Purification of the ethanol extract of O. japonica root barks afforded a new pyrrolidine alkaloid (17.9 mg) (Figure 1), the structure of which was elucidated by NMR and MS analysis. NMR and MS spectra are available in supplementary materials.

(Z)-3-(4-Hydroxybenzylidene)-4-(4-hydroxyphenyl)-1-methylpyrrolidin-2-one was obtained as white crystal, and displayed a molecular formula of C₁₈H₁₇NO₃ ([M − H]⁻, calcd., m/z 294.1130; found, m/z 294.1135) as determined by high-resolution electrospray ionization mass spectrometry (HRESIMS). The ¹H- and ¹³C-NMR spectra showed 15 proton and 14 carbon signals (Table 1). According to the HSQC spectrum and a previous report, proton signals at δ_H 6.66 (2H, d, J = 8.1 Hz), 7.23 (2H, d, J = 8.2 Hz), together with their corresponding carbon signals at δ_C 114.9 and 131.8, respectively, as well as signals at δ_H 6.72 (2H, d, J = 8.0 Hz) and 7.05 (2H, d, J = 8.0 Hz) with their corresponding carbon signals at δ_C 115.3 and 127.7, respectively, suggested the presence of two symmetrical para-substituted aromatic protons [20]. Besides, signals at δ_H 4.48 (1H, d, J = 7.8 Hz), 3.27 (1H, d, J = 9.9 Hz), and 3.98 (1H, d, J = 9.3 Hz), together with their corresponding carbon signals at δ_C 40.33 and 57.0, respectively, and a carbonyl carbon signal at δ_C 170.8 revealed a pyrrolidine-2-one ring [21]. Furthermore, the existence of a methyl group at δ_H 2.98 (3H, s) was noticeable in the ¹H-NMR spectrum.

In the HMBC spectrum, correlations from H-6 (δ_H 7.39) to C-8 (δ_C 131.8) and C-12 (δ_C 131.8), from H-8 (δ_H 7.23) to C-6 (δ_C 132.2) and C-10 (δ_C 158.2), from H-9 (δ_H 6.66) to C-7 (δ_C 125.9) and C-10 (δ_C 158.2), from H-11 (δ_H 6.66) to C-7 (δ_C 125.9) and C-10 (δ_C 158.2), and from H-12 (δ_H 7.23) to C-6 (δ_C 132.2) and C-10 (δ_C 158.2) revealed fragment A (Figure 2). Besides, HMBC correlations from H-14 (δ_H 7.05) to C-4 (δ_C 40.3) and C-16 (δ_C 156.1), from H-15 (δ_H 6.72) to C-13 (δ_C 133.5) and C-16 (δ_C 156.1), from H-17 (δ_H 6.72) to C-13 (δ_C 133.5) and C-16 (δ_C 156.1), and from H-18 (δ_H 7.05) to C-14 (δ_C 127.7) and C-16 (δ_C 156.1) confirmed fragment B (Figure 2). Furthermore, correlations from H-5 (δ_H 3.27) to C-2 (δ_C 170.8) and from H-19 (δ_H 2.98) to C-2 (δ_C 170.8) and C-5 (δ_C 57.0) proved the existence of fragment C (Figure 2). According to HMBC correlations from H-6 (δ_H 7.39) to C-2 (δ_C 170.8) and C-4 (δ_C 40.3), fragment A was attached to C-3 of fragment C (Figure 2). HMBC correlations from H-4 (δ_H 4.48) to C-14 (δ_C 127.7) and C-18 (δ_C 127.7) revealed fragment B connected with C-4 of fragment C. Thus, the compound was determined as (Z)-3-(4-hydroxybenzylidene)-4-(4-hydroxyphenyl)-1-methylpyrrolidin-2-one.

The isolated pyrrolidine alkaloid was tested for its larvicidal activity against the fourth instar larvae of Ae. aegypti, An. sinensis, and C. pipiens pallens. The fourth instar larvae of An. sinensis was the most susceptible with a 24 h LC₅₀ value of 49.91 μg/mL (Table 2), while C. pipiens pallens and Ae. aegypti were 3.2 and 4.7 times less sensitive than An. sinensis, respectively. However, such differences in sensitivity to chemicals need to be further investigated. When compared with a positive control, commercial rotenone, the isolated compound was 62–85 times less toxic to the three species of mosquitoes.

The isolated alkaloid also possessed nematicidal activity against juveniles of Bursaphelenchus xylophilus and Meloidogynein congnita with 72 h LC₅₀ values of 391.50 μg/mL and 134.51 μg/mL (Table 3), respectively; M. incongnita was 2.9 times more susceptible than B. xylophilus. However, this susceptibility difference is not yet clear. In comparison to a positive control, commercial avermectin, the isolated compound was 5000 times less toxic to the two species of nematodes.

On the basis of a literature survey, compounds of the pyrrolidine type have exhibited various biological activities, such as antioxidative, antifungal, and antibacterial activity; antitumor properties; cytotoxicity; and glycosidase inhibitory; and platelet aggregation inhibitory effects [22,23,24]. It is worth mentioning that pyrrolidines have been found to possess larvicidal activity against Ae. aegypti at a concentration less than 140 ppm [25]. Pyrrolidines also possess considerable insecticidal activity when evaluated against termite workers [26]. Moreover, pyrrolidine alkaloids have also been found to possess DNA-binding affinity and cytotoxicity [27], which might be the mechanism of action of the new alkaloid against larval mosquitoes and plant-parasitic nematodes. The above findings suggested the new alkaloid with a pyrrolidine ring isolated from the ethanol extract of O. japonica has the potential to be developed into an alternative larvicide and nematicide.

NMR spectra were recorded on Bruker (Billerica, MA, USA) Avance 500 instruments using MeOD as solvents. HRESIMS analysis was performed on an APEX II spectrometer (Bruker Daltonic Inc., Billerica, MA, USA).

Eggs of Ae. aegypti and An. sinensis and egg masses of C. pipiens pallens were obtained from the Department of Vector Biology and Control, National Institute for Communicable Disease Control and Prevention, Chinese Center for Disease Control and Prevention. Adult mosquitoes were reared in a cage (60 cm × 30 cm × 30 cm) which was placed in a growth chamber (14:10 h light:dark (L:D), 26–28 °C, 70%–80% RH). Adults were maintained on a 10% glucose solution and were allowed to blood-feed on live mice. Adults of both Ae. aegypti and An. sinensis deposited on strips of moistened filter paper, which were kept in glass beakers. Paper strips with Ae. aegypti eggs were kept wet for 24 h and then dehydrated, while An. sinensis eggs need to be kept wet. However, egg masses of C. pipiens pallens were oviposited in distilled water directly. Then, the egg masses were transported to a clean, white porcelain basin containing distilled water. Eggs or egg masses were kept in distilled water under the same conditions described above. Mosquito larvae were provided with a mixture of pig liver, fish food, and yeast powder (1:1:1) and were not utilized for experiments until they reached the fourth instar.

Colonies of B. xylophilus were maintained on Botrytis cinerea cultures. The fungus B. cinerea was cultured on potato dextrose agar (PDA) in a growth chamber (25–28 °C in dark). When B. cinerea was fully grown on PDA, B. xylophilus were collected through the modified Baermann funnel technique, washed with a mixture of 0.1% streptomycin sulfate and 0.002% actinone three times to remove any surface bacterial or fungal contaminants and then inoculated on the plate [28]. The plate was maintained in the growth chamber (25–28 °C in dark) until the fungal mycelium were completely consumed by B. xylophilus. Then, B. xylophilus were collected, washed thoroughly with sterilized distilled water, and used for bioassays immediately. All the bioassays were performed under laboratory conditions at 25–28 °C.

Eggs of M. incongnita were extracted from infected roots of tomato (Lycopersicone sculentum Mill.). All the tomato plants were reared in a growth chamber (16:8 h L:D, 25–28 °C, 75%–80% RH). When they reached a five-leaf stage, the tomatoes were used for inoculations. After 42 days, infected tomatoes were uprooted and the roots were washed free of soil with distilled water. Egg masses were hand-picked using sterilized tweezers from infected roots and then placed on a mesh nylon filter (openings 30 μm in diameter) [29]. J2s that passed through the filter were collected daily and used for bioassays immediately.

Fresh root barks of O. japonica were collected in Xixiu District, Anshun City, Guizhou Province, China (39.90°N and 116.41°E) in August 2014, and identified by Dr. Liu QR, College of Life Sciences, Beijing Normal University, Beijing. A voucher specimen of O. japonica (Rutaceae-Orixa japonica-Guizhou-2014-08) was deposited at the museum of the Department of Entomology, China Agricultural University, Beijing, China.

Root barks of O. japonica (20 kg) were air-dried, cut into pieces, and then successively extracted with 40L of different concentrations of ethanol/distilled water mixtures (95%, 75%, 50%, by volume) and distilled water at room temperature for 3 days. The extracts were then filtered, mixed and removed from the solvent under reduced pressure to afford crude extract (200 g). After that, the crude extract was suspended in distilled water (2 L) and then successively partitioned with the same volume of n-hexane, chloroform, ethyl acetate, and n-butyl alcohol to obtain four solvent fractions (37 g, 69 g, 25 g, 33 g). The n-butyl alcohol fraction was subjected to chromatography on a macroporous resin (AB-8, 1000 g) column (85 mm i.d., 850 mm length), eluting with distilled water containing increasing amounts of ethanol (up to 90%, by volume) to yield eight fractions. Fraction 8 (5.3 g) was then subjected to silica gel (Merck 9385) column (50 mm i.d., 500 mm length), eluting with CH₂Cl₂–EtOH (100:0–100:50) to afford 25 fractions. Fraction 10 (114 mg) was separated by a silica gel (Merck 9385) column (25 mm i.d., 500 mm length), eluting with CH₂Cl₂–EtOH (100:5–100:20) to afford 13 fractions. Fraction 6 (9.2 mg) was further purified by Sephadex LH-20 column (15 mm i.d., 500 mm length) chromatography, eluting with ethanol alone to yield the new pyrrolidine alkaloid (17.9 mg).

All bioactivity assays were performedunder laboratory conditions at 25–28 °C. The compound was dissolved in ethanol and diluted with tap water. The larvicidal bioassays were as recommended by WHO [30]. Solutions (250 mL) of the tested material at various concentrations were placed in glass beakers, and then 10 larvae were delivered toeach beaker. Each test was composed of six concentrations with five replicates. Commercial rotenone (purchased from Aladdin-Reagent Company, Shanghai, China) served as a positive control and ethanol was used as a negative control. Both treated and control larvae were placed in the growth chamber where mosquitoes were reared. Mortality recordings were taken after 24 h of exposure. Larvae that showed no movements were considered to be dead.

The standard nematode suspensions were prepared by appropriate dilution with sterilized distilled water to get approximately 100 juveniles/mL. Then, each well of 24-well tissue culture plates were added with 500 μL standard juvenile suspension. Numbers of active juveniles in every well were counted under a stereoscope at 10× and 5× before 500 μL stock solution was added to the corresponding well. However, the final concentration of ethanol in each treatment never exceeded 1% (by volume) [31]. Plates were then covered with “Xuan paper” (a high-quality rice paper made for traditional Chinese painting and calligraphy) to avoid evaporation. Each test was composed of five concentrations with three replicates. Commercial avermectin (purchased from Aladdin-Reagent Company, Shanghai, China) was used as a positive control, and distilled water containing ethanol (1%, by volume) served as a negative control. Both treated and control juveniles were placed in a growth chamber at 25–28 °C in dark. Mortality recordings were taken 72 h after treatment. Juveniles that showed no movements when stimulated with a fine needle were considered to be dead.