Inventory of leaf and flower odorants in plants associated with the life cycle of Japanese Papilio butterflies

based on the odorant volatiles, which did not correspond to the current phylogenetic classification. Similarly, floral odorant analysis of the six plant species ( Clerodendrum trichotomum , Cayratia japonica , Robinia pseudoacacia , Lonicera japonica , C. deliciosa and Z. ailanthoides ) visited by Papilio butterflies for nectaring revealed the presence of linalool in all flowers. Floral volatiles in C. deliciosa and Z. ailanthoides exhibited a moderate resemblance to their respective leaf volatiles. Interestingly, the results obtained for C. trichotomum were not in complete agreement with those of previous reports, emphasizing the need for newer methods of extraction and analysis.

Although drumming behavior is commonly observed during oviposition, we often observe that Papilio butterflies oviposit without contacting the plant leaves with their forelegs; furthermore, butterflies sometimes oviposit their eggs on non-host plants or objects adjacent to host plants [20]. These behaviors could be attributed to plant volatile substances, which may excite the sensitive olfactory receptors of the butterflies. In fact, butterflies are known to locate the host plant habitats by following olfactory cues and consecutively identifying the host plant by taste cues. This has been demonstrated in P. demoleus [21], P. polyxenes [22][23][24], P. machaon in Japan [23], P. troilus [23], P. cresphontes [25], P. glaucus, and P. canadensis [26]. A recent study analyzed the volatile components released by host plants of Papilio indra occurring in the Rocky Mountains [27,28]. In addition, plant volatiles of Australasian Rutaceae plants have been analyzed in previous studies [29,30], and the relationship between plant volatiles and the oviposition behavior of two Japanese Leptocircini species has been previously examined [31,32]. In addition, several other studies have been conducted from a pharmacological, fungal defense, or pollination perspective [33][34][35][36][37][38][39][40]. However, a comprehensive study on Papilio butterflies and the odorants of their egg-laying plants in Japan has not been previously undertaken.
For the identification of volatile substances in the leaves and flowers, all plant specimens were collected around noon, when the production of plant volatiles is expected to be maximal during the day. Samples were transferred to the Showa Denko Materials Techno Service laboratory and stored at -40°C until further analysis.

Analysis of plant volatiles (Method-1)
The leaf volatiles were subjected to Gas Chromatography-Mass Spectrometry (GC-MS) on an Agilent system consisting of a model 7890 gas chromatograph, a model 5977 mass-selective detector (EIMS, electron energy of 70 eV), and an Agilent ChemStation data system (Santa Clara, CA, USA). Volatiles in the leaf were trapped in an odorant-collecting cartridge (Tenax TA, GL Sciences, Tokyo, Japan) and subjected to GC using a newly developed nonsolvent method known as dynamic headspace and thermal desorption system (Gerstel, Überhausen, Germany) with a COMPS2XLxt multipurpose sampler. The GC column was a DB-VRX column (Agilent, USA) with a film thickness of 1.44 μm, length of 60 m, and internal diameter of 0.25 mm. The carrier gas was helium, with a flow rate of 2.1 mL/min. The GC oven temperature was regulated as follows: 40°C initial temperature held for 3 min; increased at 5°C/min to 260°C and held for 8 min. Leaf samples (0.2 g) were placed in 20 mL vials and measured using the dynamic headspace technique. The mass-selective detector was set at 230°C. The volatiles were identified by comparing their MS fragmentation patterns to those in the MS library (NIST14 database).

Analysis of volatiles from A. keiskei (Method-2)
Angelica keiskei leaves were dried at 40°C, crushed to a powder, and the volatiles from 0.2 g of this powder were trapped at room temperature (22°C) for 24 h using an odorant-collecting cartridge (RCC18; GL Science). RCC18 was dipped in acetone (Wako, Fujifilm, Ôsaka, Japan) to extract volatiles that were then analyzed on a GC-MS QP2010 system (Shimadzu, Japan). The GC column was a DB-5MS column (Agilent, USA) with a film thickness of 0.25 μm, length of 30 m, and internal diameter of 0.25 mm. The carrier gas was helium, at a flow rate of 2.1 mL/min. The GC oven temperature was regulated as follows: 40°C initial temperature held for 2 min; increased at 5°C/min to 240°C and held for 5 min. The mass-selective detector temperature was set at 250°C. The volatiles were identified by comparing their MS fragmentation patterns to those in the MS library (NIST14 database). This analysis was performed as a part of Otani's bachelor's degree thesis.

Volatiles of the host plant leaves
The analysis of leaf volatiles of Papilio butterfly host plants using Method-1 is shown in Table 1. We detected a total of 62 volatiles in the 14 plant species studied. Based on the number and type of volatiles detected, these host plants were roughly divided into six groups. Group A consisted of C. deliciosa, F. vulgare and A. keiskei. The characteristic of this group was that the number of detected compounds was 11-18, and included mostly terpenes; alcohols, ketones, and aldehydes were not detected. Group B consisted of B. albiflora, O. japonica, P. amurense and Z. ailanthoides. The number of compounds was 6-11 and included β-caryophyllene among sesquiterpenes (except for Z. ailanthoides), along with other monoterpenes. Group C consisted of Z. schinifolium, Z. piperitum and S. japonica. The number of compounds was 10-14 and included predominantly monoterpenes containing toluene (except β-caryophyllene in Z. schinifolium and the sesquiterpene cubebene in S. japonica). Group D consisted of only R. graveolens. This species contained 18 different types of compounds containing furan. Group E consisted of H. lanatum and A. sylvestris. This group contained 10-12 compounds, and included predominantly monoterpenes, along with alcohols, ketones, and sesquiterpenes. Group F consisted of E. meliifolia that contained alcohols, ketones, aldehydes, and only β-caryophyllene among terpenes (Table 1).
A comparison of the volatiles of A. keiskei analyzed using Methods-1 and 2 are shown in Table 2. Although low-boiling point compounds were generally detected in both methods, Method-2 failed to detect high-boiling point compounds, such as isocaryophyllene, elixene, germacrene D or germacrene B ( Table 2).

Volatiles of the flowers
The analysis of volatile substances in the flowers of six species often visited by Papilio butterflies is shown in Table 3. We detected a total of 45 volatiles. These plants were divided into two groups, G and H. The former consisted of C. trichotomum, C. japonica, R. pseudoacacia and L. japonica, whereas the latter consisted of C. deliciosa and Z. ailanthoides. Group G species contained mainly aldehydes, ketones, and alcohols, whereas those of group H contained mainly monoterpenes, with the floral volatiles partially resembling the leaf volatiles. Although the flowers of C. japonica and Z. ailanthoides appear physically similar, the volatiles did not correspond between the species. Linalool was detected in all plants. Unlike the leaves, no sesquiterpenes were detected in the flowers ( Table 3).
The use of plant parts in the present study complies with international, national and/or institutional guidelines.

DISCUSSION
The analysis of plant volatiles from the leaves revealed unexpected results, mainly because their detection did not correlate with either the phylogenetic classification or the host plant preference of Papilio species. Interestingly, β-myrcene was detected in 12 species (except R. graveolens and E. meliifolia), d-limonene in ten species, and β-caryophyllene in nine species. These three substances were detected in at least one of the 14 plants examined in this study, except R. graveolens; thus, these substances might be associated with the oviposition behavior of butterflies of the Papilio species. As R. graveolens is not indigenous to Japan, it may be necessary to consider the relationship between this species and Japanese Papilio species from a different perspective.
According to our observation in Tsukuba city, the grouping of F. vulgare with C. deliciosa, O. japonica and S. japonica was supported by the observation that female P. machaon sometimes oviposit on these plants (not published).
The criteria for host plant selection in female P. xuthus and P. helenus, both of which lay their eggs on C. deliciosa, Z. ailanthoides and E. meliifolia, could not be explained by the composition of plant volatiles; namely, E. meliifolia shared no common volatiles with C. deliciosa and Z. ailanthoides, except for β-caryophyllene that was not detected in Z. ailanthoides. The odor of E. meliifolia leaves is similar to that of apple fruits; this is completely different from that of other Rutaceae plants in Japan. In fact, hexanal and 2-hexenal, detected in E. meliifolia but not in the other 12 species (only detected from R. graveolens other than E. meliifolia), have been detected in apples [43]. Thus, these substances might be also associated with the oviposition behavior of the Papilio species. Similarly, cubebene, detected in Z. piperitum and S. japonica, is found in Piper cubeba (Piperaceae) [44], but not in other plants examined in this study. In fact, in our observation, seeds and leaves of Z. piperitum are used as "Japanese pepper" and S. japonica has an odor similar to pepper (not published). These results suggest the existence of other key plant volatiles responsible for the oviposition behavior in Papilio species; however, these were not detected in this study. Mozuraitis et al. [45] also suggested that volatiles released from the foliar extract of host plants enhance the landing rates of gravid Polygonia c-album (Nymphalidae) females but do not stimulate oviposition. The host plant selection system of Papilio and other butterflies based on plant volatiles is more complicated than previously expected, and cannot be elucidated exclusively on the basis of scent components in plants. Further analysis of the components in leaves is warranted to address this.
Although the six nectaring plants were divided into two groups on the basis of their floral volatiles, linalool was detected in all species; thus, linalool may be a key substance inducing the nectaring behavior of Papilio butterflies. The result obtained for C. trichotomum have been previously reported [46] and the results from a previous study are summarized in Table 4 along with the results obtained in this study. This study revealed an important issue; in Miyake et al.'s [46] study, almost all volatiles with a comparatively low-boiling point were not detected. The authors, in their study methods, describe the following: "The compounds trapped by the absorption tube were eluted with 2 ml of diethyl ether for pesticide residue analysis. The eluate was their study, low-boiling point volatiles were lost during concentration. In contrast, this study was unable to detect the four esters observed in their study. Therefore, those esters might have appeared during the concentration process used in the other study. Thus, this study emphasizes the need to reanalyze plant volatiles using the latest methods to identify novel candidates (Table 4). concentrated to the limit by the passing of N 2 across its surface." Thus, in    Papilio butterflies to these volatiles. These results will be shared in a separate article.

AUTHOR CONTRIBUTIONS
H.O. conceived the experiments, T.A.I and K.N conducted the experiments, and T.F analyzed the results. All authors reviewed and approved the manuscript.

CONCLUSION
Thus, Nymphalinae and Papilioninae butterflies may respond to the same substances differently. Interestingly, we noted differences in volatiles determined using Methods 1 and 2 in the same plant (A. keiskei). Hence, our study highlights that different methods of extraction and analysis may be necessary to obtain a comprehensive profile of the plant volatiles involved in the oviposition behavior of butterflies. We have already obtained the results of the relationship between some of these plant volatiles and the response of