Introduction of Floral Scent Traits into Asiatic Hybrid Lilies (Unscented) by Crossbreeding with Lilium cernuum (Scented)

Abstract

Although a strong floral scent is typical of hybrid lilies, Asiatic hybrid lilies (Lilium spp.) have only a weak scent. Therefore, developing new cultivars with pleasant floral scents is an important objective in Asiatic hybrid lily breeding programs. Among the wild species contributing to Asiatic hybrid lily establishment, L. cernuum has a characteristic scent that the nose can perceive. We crossed Asiatic hybrid lily cultivars twice with L. cernuum to introduce floral scent traits into Asiatic hybrid lilies, and scent emission in the F₁ plants and their progeny lines was evaluated instrumentally. Flowers of L. cernuum emitted benzenoids/phenylpropanoids (mainly 2-phenylethanol and benzaldehyde), monoterpenes (predominantly linalool and linalool oxides), and a fatty acid (iso-valeric acid). Lilium cernuum gave off unpleasant odors mainly because of iso-valeric acid emission. The F₁ flowers also emitted these three chemical classes. Among the lines derived from the crosses between Asiatic hybrid lily cultivars and the F₁ plant, lines that emitted all three chemical classes and lines that released one or two of the three chemical classes were segregated, and some lines emitted benzenoids/phenylpropanoids or terpenoids without emitting fatty acids. We successfully selected elite breeding lines that predominantly emitted pleasant scents. Therefore, this study demonstrated the introduction of pleasant scent production capability into Asiatic hybrid lilies using interspecific hybridization with L. cernuum.

Introduction

Plants use floral scents to attract pollinators in addition to flower shape, color, and nectar. Different emission rates, chemical compounds, and diurnal rhythms of floral scents attract different pollinators (Amrad et al., 2016; Ando et al., 2001; Byers et al., 2014; Klahre et al., 2011). Furthermore, human beings find floral scents attractive; floral scents have been loved by humans since ancient times and are widely used in the perfumery, food flavoring, and cosmetics industries, etc. In addition, they are the basis for critical commercial traits of floricultural plants, along with flower color and shape (Muhlemann et al., 2014). However, many essential traits of floral scents are often overlooked in breeding programs, and many scent traits have been lost over time because of artificial selection. Extensive breeding programs have been undertaken to improve essential traits of floricultural plants, including flower shape, petal color, and flower longevity (vase life), which can be judged by appearance. Less attention has been devoted to enhancing or maintaining scent production, mainly because perception assessments are generally limited to sensory evaluations by the human nose, for which the odor threshold perception is much lower than that of insect antennae (Dudareva and Pichersky, 2008). As a result, many current cultivars release fewer floral scents than older cultivars (Bendahmane et al., 2013; Cherri-Martin et al., 2007; Fu et al., 2007).

Floral scents are complex mixtures of volatile compounds mainly divided into benzenoids/phenylpropanoids, terpenoids, and fatty acid derivatives (Muhlemann et al., 2014). Benzenoids/phenylpropanoids are further divided into three subclasses: benzenoid (C₆-C₁), phenylpropanoid-related (C₆-C₂), and phenylpropanoid (C₆-C₃) compounds. C₆-C₁, C₆-C₂, and C₆-C₃ compounds are biosynthesized from phenylalanine via different pathways (Fig. 1A; Mostafa et al., 2022). Terpenoid volatiles are classified as monoterpenes, sesquiterpenes, and ionones, which are metabolized by distinct pathways (Fig. 1B). Monoterpenes and sesquiterpenes are synthesized in the plastids and cytosol, respectively (Nagegowda, 2010; van Schie et al., 2006), and ionones are derived from the cleavage of carotenoids (Huang et al., 2009; Rubio et al., 2008). Fatty acids and their derivatives are classified according to their aliphatic tail lengths. Iso-valeric acid (3-methylbutanoic acid) is a short-chain fatty acid with five carbon atoms. Generally, benzenoids/phenylpropanoids and terpenoids are perceived as pleasant scents emitted by flowers, fruits, and herb leaves, whereas many fatty acids and their derivatives have unpleasant odors.

Fig. 1

Biosynthesis pathways of benzenoids/phenylpropanoids (A) and terpenoids (B) and their main products detected in this study. A: Benzenoid (C₆-C₁), phenylpropanoid-related (C₆-C₂), and phenylpropanoid (C₆-C₃) compounds are derived from phenylalanine, which is synthesized through the shikimate pathway. B: Monoterpenes and ionones are produced in plastids using isopentenyl pyrophosphate (IPP) and its isomer dimethylallyl pyrophosphate (DMAPP) as substrates, which are synthesized via the plastidial methylerythritol phosphate (MEP) pathway, while sesquiterpenes are generated in cytosol using the same substrates, which are produced through the cytosolic mevalonic acid (MVA) pathway.

The genus Lilium consists of more than 100 wild species classified into seven sections based on morphology and seed germination type: Martagon, Pseudolirium, Liriotypus, Archelirion, Sinomartagon, Leucolirion, and Daurolirion (Comber, 1949). Cultivated lily plants, one of the world’s best-known ornamental plants, have been developed by hybridization among wild species. As a result, six major hybrid groups, Asiatic (A), Oriental (O), Longiflorum (L), Trumpet (T), LA (Longiflorum × Asiatic), and OT (Oriental × Trumpet), have been established (Marasek-Ciolakowska et al., 2018; van Tuyl and Arens, 2011). Lilies have a wide variety of floral scents ranging from weak to strong. For example, Oriental hybrid lily cultivars and their wild parental species belonging to section Archelirion have a strong scent (Morinaga et al., 2009; Oyama-Okubo et al., 2011; Yokota and Yahara, 2012). Oriental hybrid lily flowers emit C₆-C₁ and C₆-C₃ benzenoid/phenylpropanoid and monoterpene compounds (Kong et al., 2012; Oyama-Okubo et al., 2011; Yoshida et al., 2018). In contrast, most Asiatic hybrid lily cultivars have a weak scent (Kong et al., 2012, 2017b), and only small amounts of benzenoids/phenylpropanoids and terpenoids have been detected in Asiatic hybrid lily flowers (Kong et al., 2012), although they exhibit valuable characteristics including a variety of tepal colors and color patterns (Yamagishi and Sakai, 2020; Yamagishi et al., 2014). Therefore, to compensate for this critical missing trait, one of the most important objectives in Asiatic hybrid lily breeding programs is to develop new cultivars with pleasing floral scents (Kong et al., 2012). Asiatic hybrids are derived from crosses between wild species in the sections Sinomartagon and Daurolirion (Marasek-Ciolakowska et al., 2018). Lilium cernuum, in the section Sinomartagon, is a wild species native to the Korean Peninsula, Primorsky Krai of Russia, and northeastern China and is one of the most important species used in the breeding of Asiatic hybrid lilies (McRae, 1998). For example, the trait of splatter-type tepal spots has been introduced from L. cernuum into Tango series cultivars (Yamagishi, 2020). While most wild parental species contributing to Asiatic hybrid lily development have no scent, L. cernuum has a characteristic scent that the human nose can perceive. Therefore, L. cernuum is a candidate for introducing scent traits into Asiatic hybrid lilies using interspecific hybridization. High amounts of a fatty acid and its derivative are present in the volatile chemicals of L. cernuum flowers, along with benzenoids/phenylpropanoids (Kong et al., 2017a). Since many fatty acids and their derivatives have unpleasant oily odors, their inclusion is unsuitable for pleasant floral scents. However, benzenoids/phenylpropanoids, terpenoids, and fatty acid derivatives are biosynthesized using different metabolic pathways. Thus, it is expected that plants releasing both or either of benzenoids/phenylpropanoids and terpenoids without producing fatty acid derivatives will be segregated in any progenies developed by interspecific hybridization.

In this study, we attempted to introduce the floral scent traits of L. cernuum into Asiatic hybrid lilies. We crossed the Asiatic hybrid lily cultivar ‘Mirella’ with L. cernuum to obtain F₁ individuals and then crossed the Asiatic hybrid cultivars ‘Mirella’ or ‘Connecticut King (CK)’ with the F₁ plants to develop second progeny lines. Volatiles were evaluated instrumentally: chemicals emitted from the flowers were collected using a static headspace sampling system or a dynamic headspace sampling system and analyzed using a gas chromatography-mass spectrometry (GC-MS) system equipped with thermal desorption. We selected new breeding lines that exhibited pleasant floral scents.

Materials and Methods

Plant materials

To introduce floral scent traits from L. cernuum to Asiatic hybrid lilies, 11 Asiatic hybrid lily cultivars were hybridized with L. cernuum (cut-style pollination and embryo rescue techniques were not used), and a single F₁ plant was obtained from the cross combination of the cultivar ‘Mirella’ × L. cernuum. Then, ‘Mirella’ and another Asiatic hybrid lily ‘CK’ were further crossed with the F₁ plant to produce second progeny lines (Fig. 2). From now on, the second progeny lines derived from ‘Mirella’ × (‘Mirella’ × L. cernuum) and ‘CK’ × (‘Mirella’ × L. cernuum) were called ‘Mirella’ progeny lines and ‘CK’ progeny lines, respectively.

Fig. 2

Plant materials. A: ‘Mirella’. B: L. cernuum. C: F₁ (‘Mirella’ × L. cernuum). D: ‘CK’. E: Six (i–vi) ‘Mirella’ progeny lines. F: four (i–iv) ‘CK’ progeny lines. Scale bar = 3 cm.

Lily plants were potted and grown in a greenhouse (with an unheated and natural photoperiod) at an experimental farm at Hokkaido University, Hokkaido, Japan. When the first flower opened, the whole plant was transferred to a growth chamber maintained at a constant temperature of 20°C with a 12-h light (6:00–18:00) and 12-h dark (18:00–6:00) photoperiod to estimate volatile compounds from flowers. The anthers were removed after flowering.

To compare the flower sizes of progeny lines with those of the parents, flower bud length was measured one day before anthesis. Statistical differences were analyzed using Tukey’s honestly significant difference (Tukey’s HSD) test using R v3.3.1 (https://www.R-project.org/) following a one-way analysis of variance.

Static headspace volatile collection

Volatiles released from flowers were collected two days after flowering using a static headspace sampling system (Oyama-Okubo and Tsuji, 2013). A stir bar sorptive extraction twister (Gerstel Inc., Linthicum, MD, USA) was clipped on a flower filament, and the whole flower was covered with a clear plastic wrapper, Saran Wrap (Asahi Kasei, Tokyo, Japan). The twister was allowed to absorb volatile compounds in the growth chamber for 2 h from 18:00 to 20:00. The twisters were stored in small sealed vials until analysis. Two flowers from each parental plant, one F₁ plant, six (i–vi) ‘Mirella’ progeny lines, and three (i–iii) ‘CK’ progeny lines were used.

Dynamic headspace volatiles collection

Volatiles released from flowers were collected using a dynamic headspace sampling system (Oyama-Okubo et al., 2005) two days after flowering. The entire flower was covered with a Tedlar Bag (500 mL; GL Science, Tokyo, Japan). Air filtered through the activated carbon constantly flowed into the bag at a flow rate of 500 mL·min⁻¹. Volatiles were captured on a Tenax column (180 mg; GL Science) in a glass tube for 24 h from 12:00 to 12:00. A single flower from ‘Mirella’ and two flowers from L. cernuum, ‘CK’, the F₁ plant, the four (i, iii, iv, and v) ‘Mirella’ progeny lines, and the four (i–iv) ‘CK’ progeny lines were used. ‘Mirella’ progeny line-ii and ‘Mirella’ progeny line-vi were not included in this experiment mainly because these lines did not set sufficient flowers, and the ‘CK’ progeny line-iv was additionally used instead.

GC-MS analysis

The volatiles captured by the twisters were analyzed using a GC-MS (Agilent 5975C; Agilent Technologies, Wilmington, DE, USA) coupled to a thermal desorption system 3 (TDS3; Gerstel Inc., Oyama-Okubo et al., 2018). The thermal desorption conditions were as follows: heating from 30°C to 220°C at 60°C·min⁻¹, holding at 220°C for 10 min, and cryofocusing at −50°C using a Cold Injection System 4 (CIS4; Gerstel Inc.). Following desorption from the twister, the CIS4 was heated to 300°C at a rate of 12°C·s⁻¹ in splitless mode to transfer the analytes to the GC equipped with a capillary DB-WAX column (30 m length, 0.25 mm i.d., 0.25 μm film thickness; Agilent Technologies). Helium was used as the carrier gas at a flow rate of 1.0 mL·min⁻¹. The oven temperature was set to 45°C for 2 min, then heated to 180°C at 3.5°C·min⁻¹, from 180°C to 250°C at 15°C·min⁻¹, and held at 250°C for 8 min. The ion source temperature was 250°C. Ionization was performed in electron impact mode at 70 eV, and a mass scan range of 30–350 m/z was monitored. Volatile compounds were identified using the Wiley 9^th/NIST 2011 library search system provided with GC-MS software (Agilent Technologies) and crosschecked by comparing the mass spectra and retention times with authentic samples analyzed under the same conditions. The amount of each compound was calculated based on the corresponding peak area in the total ion chromatogram of the authentic sample.

The volatiles captured on the Tenax column were analyzed in the same manner as described above, with some modifications. Thermal desorption conditions were as follows: heating from 30°C to 250°C at 60°C·min⁻¹, holding at 250°C for 10 min, and then freezing at −50°C using CIS4. The oven temperature program of the GC was as follows: 45°C for 2 min, followed by an increase to 220°C at 3°C·min⁻¹, and held at 220°C for 10 min.

Results

Appearances in lily lines developed in this study

Lilium cernuum was used as a hybridization partner to introduce floral scent traits into Asiatic hybrid lilies. As L. cernuum had some unfavorable characteristics compared to modern cultivars, agronomic traits were evaluated in the progeny lines. Upward-facing flowers are typical features of Asiatic hybrid lilies. The Asiatic hybrid lily cultivars ‘Mirella’ and ‘CK’ had upward-facing flowers, whereas L. cernuum flowers were downward-facing (Fig. 2A, B, D). The F₁ plants bloomed horizontally, intermediate between the flower facing of the parents (Fig. 2C). Among the ‘Mirella’ and ‘CK’ progeny lines, one plant, ‘Mirella’ progeny line-ii, bloomed sideways, while the flowers of other lines were upward-facing (Fig. 2E, F).

Lily cultivars with large flowers are desirable. The flower bud length in L. cernuum was shorter than in ‘Mirella’ and ‘CK’ (Fig. 3). The flower bud length of F₁ was intermediate between that of L. cernuum and ‘Mirella’. Flower bud length of the ‘Mirella’ and ‘CK’ progeny lines was intermediate between those in the F₁ and the Asiatic parental cultivars, and the ‘Mirella’ progeny line-i and ‘CK’ progeny line-ii showed no significant differences in the flower bud length of the two Asiatic hybrid lily cultivars.

Fig. 3

Flower bud length in L. cernuum, ‘Mirella’, ‘CK’, F₁ (‘Mirella’ × L. cernuum), six (i–vi) ‘Mirella’ progeny lines, and three (i, ii, and iii) ‘CK’ progeny lines. Vertical bars show the standard error of the means of three flower buds. Columns bearing the same letters indicate no significant difference (P < 0.05; Tukey’s HSD test).

Most Asiatic hybrid lily cultivars exhibit early flowering. The flowering date was similar between L. cernuum, ‘Mirella’, and ‘CK’, and no significant difference was found in the flowering date among the F₁ and the ‘Mirella’ and ‘CK’ progeny lines (data not shown).

Scent traits estimated using the static headspace method

The scent compounds in two flowers from each plant were estimated using the static headspace volatile collection method (Table S1), and the intermediary values of the two flowers are shown in Figure 4. Lilium cernuum flowers emitted benzenoids/phenylpropanoids, whereas no quantifiable components were detected from ‘Mirella’ and ‘CK’ flowers. Benzenoids/phenylpropanoids include C₆-C₁, C₆-C₂, and C₆-C₃ compounds (Fig. 1). C₆-C₁ (benzaldehyde and benzyl alcohol), C₆-C₂ (2-phenylethanol and phenylacetaldehyde), and C₆-C₃ (eugenol and 3-hydroxy-4-phenyl-2-butanone) were emitted from L. cernuum flowers. In flowers of the F₁ and the ‘Mirella’ and ‘CK’ progeny lines, C₆-C₁ compounds (benzaldehyde and benzyl benzoate) were mainly detected, followed by C₆-C₂ compounds (2-phenylethanol). Eugenol (C₆-C₃ compound) was among the main scent components of L. cernuum, but was only faintly detected in the flowers of the progeny generations. Relatively high amounts of benzenoids/phenylpropanoids were emitted by ‘Mirella’ progeny line-iii, ‘Mirella’ progeny line-iv, and ‘Mirella’ progeny line-v among the second progeny lines.

Fig. 4

Amounts of scent compounds emitted from flowers of L. cernuum, ‘Mirella’, ‘CK’, F₁ (‘Mirella’ × L. cernuum), six (i–vi) ‘Mirella’ progeny lines, and three (i, ii, and iii) ‘CK’ progeny lines two days after flowering. A: Benzenoids/phenylpropanoids. C₆-C₁ compound: blue-toned; C₆-C₂ compound: pink-toned; and C₆-C₃ compound: orange-toned. B: Terpenoids. Monoterpenes: green-toned; sesquiterpenes: brown-toned; and ionones: gray-toned. The intermediary values of the two flowers are shown.

Terpenoid volatiles are classified as monoterpenes, sesquiterpenes, and ionones (Fig. 1). Monoterpenes [linalool, linalool oxide (pyranoid), and linalool oxide (furanoid)] were major terpenoids emitted from L. cernuum flowers. Relatively small amounts of β-ionone were also included in the volatiles. ‘Mirella’ and ‘CK’ flowers emitted little amounts of monoterpenes and ionones, but ‘Mirella’ flowers released small amounts of sesquiterpenes (α-farnesene and caryophyllene). ‘Mirella’ possessed the ability to release sesquiterpenes, although only in small amounts. Among the ‘Mirella’ and ‘CK’ progeny lines, relatively higher amounts of monoterpenes were emitted from flowers in ‘Mirella’ progeny line-i, ‘Mirella’ progeny line-iv, ‘Mirella’ progeny line-vi, and ‘CK’ progeny line-iii, but no or only faint amounts were detected in other lines.

Although a fatty acid and its derivative were detected in L. cernuum flowers in a previous study (Kong et al., 2017a), none were detected by the static headspace method using twisters.

Scent traits estimated using the dynamic headspace method

The dynamic headspace method was performed on one flower from ‘Mirella’ and two flowers from the other plants/lines to evaluate scent compounds (Table S2), focusing on fatty acids and their derivatives (Fig. 5). A single fatty acid, iso-valeric acid, was detected in L. cernuum, but not in ‘Mirella’ or ‘CK’. The F₁ and one of the eight second-progeny lines, ‘Mirella’ progeny line-iv, emitted iso-valeric acid. The emissions of benzenoids/phenylpropanoids and terpenoids estimated using the dynamic headspace method were similar to those detected using the static headspace method: for example, linalool and linalool oxides were the main terpenoids emitted from these flowers, and lines that emitted relatively high amounts of these were included in the ‘Mirella’ and ‘CK’ progeny lines (Table S2).

Fig. 5

Amounts of fatty acid emitted from flowers of L. cernuum, ‘Mirella’, ‘CK’, F₁ (‘Mirella’ × L. cernuum), four (i, iii, iv, and v) ‘Mirella’ progeny lines, and four (i–iv) ‘CK’ progeny lines two days after flowering. Only iso-valeric acid was detected among fatty acids and their derivatives. The intermediary values of the two flowers are shown except for ‘Mirella’, of which one flower was evaluated.

Discussion

Wild species that can hybridize with cultivated species are essential genetic resources for improving cultivated species. Many traits, such as disease resistance and fruit quality, have been introduced from wild species to modern cultivars (Cao et al., 2017; Labate and Robertson, 2012; Ronald et al., 1992). Novel scent traits of wild species have also been introduced into cultivated species. Interspecific hybridization has been performed several times to produce novel scent traits during the development of modern roses (Rosa × hybrida, Bendahmane et al., 2013). The introgression of scent traits from related wild species has recently been examined in Alstroemeria, Cyclamen, and carnations (Aros et al., 2019; Ishizaka et al., 2002; Kishimoto et al., 2013). These studies clarified the transmission of scent traits from parental species to their interspecific hybrids, but the expression of scent traits in further generations, such as F₂ or BC₁, has not been evaluated. In this study, floral scents in the first (F₁) and second progeny plants were evaluated to confirm the introduction of scent traits of L. cernuum into Asiatic hybrid lilies.

Wild species are significant genetic resources for plant breeding. However, wild species often have unfavorable traits such as low yield, poor quality, or difficulty in terms of cultivation; therefore, subsequent backcrossing with cultivated species is usually necessary to remove these unfavorable traits. Flowers of L. cernuum are small compared to Asiatic hybrid lily cultivars and downward-facing (upward-facing is preferred for the bouquet arrangements). This study showed that the size (bud length) and angle of flowers in the F₁ plants were intermediate, and those in many second-progeny lines were similar to those in the Asiatic hybrid lily cultivars. Therefore, we believe unfavorable traits in L. cernuum can be eliminated during subsequent crossings.

Lilium cernuum was used as the donor for floral scent traits. As for benzenoids/phenylpropanoids, L. cernuum produced C₆-C₁, C₆-C₂, and C₆-C₃ compounds, whereas C₆-C₁ and C₆-C₃ compounds are the major compounds, and C₆-C₂ compounds are the minor ones in Oriental, Trumpet, and OT hybrids (Johnson et al., 2016; Kong et al., 2012, 2017b; Oyama-Okubo et al., 2011; Yoshida et al., 2018). As for terpenoids, linalool and linalool oxide were mainly emitted from L. cernuum flowers, whereas linalool and ocimene are the primary compounds in Oriental hybrid lilies, and linalool, ocimene, and eucalyptol (cineole) are the main products in Trumpet and OT hybrid lilies (Johnson et al., 2016; Kong et al., 2012, 2017b; Oyama-Okubo et al., 2011; Yoshida et al., 2018). In addition, L. cernuum emitted high amounts of fatty acids, which are barely detected in other hybrid lilies. Consequently, L. cernuum appears to be unique in its volatile composition among Lilium species.

C₆-C₂ compounds (phenylacetaldehyde and 2-phenylethanol) accounted for approximately half of the benzenoid/phenylpropanoid components in L. cernuum. However, in the F₁ and the ‘Mirella’ and ‘CK’ progeny lines, C₆-C₁ compounds (benzaldehyde and benzyl benzoate) were mainly detected, followed by C₆-C₂ compounds (2-phenylethanol). The total amounts of benzenoid/phenylpropanoid compounds varied among the ‘Mirella’ and ‘CK’ progeny lines, and the ‘Mirella’ progeny line-iii, ‘Mirella’ progeny line-iv, and ‘Mirella’ progeny line-v released relatively large amounts of these compounds. These results indicate that we can select elite lines that emit sufficient amounts of C₆-C₁ and C₆-C₂ compounds in the progenies. Monoterpenes were detected in L. cernuum and F₁ flowers. Four of the nine second-progeny lines released relatively high amounts of monoterpenes, indicating that we can select lines producing desirable monoterpene scents in the progenies.

The aromatic description of iso-valeric acid is [sweat, acid, rancid], according to Flavornet (www.flavornet.com). Iso-valeric acid was detected in L. cernuum, but not in ‘Mirella’ or ‘CK’. It was also detected in F₁ and one of the eight second-progeny lines, the ‘Mirella’ progeny line-iv. These results indicate that plants that release iso-valeric acid and plants that do not release iso-valeric acid are segregated in the progeny lines. While the ‘Mirella’ progeny line-iv emitted iso-valeric acid, benzenoids/phenylpropanoids, and terpenoids, some second-progeny lines released high amounts of benzenoids/phenylpropanoids or terpenoids without emitting iso-valeric acid, indicating that the traits that release compounds of either benzenoids/phenylpropanoids, monoterpenes, or fatty acids are independently transmitted from L. cernuum to Asiatic hybrid lilies. Independent transmission is beneficial for developing ideal lily lines that predominantly emit favorable scents.

Oriental hybrid lilies often exhibit strong scents, which are sometimes overpowering for people, resulting in restricted usage of their cut flowers to confined spaces and restaurants (Oyama-Okubo et al., 2011). Compared to Oriental hybrid lilies, L. cernuum flowers have a weak, but unpleasant scent. The unpleasant odors were mainly due to the emission of iso-valeric acid. In contrast, weak, but pleasant scents were perceived by the flowers from some second-progeny lines. This could be due to the emission of benzaldehyde, linalool, and/or linalool oxide, for which the aromatic descriptions are [almond, burnt sugar], [flower, lavender], and [flower], respectively (Flavornet, www.flavornet.com).

In conclusion, we successfully isolated progeny lines, for which the main scent components were benzenoids/phenylpropanoids or terpenoids, that did not emit fatty acids. Lilium cernuum has some unfavorable morphological flower characteristics that can be eliminated during subsequent crossing. Therefore, this study demonstrated the introduction of scent traits into Asiatic hybrid lilies using L. cernuum and showed that L. cernuum is a promising hybridization partner for improving scent traits.

Acknowledgements

Lilium cernuum was obtained from collections at the Yurigahara (lily garden) Park, Sapporo, Japan.

Literature Cited

•  Amrad,  A.,  M.  Moser,  T.  Mandel,  M.  De Vries,  R. C.  Schuurink,  L.  Freitas and  C.  Kuhlemeier. 2016. Gain and loss of floral scent production through changes in structural genes during pollinator-mediated speciation. Curr. Biol. 26: 3303–3312.
•  Ando,  T.,  M.  Nomura,  J.  Tsukahara,  H.  Watanabe,  H.  Kokubun,  T.  Tsukamoto,  G.  Hashimoto,  E.  Marchesi and  I. J.  Kitching. 2001. Reproductive isolation in a native population of Petunia sensu Jussieu (Solanaceae). Ann. Bot. 88: 403–413.
•  Aros,  D.,  M.  Suazo,  C.  Rivas,  P.  Zapata,  C.  Úbeda and  M.  Bridgen. 2019. Molecular and morphological characterization of new interspecific hybrids of alstroemeria originated from A. caryophylleae scented lines. Euphytica 215: 93.
•  Bendahmane,  M.,  A.  Dubois,  O.  Raymond and  M.  Le Bris. 2013. Genetics and genomics of flower initiation and development in roses. J. Exp. Bot. 64: 847–857.
•  Byers,  K. J. R. P.,  H. D.  Bradshaw and  J. A.  Riffell. 2014. Three floral volatiles contribute to differential pollinator attraction in monkeyflowers (Mimulus). J. Exp. Biol. 217: 614–623.
•  Cao,  X.,  Z.  Qiu,  X.  Wang,  T.  Van Giang,  X.  Liu,  J.  Wang,  X.  Wang,  J.  Gao,  Y.  Guo,  Y.  Du,  G.  Wang and  Z.  Huang. 2017. A putative R3 MYB repressor is the candidate gene underlying atroviolacium, a locus for anthocyanin pigmentation in tomato fruit. J. Exp. Bot. 68: 5745–5758.
•  Cherri-Martin,  M.,  F.  Jullien,  F.  Heizmann and  S.  Baudino. 2007. Fragrance heritability in hybrid tea roses. Sci. Hortic. 113: 177–181.
• Comber, H. F. 1949. A new classification of the genus Lilium. p. 86–105. In: Lily yearbook, vol. 13. Royal Horticultural Society, London.
•  Dudareva,  N. and  E.  Pichersky. 2008. Metabolic engineering of plant volatiles. Curr. Opin. Biotechnol. 19: 181–189.
•  Fu,  Y.,  X.  Gao,  Y.  Xue,  Y.  Hui,  F.  Chen,  Q.  Su and  L.  Wang. 2007. Volatile compounds in the flowers of freesia parental species and hybrids. J. Integrative Plant Biol. 49: 1714–1718.
•  Huang,  F. C.,  G.  Horvath,  P.  Molnar,  E.  Turcsi,  J.  Deli,  J.  Schrader,  G.  Sandmann,  H.  Schmidt and  W.  Schwab. 2009. Substrate promiscuity of RdCCD1, a carotenoid cleavage oxygenase from Rosa damascena. Phytochemistry 70: 457–464.
•  Ishizaka,  H.,  H.  Yamada and  K.  Sasaki. 2002. Volatile compounds in the flowers of Cyclamen persicum, C. purpurascens and their hybrids. Sci. Hortic. 94: 125–135.
•  Johnson,  T. S.,  M. L.  Schwieterman,  J.  Young,  K. H.  Cho,  D. G.  Clark and  T. A.  Colquhoun. 2016. Lilium floral fragrance: a biochemical and genetic resource for aroma and flavor. Phytochemistry 122: 103–112.
•  Kishimoto,  K.,  M.  Yagi,  T.  Onozaki,  H.  Yamaguchi,  M.  Nakayama and  N.  Oyama-Okubo. 2013. Analysis of scents emitted from flowers of interspecific hybrids between carnation and fragrant wild Dianthus species. J. Japan. Soc. Hort. Sci. 82: 145–153.
•  Klahre,  U.,  A.  Gurba,  K.  Hermann,  M.  Saxenhofer,  E.  Bossolini,  P. M.  Guerin and  C.  Kuhlemeier. 2011. Pollinator choice in Petunia depends on two major genetic loci for floral scent production. Curr. Biol. 21: 730–739.
•  Kong,  Y.,  J.  Bai,  L.  Lang,  F.  Bao,  X.  Dou,  H.  Wang and  H.  Shang. 2017a. Floral scents produced by Lilium and Cardiocrinum species native to China. Biochem. Syst. Ecol. 70: 222–229.
•  Kong,  Y.,  J.  Bai,  L.  Lang,  F.  Bao,  X.  Dou,  H.  Wang and  H.  Shang. 2017b. Variation in floral scent compositions of different lily hybrid groups. J. Amer. Soc. Hort. Sci. 142: 175–183.
•  Kong,  Y.,  M.  Sun,  H.  Pan and  Q.  Zhang. 2012. Composition and emission rhythm of floral scent volatiles from eight lily cut flowers. J. Amer. Soc. Hort. Sci. 137: 376–382.
•  Labate,  J. A. and  L. D.  Robertson. 2012. Evidence of cryptic introgression in tomato (Solanum lycopersicum L.) based on wild tomato species alleles. BMC Plant Biol. 12: 133. DOI: 10.1186/1471-2229-12-133.
•  Marasek-Ciolakowska,  A.,  T.  Nishikawa,  D. J.  Shea and  K.  Okazaki. 2018. Breeding of lilies and tulips—interspecific hybridization and genetic background. Breed. Sci. 68: 35–52.
• McRae, E. A. 1998. Lilies: A guide for growers and collectors. Timber Press Inc., Portland, Oregon.
•  Morinaga,  S.,  Y.  Kumano,  A.  Ota,  R.  Yamaoka and  S.  Sakai. 2009. Day–night fluctuations in floral scent and their effects on reproductive success in Lilium auratum. Popul. Ecol. 51: 187–195.
•  Mostafa,  S.,  Y.  Wang,  W.  Zeng and  B.  Jin. 2022. Floral scents and fruit aromas: Functions, compositions, biosynthesis, and regulation. Front. Plant Sci. 13: 860157. DOI: 10.3389/fpls.2022.860157.
•  Muhlemann,  J. K.,  A.  Klempien and  N.  Dudareva. 2014. Floral volatiles: from biosynthesis to function. Plant Cell Environ. 37: 1936–1949.
•  Nagegowda,  D. A. 2010. Plant volatile terpenoid metabolism: Biosynthetic genes, transcriptional regulation and subcellular compartmentation. FEBS Lett. 584: 2965–2973.
•  Oyama-Okubo,  N. and  T.  Tsuji. 2013. Analysis of floral scent compounds and classification by scent quality in tulip cultivars. J. Japan. Soc. Hort. Sci. 82: 344–353.
•  Oyama-Okubo,  N.,  T.  Ando,  N.  Watanabe,  E.  Marchesi,  K.  Uchida and  M.  Nakayama. 2005. Emission mechanism of floral scent in Petunia axillaris. Biosci. Biotechnol. Biochem. 69: 773–777.
•  Oyama-Okubo,  N.,  T.  Haketa,  H.  Furuichi and  S.  Iioka. 2018. Characteristics of floral scent compounds in a new fragrant petunia cultivar TX-794 ‘Evening Scentsation’. Hort. J. 87: 258–263.
•  Oyama-Okubo,  N.,  M.  Nakayama and  K.  Ichimura. 2011. Control of floral scent emission by inhibitors of phenylalanine ammonia-lyase in cut flower of Lilium cv. ‘Casa Blanca’. J. Japan. Soc. Hort. Sci. 80: 190–199.
•  Ronald,  P. C.,  B.  Albano,  R.  Tabien,  L.  Abenes,  K.  Wu,  S.  McCouch and  S. D.  Tanksley. 1992. Genetic and physical analysis of the rice bacterial blight disease resistance locus, Xa21. Mol. Gen. Genet. 236: 113–120.
•  Rubio,  A.,  J. L.  Rambla,  M.  Santaella,  M. D.  Gómez,  D.  Orzaez,  A.  Granell and  L.  Gómez-Gómez. 2008. Cytosolic and plastoglobule-targeted carotenoid dioxygenases from Crocus sativus are both involved in β-ionone-release. J. Biol. Chem. 283: 24816–24825.
•  van Schie,  C. C. N.,  M. A.  Haring and  R. C.  Schuurink. 2006. Regulation of terpenoid and benzenoid production in flowers. Curr. Opin. Plant Biol. 9: 203–208.
•  van Tuyl,  J. M. and  P.  Arens. 2011. Lilium: breeding history of the modern cultivar assortment. Acta Hortic. 900: 223–230.
•  Yamagishi,  M. 2020. White with partially pink flower color in Lilium cernuum var. album is caused by transcriptional regulation of anthocyanin biosynthesis genes. Sci. Hortic. 260: 108880. DOI: 10.1016/j.scienta.2019.108880.
•  Yamagishi,  M. and  M.  Sakai. 2020. The microRNA828/MYB12 module mediates bicolor pattern development in Asiatic hybrid lily (Lilium spp.) flowers. Front. Plant Sci. 11: 590791. DOI: 10.3389/fpls.2020.590791.
•  Yamagishi,  M.,  S.  Toda and  K.  Tasaki. 2014. The novel allele of the LhMYB12 gene is involved in splatter-type spot formation on the flower tepals of Asiatic hybrid lilies (Lilium spp.). New Phytol. 201: 1009–1020.
•  Yokota,  S. and  T.  Yahara. 2012. Pollination biology of Lilium japonicum var. abeanum and var. japonicum: evidence of adaptation to the different availability of diurnal and nocturnal pollinators. Plant Species Biol. 27: 96–105.
•  Yoshida,  K.,  N.  Oyama-Okubo and  M.  Yamagishi. 2018. An R2R3-MYB transcription factor ODORANT1 regulates fragrance biosynthesis in lilies (Lilium spp.). Mol. Breed. 38: 144.