Cinnamyl alcohol: host-recognizing kairomone of the ower thrips Frankliniella intonsa

Flower-inhabiting thrips nd hosts using olfactory and visual cues. In this study, we report the identication of a plant-produced kairomone of the ower thrips Frankliniella intonsa (Trybom), an important agricultural pest in northeast Asia. GC-MS analysis of solid-phase microextraction samples from blueberry owers, Vaccinium corymbosum L., that mediate the attraction of adult F. intonsa revealed that the major component was cinnamyl alcohol, followed by cinnamyl acetate, cinnamaldehyde, germacrene D, β-bourbonene, β-caryophyllene, and benzyl benzoate. The biological activity of the oral compounds was investigated using commercial cinnamaldehyde, cinnamyl alcohol, β-caryophyllene, cinnamyl acetate, and benzyl benzoate in hot pepper (Capsicum annuum L.) elds. Signicantly more F. intonsa males and females were caught in red delta traps with cinnamyl alcohol than in all other traps. Cinnamaldehyde and cinnamyl acetate attracted adult F. intonsa but were not as attractive as cinnamyl alcohol. β-Caryophyllene and benzyl benzoate were not attractive. Furthermore, the addition of four minor components to cinnamyl alcohol did not result in increased trap catches relative to cinnamyl alcohol alone, indicating that cinnamyl alcohol is responsible for attracting adult F. intonsa toward blueberry owers. Therefore, this phenylpropanoid could be used as a selective, effective lure for monitoring and controlling F. intonsa.


Introduction
Thrips (Thysanoptera: Thripidae) are economically important pests of various agricultural crops in northeast Asia, causing damage by feeding and through the transmission of plant viruses (Okazaki and Sakurai, 2005;Moon et al., 2006;Wei et al., 2012). To control thrips, growers have depended exclusively on insecticide applications at regular intervals, but this can result in poorly timed or unnecessary insecticide applications. In addition, the heavy reliance on broad-spectrum insecticides to control these pests has resulted in insecticide resistance (Bielza, 2008;Cho et al., 2018). Consequently, alternative control methods are urgently needed. Volatile semiochemicals could be used to monitor the seasonal abundance of thrips and to control thrips by mass trapping and attract-and-kill tactics (Kirk et al., 2021).
In the course of a eld study of crop damage caused by insect pests, we serendipitously observed the attraction of large numbers of the ower thrips Frankliniella intonsa (Trybom), an important pest of hot pepper in Korea (Moon et al., 2006), to highbush blueberry Vaccinium corymbosum L. (cultivar Darrow) owers, suggesting the existence of a host kairomone that mediates the attraction of adult F. intonsa to the owers. Here, we report the results of chemical analyses of solid-phase microextraction samples collected from blueberry owers and evaluate the biological activity of the identi ed compounds in hot pepper elds.

Collection of headspace volatiles
Solid-phase microextraction (SPME) was used to collect headspace volatiles of highbush blueberry owers. An SPME ber (100-µm polydimethylsiloxane coating; Supelco, Bellefonte, PA, USA) was conditioned before use for 30 min in a gas chromatography injector (250°C). The blueberry owers were cut and immediately introduced into a 500-mL glass jar. The SPME ber was then inserted into the jar through a small hole in the cap. After a 3-h exposure, the ber was immediately subjected to gas chromatography-mass spectrometry (GC-MS).

Chemical analysis
The SPME-collected volatiles were analyzed on an Agilent 8890 GC system (Santa Clara, CA, USA) interfaced to an Agilent 5977B mass-selective detector. Samples were run on a DB-5MS column (60 m × 0.25 mm × 0.25 µm lm; J&W Scienti c, Folsom, CA, USA). The injector and transfer line temperatures were 250°C. To start the GC-MS analysis, the loaded SPME bers were desorbed for 1 min in the injection port. The GC oven temperature was programmed from 80°C (1-min hold) to 200°C at 5°C/min and held for 10 min. Helium was the carrier gas (1 mL/min) and the injections were splitless. Electron impact mass spectra were monitored at 70 eV in a mass range of 40-300 amu, and the source was maintained at 230°C. Compounds were identi ed by matching mass spectra to the Wiley Mass Spectral Database; these identi cations were con rmed by comparison of the retention times and mass spectra with those of authentic standards.

Field trials
Field trials were conducted in 'Bangtanbog' hot pepper (Capsicum annuum L.) elds in Kochang, South Korea (35.3°N, 126.4°E), during July and August 2021. Trial 1 investigated the attraction of adult F. intonsa to cinnamaldehyde, cinnamyl alcohol, β-caryophyllene, cinnamyl acetate, and benzyl benzoate singly. Trial 2 tested the attraction of adult F. intonsa to cinnamyl alcohol alone or in combination with cinnamaldehyde, β-caryophyllene, cinnamyl acetate, or benzyl benzoate in the ratio found in the blueberry ower extract. Blank traps served as negative controls in all experiments and traps baited with a mixture of the ve compounds were used as a positive control in Trial 2.
Test compounds were dissolved in 100 µL diethyl ether and then stabilized by adding 1% butylated hydroxytoluene as an antioxidant. Candidate attractants were loaded into pink polyethylene sachets (2.5 × 6.0 cm, 100 µm wall thickness; AD Corp., Andong, South Korea) and heat-sealed. Red delta traps containing a white adhesive-coated panel at the base (Green Agro Tech, Gyeongsan, South Korea) were baited with the sachets. Red delta traps were used because of their relative low catch of non-target species (El-Sayed et al., 2009). The delta traps were suspended from metal poles at the height of the red pepper canopy. The eld test used a completely randomized block design with four replicates of each treatment. Traps were placed ~15 m apart within each block and the distance between blocks was approximately 50 m. Sticky bases were removed at 3-to 4-day intervals during each study. Thrips captured in the traps were identi ed and counted under a stereoscopic microscope.
Because the trap catch data were not normally distributed, differences were analyzed via nonparametric Kruskal-Wallis ANOVA followed by comparison of means using the Wilcoxon paired sample test (SAS Institute, 2014). In all analyses, α = 0.05. Males and females were analyzed separately.

Field trials
The eld trial testing single compounds showed that signi cantly more females and males of F. intonsa were caught in traps with cinnamyl alcohol than in all other traps (females, x² = 19.6703, P = 0.0014; males, x² = 15.5587, P = 0.0082) (Fig. 1). Cinnamyl alcohol caught 42 times more females and 33 times more males than controls. Traps with cinnamaldehyde and cinnamyl acetate also caught more F. intonsa females than traps without volatiles (Fig. 1A). By contrast, there were no differences in the number of male thrips caught in traps baited with cinnamaldehyde or cinnamyl acetate and the unbaited control (Fig. 1B). This may be due to the relatively low numbers of males caught during the summer. The catches of male and female F. intonsa in traps baited with β-caryophyllene and benzyl benzoate did not differ from the catch in control traps.
Another eld trial testing possible synergistic effects between cinnamyl alcohol and the other chemicals identi ed in blueberry owers caught thousands of thrips, again indicating the strong attractiveness of cinnamyl alcohol (females, x² = 13.6961, P = 0.0332; males, x² = 15.4339, P = 0.0171) (Fig. 2). However, cinnamaldehyde, β-caryophyllene, cinnamyl acetate, and benzyl benzoate did not increase the trap catches when added to cinnamyl alcohol. Moreover, the full ve-component blend did not increase the capture of adult male or female thrips relative to cinnamyl alcohol alone.

Discussion
Our data con rmed that cinnamyl alcohol is the major component of oral volatiles of highbush blueberry (Rodriguez-Saona et al., 2011). In the eld, we found that traps with cinnamyl alcohol caught more adult F. intonsa than traps with the other compounds. Cinnamaldehyde and cinnamyl acetate also increased the trap capture of female F. intonsa, but not to the same extent as cinnamyl alcohol. In attractant eld-screening experiments in the United States, Morgan and Crumb (1928) found that signi cantly more adults of Frankliniella tritici Fitch were caught in traps with cinnamaldehyde than in traps with cinnamyl alcohol in an oat eld. The large difference between the two Frankliniella species in their responses to cinnamyl alcohol and cinnamaldehyde points to the functional and behavioral signi cance of oral scents in host location by ower-inhabiting thrips.
In the eld trial testing chemical interactions between cinnamyl alcohol and the other compounds identi ed from blueberry owers, the four minor compounds had no synergistic activity with cinnamyl alcohol alone. The lack of synergism from compound mixtures is similar to that observed for Thrips obscuratus (Crawford) (El-Sayed et al., 2009, 2014 and Thrips tabaci Lindeman (Teulon et al., 2007). This suggests relatively simple chemosensory input with single receptors for generic compounds or poor sensory integration between receptor neurons in thysanopteran insects.
It is unclear whether the response to oral volatiles differs between the sexes of ower-dwelling thrips. We observed that red delta traps with cinnamyl alcohol caught more male and female adult F. intonsa than delta traps without cinnamyl alcohol in hot pepper elds. Our result supports the nding by Teulon et al. (1993Teulon et al. ( , 1999) that the addition of oral volatiles to traps signi cantly increases the capture of adult male and female thrips. Conversely, in a Y-tube olfactometer test, Cao et al. (2018) reported that female Frankliniella occidentalis preferred host plant oral volatiles, while male F. occidentalis did not show a preference for any oral odors. Thus, further tests with different volatile blends, and with different experimental conditions (e.g., distance to source and wind velocity) are required to clarify the sex-speci c differences of the olfactory response to oral volatile components in thrips.
Many thrips species integrate visual information with olfactory information speci c for host plant selection (Teulon et al., 1999;van Tol et al., 2020). Adult F. intonsa are predominantly attracted to blue and white over other colors (Seo et al., 2006;Mao et al., 2018). Besides cinnamyl alcohol identi ed here, the aldehydes p-anisaldehyde and benzaldehyde attracted F. intonsa during eld screening tests (Kirk, 1985;Teulon et al., 1993). Consequently, it will be interesting to investigate the interaction between trap color and the presence of odors to obtain more effective monitoring and control agents for this thrips. Figure 1 Gas chromatography-mass spectrometry (GC-MS) total ion chromatograms of SPME-collected volatiles from highbush blueberry owers on a DB-5MS column.

Figure 3
The effect of adding four minor compounds to the standard (cinnamyl alcohol; 10 mg) in the ratio found in blueberry ower extract on the capture of Frankliniella intonsa females (A) and males (B) in red pepper elds in Kochang, South Korea, 28 July to 3 August 2021 (n = 5). Bars with the same letter are not signi cantly different based on the Wilcoxon paired sample test (P > 0.05).