous and tree peonies, including high ornamental value, good disease resistance, strong growth potential (Sun et al., 2021), a flowering period similar to that of herbaceous peonies, strong adaptability, and abundant flowers (Yang et al., 2020). Based on the nucleotide sequence of herbaceous peony chloroplast DNA (P. lactiflora chloroplast MK860971) and sequences of HPB primers, the assumed band size amplified by the HPB primer was 395 bp for herbaceous peonies. The two bands (502 bp and 395 bp) amplified by the STS primer HPB were consistent with chloroplast DNA sequences as determined from the homology search. The size difference in the two bands obtained by the HPB primer in the same region may indicate the 107 bp deletion in herbaceous peonies based on the chloroplast DNA sequence information. The 502 bp band was consistent with tree peony-type chloroplast DNA and was specifically detected in tree peonies, whereas the 395 bp band was consistent with herbaceous peony-type chloroplast DNA and was specifically amplified in herbaceous peony cultivars. Given that the two bands amplified by the STS primer HPB are derived from chloroplast DNA and are inherited maternally (Kobayashi et al., 2013), the bands can be used to estimate the seed parent of intersubgeneric hybrids.
The introduction of yellow pigment greatly enhanced variation in flower color, and intersubgeneric crosses with herbaceous peony gave rise to cultivars harboring characteristics of both tree and herbaceous peonies. P. lutea and P. delavayi have been involved in the breeding of French and American cultivars, and in the breeding of intersubgeneric hybrids of herbaceous and tree peony cultivars, for which the seed parent was herbaceous peony and the pollen parent was Lutea hybrid. The present results obtained with the two DNA markers developed in this study were consistent with these observations. The results of DNA markers are consistent with breeding history, in which P. lutea or P. delavayi contributed to the development of French and American cultivars, and acted as a bridge plant in the generation of intersubgeneric hybrids between herbaceous and tree peony cultivars. In the development of tree peony cultivars, interspecific crosses and intersubgeneric crosses were required to introduce new traits. However, the frequency of viable seeds in interspecific or intersubgeneric crosses is extremely low. Moreover, tree peonies require a considerable number of years to bloom from seed. Given that the DNA markers developed in this study are new markers for determining gene transfer of a species, they may be used for marker selection for interspecific and intersubgeneric hybrids.
In conclusion, we developed two DNA markers useful for peony breeding: the LuDeB marker, which is able to detect the genetic relationship with P. lutea and P. delavayi; and the HPB marker which is able to identify the cpDNA of herbaceous peony in intersubgeneric hybrid cultivars between herbaceous and tree peony. The possession of bands obtained in this study also elucidates the breeding history of cultivar development in the genus Paeonia.
We are grateful to the Botanic Gardens of Toyama for providing material of the wild species P. lutea (BGT30226), and Shimane-hananosato for providing some of the intersubgeneric hybrids. We thank Yuji Kurashige, Principal of the Niigata Prefectural Botanical Garden, for fruitful discussions.
