Aggregation Pheromone of the Guam Strain of Coconut Rhinoceros Beetle, Oryctes Rhinoceros, and Revision of Stereochemistry

The coconut rhinoceros beetle, Oryctes rhinoceros (Linnaeus (Coleoptera: Dynastinae) (CRB), is endemic to tropical Asia where it damages both coconut and oil palm. A 48 new invasion by CRB occurred on Guam in 2007 and eradication attempts failed using 49 commonly applied O. rhinoceros nudivirus (OrNV) isolates. This and subsequent invasive 50 outbreaks were found to have been caused by a previously unrecognized haplotype, CRB-G, 51 which appeared to be tolerant to OrNV. The male-produced aggregation pheromone of the 52 endemic, susceptible strain of O. rhinoceros (CRB-S) was previously identified as ethyl 4- 53 methyloctanoate. There were anecdotal reports that the commercial lures containing this 54 compound were not attractive to CRB-G and the aim of this work was to identify the pheromone 55 of CRB-G. Initial collections of volatiles from virgin male and female CRB-G adults failed to 56 show any male- or female-specific compounds as candidate pheromone components. Only after 57 five months were significant quantities of ethyl 4-methyloctanoate and 4-methyloctanoic acid 58 produced by males but not by females. No other male-specific compounds could be detected, in 59 particular methyl 4-methyloctanoate, 4-methyl-1-octanol or 4-methyl-1-octyl acetate, 60 compounds identified in volatiles from some other species of Oryctes . Ethyl 4-methyloctanoate 61 elicited a strong electroantennogram response from both male and female CRB-G, but these 62 other compounds, including 4-methyloctanoic acid, did not. The enantiomers of ethyl 4- 63 methyloctanoate and 4-methyloctanoic acid were conveniently prepared by enzymatic resolution 64 of the commercially-available acid, and the enantiomers of the acid, but not the ester, could be 65 separated by gas chromatography on an enantiospecific cyclodextrin phase. Using this approach, both did not. Reasons for the previous misassignment of the configuration of the components of the 75 CRB-S pheromone are discussed along with the practical implications of these results.


Declarations
Introduction 81 The coconut rhinoceros beetle, Oryctes rhinoceros (Linnaeus 1758) (Coleoptera: Scarabaeidae: 82 Dynastinae) (CRB), is endemic to tropical Asia where it damages both coconut and oil palm, and 83 can kill palms when adults bore into crowns to feed on sap (Bedford, 2013a(Bedford, , 2013b. The beetle 84 was introduced into the Pacific in 1909 via potted soil with rubber tree seedlings from Sri Lanka 85 (Catley, 1969). The pest rapidly multiplied and spread throughout the Pacific Islands. However, 86 it was brought under control by the discovery and distribution of a viral biocontrol agent, Oryctes 87 rhinoceros nudivirus (OrNV), previously known as Rhabdiovirus oryctes and Baculovirus 88 oryctes (Bedford, 2013b;Huger, 2005). Further spread of CRB was suppressed for 30 years 89  Marshall et al. (2017) reported that all of these outbreaks have been 96 caused by a previously unrecognized haplotype, CRB-G, which appeared to be tolerant to OrNV. 97

Collection of Volatiles 129
Volatiles were collected from 1-7 individuals. Most collections were made from individuals of 130 one sex, but some were made from both sexes together. These were placed in a 5-litre round-131 bottomed, bolt-head flask with shredded paper and a piece of sugarcane. Some collections were 132 made with banana or tinned palm hearts as food, but these produced larger amounts of volatiles 133 than sugarcane. Air was drawn into the flask (2 l/min) through a filter containing activated 134 charcoal (20 cm x 1.5 cm; 6-10 mesh) and out through a filter made from a Pasteur pipette (4 135 mm i.d.) containing Porapak Q (200 mg; 50-80 mesh; Supelco, Gillingham, Dorset, UK). The 136 Porapak Q was extracted with chloroform in a Soxhlet apparatus for 6 h and washed with 137 dichloromethane immediately before use. Collections were made for periods from 24 h -72 h. 138 Trapped volatiles were removed with dichloromethane (Pesticide Residue Grade; 1 ml) and 139 stored at 4 ºC before analysis. 140 To convert esters in the collections of volatiles to the corresponding acids, 0.5 ml of a 141 collection containing approximately 50 µg ethyl 4-methyloctanoate was evaporated just to 142 dryness under a gentle stream of nitrogen. Ethanol (50 µl) and 2N aqueous potassium hydroxide 143 solution (50 µl) were added and the mixture left for 6 h at room temperature. Water (0.5 ml) was 144 added and the mixture was extracted with hexane (0.5 ml). After acidification with aqueous 4N 145 sulfuric acid (50 µl), the mixture was extracted with two portions of diethyl ether (0.5 ml) and 146 the combined extracts dried with anhydrous magnesium sulfate. 147

Analysis by Gas Chromatography 148
Collections of volatiles were analyzed by gas chromatography (GC) with flame ionization 149 detection (FID) using HP6850 instruments (Agilent Technologies, Stockport, Cheshire, UK) 150 fitted with a fused silica column (30 m x 0.32 mm i.d. 0.25 μ film thickness) coated with non-151 polar HP5 (Agilent) or polar DBWax (Supelco). The oven temperature was programmed from 50 152 °C for 2 min than at 10 °C/min to 250 °C. Carrier gas was helium (2.4 ml/min), injection was 153 splitless (220 °C) and detection by FID (250 °C). 7 split injection (220 °C; 20:1) and FID (220 °C). The oven temperature was programmed from 60 157 °C for 2 min then at 5 °C/min to 200 °C. 158 Collections were also analyzed by GC coupled to mass spectrometry (MS) using a Varian 159 3700 GC linked directly to a Saturn 2200 ion-trap MS (Varian, now Agilent). Columns (30 m x 160 0.25 mm i.d. 0.25 μ film thickness) were coated with polar DBWax (Supelco) or non-polar VF5 161 (Varian/Agilent). The carrier gas was helium (1 ml/min) and the oven temperature was 162 programmed from 40 °C for 2 min then at 10 °C/min to 250 °C. 163 Retention Indices for compounds were calculated relative to the retention times of n-164 alkanes. Amounts present in collections were estimated by comparison of peak areas with those 165 of external standards. The response factor for 4-methyloctanoic acid was much lower than that of 166 the corresponding acid in both GC-FID and GC-MS. 167

EAG) 169
GC-EAG Analyses were carried out on a HP6890 GC (Agilent) fitted with flame ionization 170 detector (FID) and fused silica capillary columns (30 m x 0.32 mm i.d. x 0.25 µm film thickness) 171 coated with DBWax and DB5 (Supelco). Injections onto the DBWax column were in splitless 172 mode (220 °C), carrier gas was helium (2.4 ml/min) and the oven temperature was programmed 173 from 50 °C for 2 min and then at 10 ºC/min or 20 °C/min. to 250 °C for 3 min. The effluents of 174 the two columns were combined with a glass push-fit Y-tube connector (Agilent) connected to a 175 second Y-tube connector with deactivated fused silica tubing (10 cm x 0.32 mm i.  Table 2. 256 Traps were deployed in randomized complete blocks with 50 m between traps in a block 257 and at least 100 m between blocks. There were six replicate blocks in each experiment and 258 catches were sexed, counted and discarded weekly. Total catches were transformed to log(x+1) 259 and subjected to analysis of variance, omitting treatments with zero catches. Where significant 260 differences (P < 0.05) were indicated, differences between means were tested for significance (P 261 < 0.05) by a Least Significant Difference (LSD) test. enantiomer of the acid, confirming that the ester also had the (R)-configuration (Fig. 3). 290

EAG) 292
In GC-EAG analyses of early collections of volatiles from male or female CRB-G beetles, no 293 consistent EAG responses were detected from antennae of male or female beetles (N = 6). 294 In GC-EAG analyses of synthetic standards a consistent response was observed from 295 antennae of both males and females to ethyl 4-methyloctanoate. A response was occasionally observed to the corresponding methyl ester, but no response was observed to 4-methyl-1-octanol, 297 the corresponding acetate or 4-methyloctanoic acid (Fig. 4). 298 EAG responses from antennae of both males and females were generally greater to ethyl 299 (R)-4-methyloctanoate than to the (S)-enantiomer (Supplementary Material Fig. S7). This was 300 not examined rigorously as available samples of both enantiomers contained small amounts of 301 the opposite enantiomer. However, in one example, mean responses (N = 2 delivered alternately) 302 from the antenna of a male beetle were 0.28 mV and 0.14 mV to the (R)-and (S)-enantiomers 303 (10 ng), respectively, and from the antenna of a female beetle 0.32 mV and 0.10 mV, 304 respectively. 305

Pheromone of Susceptible Strain of Oryctes rhinoceros (CRB-S) 306
Beetles the ChemTica lure, although not significantly higher than those baited with racemic ethyl 4-329 methyloctanoate in a polyethylene vial, even though the release rate was almost an order of 330 magnitude greater (9.1 mg/d and 1.0 mg/d, respectively, Table 1). 331 In the second experiment (Fig. 6) Stereospecific alkylation with a protected 3-iodo-1-propanol was followed by removal of the 416 auxiliary by hydrolysis, chain extension and oxidation. 417 In our work, the enantiomers of 4-methyloctanoic acid and the corresponding ethyl ester 418 were conveniently resolved enzymatically in multigram quantities as reference standards and for 419 field testing. Heinsman et al. (1998;2001) reported that immobilized lipase from Candida 420 antarctica acts selectively on the (R)-enantiomer, so that the acid can be selectively esterified in 421 ethanol or the ethyl ester can be selectively hydrolyzed in aqueous buffer. Moreover, the acid 422 could be separated from the ester by extraction with aqueous base and subsequent acidification 423 without resort to chromatography. The stereoselectivity of this lipase for the enantiomers of 4-424 methyloctanoic acid was not as high as for 2-alkanols and the corresponding acetates (Hall et al. 425 2012), but two cycles produced material with at least 94% ee. 426

Pheromone Production 427
In our hands, beetles of both CRB-G and CRB-S strains did not start producing pheromone for 428 several months after emergence and arrival at NRI. In attempts to induce pheromone production, 429 various substrates including palm detritus, soil and paper were tried, as were various foodstuffs 430 including banana, apple, palm hearts and sugarcane, without any obvious effects. Beetles were 431 kept under laboratory conditions and did not fly, so whether this pattern is replicated in the field 432 is unknown. This has not been reported for other Oryctes species, although Prof Didier Rochat 433 did suggest that pheromone production can be very variable and erratic (pers comm 2018).