Synthesis and Insecticidal Activity of Fire Ant Venom Alkaloid-Based 2-Methyl-6-alkyl-Δ1,6-piperideines

2,6-dialkylpiperideines found in the venom of Solenopsis (Hymenoptera, Formicidae) fire ants are a range of compounds possessing various biological activities. A series of racemic 2-methyl-6-alkyl-Δ1,6-piperideines were synthesized for chemical confirmation of the natural products found in fire ant venom, and the evaluation of their biological activity. Synthetic Δ1,6-piperideines and the natural compounds in the cis-alkaloid fraction of Solenopsis invicta had identical mass spectra and retention times. Their insecticidal activities against the third-instar larvae of cotton bollworm (Helicoverpa armigera) were evaluated by using injection and topical application methods. All three compounds exhibited no lethal effect at concentrations of 0.05–0.4 mol/L by topical treatment, but moderate lethal effect at 0.4 mol/L through injection treatment. Compound 6a showed significantly higher activity than the natural insecticide nicotine. The differences in activity among compounds 6b, 6c and nicotine were not significant. The elongation of the carbon chain at the 6-position of the piperideine ring appears to decrease insecticidal activity.


Chemistry
Racemic 2-methyl-6-alkyl-Δ 1,6 -piperideines were prepared following the synthetic route shown in Scheme 1 [39]. The large-scale preparation of glutarimide (2) was readily carried out using glutaric acid (1) and urea as starting materials, which reacted neat at 175 °C. One of the carbonyl groups on the piperidinone ring was converted to a methyl group in compound 3 by a Grignard reaction of glutarimide (2) with CH3MgBr reagent, and subsequent reduction with NaBH3CN. A key step was the generation of NBocaminoketones 5a-c from Boc-protected compound 4, which was reacted with preprepared Grignard reagents to cause ring opening. Finally, the target compounds 6a-c were obtained by catalytic cyclization of compounds 5a-c with concentrated HCl. All copies of the spectra for compounds 2-4, 5a-c, and 6a-c are available in Supplementary Materials.

Confirmation of Δ 1,6 -Piperideines in Natural Venom
Fire ant venom alkaloids were readily extracted out of worker bodies and separated into cis-and trans-alkaloid fractions. The chemical identities of these alkaloids were determined by comparing the mass spectrum and the gas chromatographic retention time with published data. In both alkaloid fractions, two series of piperideines were isolated. Δ 1,6 -piperideines were identified by characteristic mass ions at m/z 96 and 111, whereas Δ 1,2 -piperideines were identified by characteristic mass ions at m/z 96, 97 and 110 [15,16]. Ultimately, synthetic compounds were directly compared with the natural alkaloids in the cis-alkaloid fraction, which have a well-established structure and stereochemistry. The retention times of Δ 1,6 -piperideine alkaloids from the natural venom are shown in Figure  1A. The mass spectra and retention times of synthetic compounds 6a-c were completely

Confirmation of ∆ 1,6 -Piperideines in Natural Venom
Fire ant venom alkaloids were readily extracted out of worker bodies and separated into cisand trans-alkaloid fractions. The chemical identities of these alkaloids were determined by comparing the mass spectrum and the gas chromatographic retention time with published data. In both alkaloid fractions, two series of piperideines were isolated. ∆ 1,6 -piperideines were identified by characteristic mass ions at m/z 96 and 111, whereas ∆ 1,2 -piperideines were identified by characteristic mass ions at m/z 96, 97 and 110 [15,16]. Ultimately, synthetic compounds were directly compared with the natural alkaloids in the cis-alkaloid fraction, which have a well-established structure and stereochemistry. The retention times of ∆ 1,6 -piperideine alkaloids from the natural venom are shown in Figure 1A. The mass spectra and retention times of synthetic compounds 6a-c were completely consistent with those of the corresponding natural ∆ 1,6 -piperideines. All three compounds had similar fragmentation patterns in the mass spectrum, with the intense ions at m/z 96 and 111 ( Figure 1B-D). These data support the identification of ∆ 1,6 -piperideine alkaloids in the venom of Solenopsis fire ants. The mass spectra of compounds 6a-c matched those of the corresponding 2-methyl-6-undecyl-∆ 1,6 -piperideine, 2-methyl-6-tridecyl-∆ 1,6 -piperideine and 2-methyl-6-pentadecyl-∆ 1,6 -piperideine in the literature [10,14,15]. All these mass spectra showed the same base peak ion at m/z 111. However, Chen et al. [14] reported a slightly different mass pattern, with a base peak ion at m/z 96 for 2-methyl-6-tridecyl-∆ 1,6piperideine.

Toxicity against Cotton Bollworm (Helicoverpa armigera)
The insecticidal activities of compounds 6a-c were evaluated by topical and injection treatments. As shown in Table 1, the mortality of piperideine 6b and 6c at various concentrations (0.05-0.4 mol/L) was similar to the standard control (nicotine) and the solvent control (acetone), indicating that these alkaloids were inactive against the early third-instar larvae of cotton bollworm at these test concentrations. Compound 6a, with a shorter side chain, exhibited a weak effect within 48 h, through topical application at concentrations of 0.2 and 0.4 mol/L. Table 1. Contact toxicities of 6a-c against the early third-instar larvae of cotton bollworm following topical application.

Toxicity against Cotton Bollworm (Helicoverpa armigera)
The insecticidal activities of compounds 6a-c were evaluated by topical and injection treatments. As shown in Table 1, the mortality of piperideine 6b and 6c at various concentrations (0.05-0.4 mol/L) was similar to the standard control (nicotine) and the solvent control (acetone), indicating that these alkaloids were inactive against the early third-instar larvae of cotton bollworm at these test concentrations. Compound 6a, with a shorter side chain, exhibited a weak effect within 48 h, through topical application at concentrations of 0.2 and 0.4 mol/L.
For the injection treatment, a concentration of 0.4 mol/L was selected. The three piperideines exhibited moderate activities against the third-instar larvae of cotton bollworm ( Table 2). Compound 6a, with an undecyl side chain, displayed the highest mortality rates of 45.5% (24 h) and 75.6% (48 h). Compounds 6b and 6c, with a longer side chain, showed relatively lower activity. The order of mortality from high to low is as follows: 6a > 6b > 6c, which means that the piperideines with a shorter side chain have better activity. Interestingly, as shown in Table 2, 6a displayed significantly higher activity in 48 h than nicotine, which has long been regarded as a natural insecticide [42]. At the same concentration, the toxicities of the piperideines in the subcutaneous injection were evidently greater than those in the topical administration, suggesting that the piperideines may possess weak penetrating capacity through the cuticle of the cotton bollworm larva. The significant difference in toxicity between topical and injection applications may explain why fire ant workers usually inject venomous poison into their prey while hunting. Table 1. Contact toxicities of 6a-c against the early third-instar larvae of cotton bollworm following topical application.

Compound
Concentration (  The alkaloids in fire ant venom are mainly hydrophobic piperidines, along with piperideines in much lesser amounts. Piperideines have long been thought to be unstable intermediates for the biosynthesis of piperidines [10,15,35], but these remain largely unstudied. The facile syntheses of these ∆ 1,6 -piperideines make it possible to study the biosynthesis of piperidines in fire ant venom. The mixed piperideine alkaloids from the extracts of the red imported fire ants show significant inhibitory effects on the growth of Pythium ultimum [36] and Clavibacter michiganensis subsp. michiganensis [41], and relatively high toxicity to the green peach aphid, Myzus persicae [38]. As they are minor components in fire ant venom, the availability of natural ∆ 1,6 -piperideines is very limited. The synthetic version of ∆ 1,6 -piperideines has the potential to control plant disease and pest insects. Future studies should focus on the scale-up synthesis of ∆ 1,6 -piperideines and the development of ∆ 1,6 -piperideine-based pesticides.

Synthesis
All reactions were carried out in dry glassware with a magnetic stirrer. Liquid solutions were transferred via syringe. All reactions were monitored by thin-layer chromatography on silica gel GF254 pre-coated plates (0.25 mm). The Grignard reagents n-C 11 H 23 MgBr, n-C 13 H 27 MgBr and n-C 15 H 31 MgBr were prepared according to a reported method from the corresponding n-alkyl bromide [43]. 1 H and 13 C NMR spectra were recorded on Varian Inova-400 MHz spectrometers using tetramethylsilane as the internal standard, and were reported in terms of chemical shift (δ ppm). The high-resolution mass spectra data were obtained using a Bruker mass spectrometer (Daltonik, Germany). Anhydrous THF was prepared by distillation over sodium with a diphenyl ketone indicator. Anhydrous DCM was collected by heating over CaH 2 . All other commercially obtained reagents and solvents were used as received.
Glutarimide (2): A mixture of glutaric acid (53.15 g, 0.40 mol) and urea (36.24 g, 0.60 mol) was melted for 40 min at 140 • C, and then the reaction mixture was stirred for an additional 2.5 h at 175 • C. Upon cooling, the crude product was recrystallized twice from EtOH to produce glutarimide 2 as a white solid (26.22 g, yield 58%) [44]. 1 (3): Into a stirred solution of glutarimide (15.00 g, 0.13 mol) in CH 2 Cl 2 (100 mL) at −78 • C under a nitrogen atmosphere, the Grignard reagent CH 3 MgBr was added into 2-methyltetrahydrofuran (3 M, 130 mL) dropwise. Then the reaction mixture was stirred at room temperature for 12 h. NaBH 3 CN (9.83 g, 0.16 mol) was added, followed by the slow addition of a 6 N HCl solution to keep the pH at 3-4. The reaction was continued at 25 • C for 5 h. After completion, the mixture was neutralized with 1 N NaOH and extracted with CH 2 Cl 2 (5 × 100 mL). The combined organic extracts were dried over anhydrous Na 2 SO 4 and concentrated in vacuo. The crude product was purified by column chromatography on silica gel using ethyl acetate:MeOH (20:1~10:1) to obtain compound 3 as a white solid (9.75 g, yield 65%) [45]. 1  6-Methyl-2-oxo-piperidine-1-carboxylic acid tert-butyl ester (4): To a solution of compound 3 (9.75 g, 0.086 mol), Et 3 N (13.07 g, 0.13 mol) and DMAP (0.52 g, 4.3 mmol) in CH 2 Cl 2 (150 mL), (Boc) 2 O (28.12 g, 0.13 mol) was added. The solution was stirred to reflux for 18 h. When TLC showed that the reaction was complete, the solvent was evaporated and the resulting oil was purified via column chromatography on silica gel using hexanes; ethyl acetate (10:1) was used to obtain compound 4 as a yellow oil (12.86 g, yield 70%) [46]. 1

General Procedure for Preparation of N-Boc-aminoketones (5a-c):
To a stirred solution of tetramethylethylenediamine (TMEDA) (1.5 equiv.) and N-Bocpiperidinone (4) (1 equiv.) in anhydrous THF at −30 • C under a nitrogen atmosphere, the freshly prepared Grignard reagent RMgBr (1.5 equiv.) was added dropwise. The resulting suspension was allowed to be stirred at 25 • C for 10 h. After completion, 1N HCl was added to quench the reaction. The mixture was extracted with CH 2 Cl 2 . The combined organic phase was dried over anhydrous Na 2 SO 4 . The solvent was evaporated and the crude products were isolated by column chromatography on silica gel using hexanes; ethyl acetate (15:1) was used to afford the corresponding compounds.

Extraction, Isolation and Identification of ∆ 1,6 -Piperideines
The extraction and isolation of venom alkaloids from S. invicta workers were conducted following previously reported procedures [15,16,48]. Briefly, the hexane extract of ant workers (1 g) randomly selected from the colony, was concentrated to 0.5 mL under a mild stream of nitrogen and loaded onto a 13 mm i.d. gravity chromatography column (20 g of silica gel, 300-400 mesh). The sample was sequentially eluted with hexane and a series of mixtures of hexane and acetone containing 2% triethylamine. Each 2 mL of eluent was collected, and its chemistry was analyzed by a Shimadzu GC-2030 equipped with a flame ionization detector (FID). The dimensions of the capillary column used were as follows: HP-5MS, 30 m × 0.25 mm i.d., 0.25 µm (Agilent). The injector was operated at 250 • C in splitless mode with the split opened after 1 min. Nitrogen was used as the carrier (1 mL/min) and makeup gas. The GC program used was as follows: injection at 90 • C (isothermal for 2 min), increase at 15 • C/min up to 270 • C and hold for 16 min. The detector temperature was set at 280 • C. In terms of appearance of characteristic peaks and their retention times in the GC chromatograms of each collection, the cisand trans-alkaloid fractions were obtained by pooling collections containing specific groups of alkaloids. A mixture of synthetic ∆ 1,6 -piperideines 6a, 6b and 6c in equal amounts was dissolved in hexane. To confirm the identities of ∆ 1,6 -piperideines, both the cis-alkaloid fraction with a final volume of 1.5 mL and the solution of synthetic standards were subjected to GC-MS analysis using a Shimadzu GCMS-TQ8050 NX equipped with a Shimadzu SH-Rxi-5Sil MS column with exactly the same dimensions as the HP-5MS column above. The oven temperature was programmed as follows: 90 • C for 1 min, increased at 10 • C/min to 160 • C, then to 250 • C at 3 • C/min, and held for 2 min. The transfer line temperature was set at 250 • C. Mass spectra were obtained using electron impact (EI, 70 eV). Helium was used as the carrier gas (1 mL/min).

Insect Rearing Procedure
Eggs of Helicoverpa armigera were purchased from a commercial supplier, Keyun Industry Co., Ltd. (Jiyuan, China). Newly hatched H. armigera larvae were reared in groups of 50-100 in 200 mL plastic cups at 27 ± 1 • C, relative humidity 60% and 16L:8D h photoperiod, until the third instar. Small cubes of artificial diet [49] were refreshed daily. The early third-instar larvae were individually transferred to 9 cm-diameter Petri dishes and used for bioassays. The in vivo effect of piperideines 6a-c was determined through topical and injection applications.

Topical Application
Piperideines 6a-c and nicotine were dissolved in acetone and then diluted to obtain the following concentrations: 0.4, 0.2, 0.1 and 0.05 mol/L. Acetone was used as a negative control, whereas nicotine was used as a positive control. A 2 µL drop of acetone solution was applied with a micropipette onto the thoracic dorsum of an anesthetized early thirdinstar larva placed in a metal bath at 4 • C. Twenty larvae were treated for each treatment. Treated larvae were placed individually in a 10 mL glass tube and maintained on an artificial diet during the post-treatment observation period under the standard rearing conditions describe above. Larvae were considered dead if they were unable to move in a coordinated manner when touched several times with a hairbrush. The numbers of dead larvae were recorded 24 h and 48 h after treatment for the calculation of mortality.

Injection Application
Injections of 6a-c and nicotine solutions at 0.4 mol/L were performed using the Nanoject II Injector (Drummond Scientific Company, Broomall, PA, USA). The needle of the injector was carefully inserted between the second and third abdominal segments of an anesthetized third-instar larva placed in the icy metal bath at a vertical angle of about 30 • under a dissecting microscope. A total of 202.4 nL (50.6 nL × 4) of the sample was applied to a single larva. Sixteen larvae were treated in each treatment. The tests were replicated five times over 5 consecutive days. Acetone was served as the control. The treated larvae were maintained separately under standard conditions. Mortality was recorded 24 h and 48 h after treatment.

Conclusions
In summary, three minor components of the fire ant venom and 2-methyl-6-alkyl-∆ 1,6piperideines were synthesized and used to confirm the identification of the corresponding natural products. Their retention times and mass spectra data coincided with those in the cis-alkaloid fraction purified from the hexane extract of S. invicta workers. Their insecticidal activities were further evaluated with topical application and injection methods. The results of the bioassays indicated that these compounds displayed insecticidal properties against the early third-instar larvae of H. armigera. In particular, compound 6a, with an undecyl side chain, displayed promising injection activity as compared to the natural alkaloid nicotine at 0.4 mol/L. The activity of 2-methyl-6-alkyl-∆ 1,6 -piperideines is negatively correlated to the length of the carbon chain at the 6-position of the piperideine ring. Interestingly, the piperideines were found to be almost inactive when applied topically. The cuticle of the cotton bollworm may prevent the penetration of piperideines into the haemolymph, resulting in such inactivity by topical application. The insecticidal activities of compounds 6a-c provide valuable information for further research on biological activities.