Yeast associated with flower longicorn beetle Leptura ochraceofasciata (Cerambycidae: Lepturinae), with implication for its function in symbiosis

Wood is difficult for most animals to digest due to large amounts of indigestible polymers, but some wood-feeding insects are considered to be able to utilize it as food with the aid of microbial symbionts. Most members of flower longicorn beetles (Coleoptera: Cerambycidae: Lepturinae) feed on nectar and pollen of flowers as adults and wood as larvae. In some lepturines, associations with yeasts are known: female adults possess fungus-storing organs (termed mycetangia) at ovipositors, and larvae also possess such organs (termed mycetomes) in their midguts to carry the associated yeasts. Despite the high diversity of Lepturinae in the world, lepturine-yeast associations, such as the consistency of associated yeasts among the beetle’s developmental stages and ecological function of yeast symbionts, have been poorly documented. Here, we investigated the yeast symbiont of the Japanese common lepturine Leptura ochraceofasciata. X-ray computed microtomography revealed that a pair of tube-like, S-shaped mycetangia was located at the basal part of the ovipositor and that a muscle bundle joined the apex of the mycetangium to spiculum ventrale of sternum VIII. All female adults harbored only one yeast species, Scheffersomyces insectosa, in the mycetangia. All larvae harbored S. insectosa exclusively in the mycetomes. Scheffersomyces insectosa was also recovered from surfaces of eggs. Scheffersomyces insectosa assimilated wood-associated sugars including xylose, cellobiose, and xylan in culture. These results suggest the intimate association between L. ochraceofasciata and S. insectosa: S. insectosa is transmitted from the mother to offspring during oviposition and may be related to larval growth in wood.


Introduction
Wood is composed of indigestible polymers, such as cellulose, hemicelluloses, and lignin, and thus wood is unavailable for most animals as food [1]. In nature, however, diverse insects

Insects
For yeast isolation, 11 female adults of L. o. ochraceofasciata were collected in central Honshu, Japan in July and August 2018, July 2020, and June 2021 (Table 1). Of those, eight were collected when visiting flowers of Angelica pubescens (Apiaceae), Cynanchum caudatum (Apocynaceae), and Hydrangea paniculata (Hydrangeaceae). One (individual ID: Fi10) was captured when she just emerged from decayed wood of Chamaecyparis obtusa (Cupressaceae). Two (Fs1 and Fs2) originated from a piece of decayed wood of Abies sp. (Pinaceae), which was identified by microscopic observation following the procedure of Iimura et al. (2021) [24]. The sampled wood was placed in the laboratory at room temperature. One female (Fs1) emerged from the wood and the other (Fs2) was in the pupal chamber when the wood was split using a wood-cutting knife. In addition, a female adult (Fy1) of L. o. ochrotela Bates was collected on a fallen dead tree (unidentified conifer) in Kyushu, Japan on 22 July 2020 (Table 1). Water or honey solution was added to keep these adult samples alive until use.
Samples were weighed using a digital scale when dissected. Body and elytral lengths of adults were also determined using digital calipers.
To obtain eggs, ten adult females of L. o. ochraceofasciata (elytral length: mean ± SD = 11.63 ± 0.85 mm, n = 10) were captured in Inabu on 30 July, 2020. As an oviposition substrate, a rolled corkboard (4-cm diameter × 6 cm) with a piece of decayed wood of Ch. obtusa where larvae of L. o. ochraceofasciata were found in Inabu placed at the core was put in a plastic container (10 × 10 × 10 cm). Then, adult females were placed in the container and allowed to lay eggs at room temperature (ca. 25˚C) under florescent light. Twenty-six arbitrarily selected eggs were individually placed onto a Petri dish (3-cm diameter) with a piece of moistened paper and incubated at 25˚C in the dark. For yeast isolation, three 14-day-reared eggs were selected arbitrarily. The other eggs were incubated until hatching to observe behaviors of hatched larvae. To identify larval samples, we applied a molecular approach. Adults of 13 species from 9 genera in the tribe Lepturini and 2 species from 2 genera in the tribe Rhagiini used as an outgroup were collected in Japan ( Table 2). Note that all species of the genus Leptura recorded in Aichi Prefecture were used for the analysis except for an uncommon species L. kusamai Ohbayashi et Nakane [25]. The captured samples were preserved in absolute ethanol.
No specific permits were required for the described field studies. The locations are not privately owned or protected in any way. The field studies did not involve protected species. All applicable international, national, and/or institutional guidelines for the care and use of animals were followed.

Fungus-storing organs of adults and larvae
Living adults and larvae were dissected using fire-sterilized tweezers under a stereo-microscope. The presence or absence of the mycetangia in adults and mycetomes in larvae were recorded. Images of them were photographed using an EOS Kiss X8i digital camera (Canon, Tokyo, Japan). The length of either the left or right mycetangium was measured using ImageJ 1.47t [26] for female adults used for yeast isolation except for Fs1 and Fs2, and the relative mycetangial length was calculated as the mycetangial length/elytral length. These mycetangia and mycetomes were further used for yeast isolation.
In addition, living adults from Nigorigo were dissected, the mycetangia were removed using fire-sterilized tweezers, and their contents were observed by microscopy.

Micro-CT observation of fungus-storing organs of adults
Antennae and legs of a living female adult (elytral length: 12.04 mm, weight: 181.2 mg) obtained from Dando were cut using tweezers. Then, the body was fixed with Carnoy solution for 2 days at room temperature, stained with 25% Lugol's solution for 4 days at room temperature, and embedded in 0.5% agarose gel [27].
To observe the mycetangial structure, first, we scanned the abdomen without dissection. Because aggregation of membranous organs around the base of the ovipositor made it difficult to distinguish mycetangia, we carefully removed eggs, ovaries, the digestive tract, and spiculum ventrale of sternum VIII using tweezers under a stereo-microscope. The dissected genitalia was restained with 25% Lugol's solution for 1 day at room temperature, embedded with 0.5% agarose gel, and scanned. As shapes and positions of mycetangia were not different between before and after dissection, scanned data of dissected genitalia were used for analyses.
The dissected genitalia in a stand-mounted 200-μm micropipette tip (BM Equipment, Tokyo, Japan) was scanned using the ScanXmate-E090S105 (ComscanTechno, Kanagawa, Japan). The genitalia was rotated 360˚in steps of 0.24˚, generating 1500 projection images of 992 × 992 pixels. The X-ray source was set at 85 kV and 90 μA. The scan data were reconstructed at an isotropic resolution of 2.8 μm, and converted into a tiff image dataset using con-eCTexpress software (ComscanTechno). Digital cross-sections and 3D models were made using OsiriX MD (Pixmeo, Bernex, Switzerland) and Imaris v9.8 (Bitplane, Zurich, Switzerland). Finally, supplemental movies were edited using Adobe Premiere Pro (Adobe, San Jose, CA, USA) (S1 and S2 Movies).

Yeast isolation from adults, larvae, and eggs
For adults, a mycetangium was removed from the ovipositor using fire-sterilized tweezers, surface-washed with sterile water for 10 s twice, and placed in a 2-mL tube containing 1000 μL of sterile water. It was cut into small pieces using a fire-sterilized injection needle and vortexed vigorously.
For larvae, mycetomes were removed from the guts, surface-washed with sterile water for 10 s three times, and placed in a 1.5-mL tube containing 500 or 1000 μL of sterile water. They were ground using an autoclaved pestle and vortexed vigorously.
For eggs, each of them was directly placed in a 2-mL tube containing 1000 μL of sterile water, and vortexed vigorously.
Then, 50 μL of the suspension was spread over potato dextrose agar (PDA) (Difco, Detroit, MI, USA) plates (9-cm diameter) containing 20 μg/mL of rifampicin (FUJIFILM Wako Pure Chemical, Osaka, Japan). Three replicates were made for each of the samples from adults and larvae, and one for each of the egg samples. The plates were incubated at 25˚C in the dark until yeast colonies appeared. The fungal colonies that grew on the plates were roughly classified based on their morphological traits (morphotype), and the number of colonies of each morphotype (colony forming units = CFU) was counted. When too many colonies were present on the plates, 10-and 100-fold dilutions were subsequently made using the original suspension stored at 5˚C, and the above-mentioned microbial isolation was reconducted using the diluted solutions. Note that the CFU values were not comparable among samples due to variable storing periods (0 to 92 days) of suspension before use. Four or eight colonies per morphotype were selected arbitrarily for subsequent culturing and DNA analysis.

DNA sequencing analysis of yeasts
To identify yeasts isolated from L. ochraceofasciata, DNA sequences (ca. 600 bps) in the D1/ D2 domain of the 26S rRNA (26S) gene were determined. In addition, to estimate their phylogenetic positions, DNA sequences of the internal transcribed spacer region and 5.8S rRNA (ITS/5.8S) gene (ca. 600 bps) and those of the translation elongation factor-1α (TEF) gene (ca. 800 bps) were determined for the representative isolates.
The PCR products were purified using Exo SAP-IT (Thermo Fisher Scientific, Waltham, MA, USA), and directly sequenced using BigDye Terminators (Thermo Fisher Scientific) and ABI PRISM 3130xl, 3730xl Genetic Analyzer (Thermo Fisher Scientific). Nucleotide sequence data reported in this study have been deposited in the DNA Data Bank of Japan (DDBJ) with accession numbers LC732211-LC732272 (see Table 3). The sequences were subjected to BLASTn searches for identification [32]. For Scheffersomyces yeasts, multiple alignments of the nucleotide sequences were generated using the program ClustalW in MEGA X [33]. Molecular phylogenetic analysis was conducted by the neighbor-joining method using MEGA X with 1000 bootstrap replicates.

Identification of larvae
To identify larvae collected in this study, DNA sequences in the mitochondrial cytochrome oxidase subunit I (COI) gene (658 bps) were determined. DNA was extracted from muscle tissues of adults and heads of larvae using PrepMan Ultra Reagent (Life Technologies, Warrington, UK). The following primer pair was used for PCR: LCO1490 (5'-GGTCAACAAATCAT AAAGATATTGG-3') (forward) and HCO2198 (5'-TAAACTTCAGGGTGACCAAAAAATCA-3') (reverse) [34]. Purification of PCR products and sequencing were conducted in the abovementioned manners (accession numbers: LC733218-LC733232 for adults, LC733233-LC733237 for larvae) (see Table 2). Then, the DNA sequences of larvae were compared with those of adults. A neighbor-joining phylogenetic tree was constructed using MEGA X with 1000 bootstrap replicates.

Carbon assimilation test
The representative isolate (strain name: Fo1-1-1) of the yeast deriving from a female (Fo1) originating from the Kaida Highlands was cultured aerobically in 20 mL of yeast nitrogen base (YNB) (Difco) containing 0.5% glucose at 25˚C in the dark for 2 to 3 days with shaking at 85 rpm. The culture media were centrifuged and cell pellets were suspended in sterile water.
To investigate the relationship between CFU and turbidity of the yeast examined, five types of yeast suspensions (OD 600 = 0.05, 0.10, 0.50, 1.00, 1.25) were made. For each type of suspension, 50 μL of each dilution (i.e., 1/10 2 , 1/10 3 , 1/10 4 equivalent of the original suspension) was spread onto a PDA plate (9-cm diameter) containing 20 μg/mL of rifampicin (3 replicates). The plates were incubated at 25˚C in the dark for 2 days and the CFUs were counted.
For the carbon assimilation test, we made a suspension in which OD 600 was adjusted to 0.10-0.12. Then, 50 μL of the cell suspension was added into a tube (2 mL . The concentration of each carbon source was 0.5 g/L, except for xylan, at 1.5 g/L. As xylan from beech is insoluble, a high concentration of beechxylan was used. To determine whether the assimilating ability is different between beech-and corn-xylans, the concentration of corn-xylan was adjusted to 1.5 g/L. The tubes were shaken at 85 rpm and incubated at 25˚C in the dark for 7 days. Afterwards, OD 600 was recorded to determine the growth of each strain. The degree of assimilation was scored according to the difference in the turbidity increase (ΔOD 600 ) between culture media containing no and a given carbon source as follows: no growth (ΔOD 600 < 0.03), weak growth (0.03 � ΔOD 600 < 0.10), moderate growth (0.10 � ΔOD 600 < 0.40), strong growth (0.40 � ΔOD 600 < 1.00), and very strong growth (1.00 � ΔOD 600 ) [7].

Statistical analysis
Pearson's correlation coefficient and the ordinary least squares (OLS) method were used to determine the relationship between two variables. Calculations were performed using R 3.5.1 [35].

Fungus-storing organs of adults
All female adults of L. o. ochraceofasciata and L. o. ochrotela (body length: mean ± SD = 17.70 ± 1.19 mm, range = 16.20 to 19.98 mm, n = 12; elytral length: 12.32 ± 0.82 mm, range = 10.46 to 13.41 mm, n = 12; weight: 192.6 ± 37.8 mg, range = 115.2 to 247.7 mg, n = 12) had a pair of membranous, symmetrical, tube-like mycetangia (length: 2.85 ± 0.17 mm, range = 2.57 to 3.09 mm, n = 10; relative mycetangial length: 0.23 ± 0.02, range = 0.21 to 0.27, n = 10) at the base of the ovipositor (Fig 1B, S1 Table). The blind end of each mycetangium and anterior end of the spiculum ventrale of sternum VIII were connected by thin muscle tissues as reported in other species [16,17] (Fig 1B). Note that we did not record the presence/absence of a secretion gland open to the mycetangia. When a mycetangium was removed from the ovipositor, a whitish fluid came out. In a mycetangium, cysts containing large numbers of yeast cells were abundant (n = 2) (Fig 1C and 1D). Some of these yeast cells were budding (Fig 1D).

Micro-CT observation of fungus-storing organs of adults
Micro-CT revealed that paired tube-like mycetangia were located at the basal part of the ovipositor (Fig 2, S1 and S2 Movies), as observed by dissection under a stereo-microscope ( Fig 1B). Each mycetangium was S-shaped: the basal part curved posteriorly and ventrally, the mid-part was located between the lateral oviduct and vagina and curved anteriorly with surrounding the lateral oviduct, and the blind end connected with muscle tissues along the spiculum ventrale of sternum VIII (Fig 2, S2 Movie). The 3D images of the left and right mycetangia were asymmetrical in shape and position within the body (Fig 2C-2E, S2 Movie).

Identification of larvae
The DNA sequences (658 bps) for the COI gene were 99.5-100% identical to each other among five larvae and closest to that of the L. o. ochraceofasciata adult (99.7-99.8% identity).
Due to multiple alignments of the DNA sequences of the sampled larvae and lepturine adults, DNA sequences of 532 bps were used for phylogenetic analyses. The analyses revealed that all larvae formed a clade with L. o. ochraceofasciata (Fig 3). Thus, they were concluded to be L. o. ochraceofasciata.

Behaviors of hatched larvae
We observed that the larvae fed on eggshells during and immediately after hatching (n = 23) (Fig 1H).
Due to multiple alignments of the DNA sequences of the studied and reference yeasts, 570 bps (26S), 618 bps (ITS/5.8S), and 584 bps (TEF) were used for the phylogenetic analyses. The  analyses revealed that the isolated Scheffersomyces yeasts were conspecific with Sc. insectosa in the Scheffersomyces clade (Fig 4).
To determine the assimilation of wood-associated carbon sources by the yeast symbiont, 16 different carbon sources including a treatment with no carbon source were tested using liquid media. When the yeast was cultured with no carbon source for 7 days, the turbidity increase was 0.00. The yeast assimilated corn-xylan very strongly and glucose, galactose, mannose, xylose, rhamnose, fructose, sucrose, cellobiose, and beech-xylan strongly (Table 4). Arabinose, galacturonic acid, glucuronic acid, mannan and carboxymethyl cellulose were not assimilated (Table 4).

Discussion
All female adults of L. ochraceofasciata obtained from multiple locations including two subspecies (L. o. ochraceofasciata and L. o. ochrotela) harbored only one species of yeast, Sc. insectosa in their mycetangia. All larvae of L. ochraceofasciata had mycetomes and harbored Sc. insectosa exclusively in them. Scheffersomyces insectosa was also recovered from all eggs examined. Scheffersomyces insectosa assimilated wood-associated sugars, including xylose, cellobiose, and xylan in culture. These results strongly suggest an intimate association between L. ochraceofasciata and Sc. insectosa through the insect's life history. Scheffersomyces insectosa may benefit from the vectoring activity of L. ochraceofasciata from wood to wood. Given that only larvae of L. ochraceofasciata consume wood in its developmental stages, Sc. insectosa may help the larvae to digest wood. This is the first reported example of a lepturine-yeast association in Asia.
In mycetangia-bearing insects, there are many types of mycetangia, such as those showing differences in shape (pit, sac, or setal-brush) and presence/absence of secretion glands [10,36]. Mycetangium sensu stricto represents a glandular type only, while mycetangium sensu lato includes a nonglandular one. In the present study, the type of mycetangia of L. ochraceofasciata was undetermined. Thus, it is tentatively used in a broad sense.
The 3D structure of ovipositor-associated mycetangia reconstructed based on micro-CT images revealed that the tube-like mycetangia are bendable in the beetle's body. During oviposition, a female insect exposes her ovipositor to insert it into crevices of oviposition substrate and retract it after egg deposition (MK, FM, WT personal observation). The muscle bundle connected between the apex of the mycetangium and sclerotized spiculum ventrale of sternum VIII (Figs 1B and 2, S2 Movie). Thus, it is considered that a mycetangium is usually S-shaped inside the body and that it is tensioned when the ovipositor is protracted, resulting in the secretion of yeast cells. Symbiont-loading from mycetangia onto the egg surface may be synchronized with movement of the ovipositor for egg deposition.
In two European lepturines, Oxymirus cursor (Linnaeus) and R. mordax, their yeast symbionts are transmitted from mother to offspring via the surface of eggs: hatched larvae acquire yeast symbionts by ingesting the eggshells on which the yeast symbionts are present [16,17]. In L. ochraceofasciata, similarly, the hatched larvae feed on their yeast-present eggshells. Vertical transmission mechanism of yeast symbionts may have been conserved in Lepturinae.
Interestingly, Sc. insectosa was isolated from mycetangia of one female before she emerged from her pupal chamber and two females immediately after they emerged from the natal wood. This repeated isolation of Sc. insectosa from newly eclosed adults suggests that female adults of L. ochraceofasciata incorporate Sc. insectosa into their mycetangia within pupal chambers. Meanwhile, budding yeast cells were present within mycetangia of female adults that were collected on flowers (Fig 2C). It is likely that Sc. insectosa reproduces within mycetangia after the female adults emerge from the natal wood.
In the carbon assimilation test, Sc. insectosa assimilated various sugars including woodassociated mono-, di-, and polysaccharides. Particularly, this yeast showed marked ability to assimilate corn-and beech-xylans. These data must be interpreted with caution, however, because the concentrations of these xylans were higher than the other carbon sources. This high concentration of xylans would cause relatively strong growth of the yeast. Nevertheless, the L. ochraceofasciata-associated Sc. insectosa evidently assimilates xylans and the degree of assimilation ability may vary among xylans. In contrast, Sc. insectosa (strain: SICYLG3) isolated from the gut of larvae of Sinodendron cylindricum (Linnaeus) (Lucanidae) assimilates xylan weakly [37]. Larvae of Si. cylindricum inhabit decayed wood [37], while those of L. ochraceofasciata can utilize physically hard wood (FM, WT personal observation), in which hemicelluloses including xylan are likely to exist abundantly. This physiological difference between yeast strains may be related to the microhabitats of their insect hosts.
Female adults and larvae of L. ochraceofasciata harbored Sc. insectosa exclusively in their fungus-storing organs, whereas another yeast, Meyerozyma caribbica-like yeast was detected from eggs together with Sc. insectosa. Given that M. caribbica is found widely in various artificial and natural environments [19] and that the oviposition experiment in this study was conducted under unsterilized conditions, it is likely that the presence of Meyerozyma sp. on eggs is due to contamination during the experiment.
Leptura ochraceofasciata constantly and exclusively possessed Sc. insectosa with marked abundance, whereas the yeast has been isolated from other xylophagous beetles: mycetangia of a female adult of the European lepturine, Sti. maculicornis (DeGeer) (formerly, L. maculicornis) [18,38], and the gut of larvae of the lucanid Si. cylindricum obtained from decayed wood of a beech tree in Switzerland [37]. These suggest an asymmetrical interdependence between L. ochraceofasciata and Sc. insectosa. Leptura ochraceofasciata might obligatorily depend on Sc. insectosa, but Sc. insectosa might facultatively depend on L. ochraceofasciata. Alternatively, at a local scale, Sc. insectosa might obligatorily depend on L. ochraceofasciata. In Japan, L. ochraceofasciata is one of the most common lepturines in primary and secondary forests from low to high elevation areas [21]. On the other hand, Sc. insectosa is a rare yeast and has been isolated from xylophagous beetles twice in Europe [18,37,38]. Further study is required to determine to what extent Sc. insectosa depends on L. ochraceofasciata by examining many wood-inhabiting insects living sympatrically with L. ochraceofasciata.