Larvicidal activity of lignans and alkaloid identified in Zanthoxylum piperitum bark toward insecticide-susceptible and wild Culex pipiens pallens and Aedes aegypti

Background The yellow fever mosquito, Aedes aegypti, and the common house mosquito, Culex pipiens pallens, transmit dengue fever and West Nile virus diseases, respectively. This study was conducted to determine the toxicity of the three lignans (–)-asarinin, sesamin and (+)-xanthoxylol-γ,γ-dimethylallylether (XDA), and the alkaloid pellitorine from Zanthoxylum piperitum (Rutaceae) bark to third-instar larvae from insecticide-susceptible C. pipiens pallens and Ae. aegypti as well as wild C. pipiens pallens resistant to deltamethrin, cyfluthrin, fenthion, and temephos. Methods The toxicities of all isolates were compared with those of mosquito larvicide temephos. LC50 values for each species and their treatments were significantly different from one another when their 95% confidence intervals did not overlap. Results XDA was isolated from Z. piperitum as a new larvicidal principle. XDA (LC50, 0.27 and 0.24 mg/l) was 4, 53, and 144 times and 4, 100, and 117 times more toxic than pellitorine, sesamin, and asarinin toward larvae from susceptible C. pipiens pallens and Ae. aegypti, respectively. Overall, all the isolates were less toxic than temephos (LC50, 0.006 and 0.009 mg/l). These constituents did not differ in toxicity to larvae from the two Culex strains. The present finding indicates that the lignans and alkaloid and the insecticides do not share a common mode of larvicidal action or elicit cross-resistance. Conclusion Naturally occurring Z. piperitum bark-derived compounds, particularly XDA, merit further study as potential mosquito larval control agents or as lead compounds for the control of insecticide-resistant mosquito populations. Electronic supplementary material The online version of this article (doi:10.1186/s13071-017-2154-0) contains supplementary material, which is available to authorized users.


Background
The yellow fever mosquito, Aedes aegypti (Linnaeus, 1762) [1], and the common house mosquito, Culex pipiens pallens (Coquillett, 1898) [2], are found in tropical and subtropical regions of the world [3] and Eastern Asia [4], respectively, and are serious disease vectoring insect pests [5,6]. A recent study calculated that more than 2.5 billion people are at risk of dengue infection over 100 countries worldwide, and there may be 50-100 million dengue infections annually, including 22,000 deaths every year, mostly among children [7]. From 1999 to 2015, 43,937 cases of human West Nile virus disease (including 20,265 neuroinvasive disease cases) were reported in the United States (US), which resulted in 1,911 deaths [8]. The most serious problem with the mosquito species is their ability to evolve resistance to insecticides rapidly [9]. Increasing levels of resistance to the conventional insecticides have resulted in multiple treatments and excessive doses, raising serious environmental and human health concerns. Widespread insecticide resistance has been one of the major obstacles in the cost-effective integrated vector management program. In addition, the number of approved insecticides may be reduced soon in the US by the US Environmental Protection Agency as reregistration occurs [10]. Reregistration requirement is also a concern in other regions including in the European Union, where it is under the control of the Commission Regulation (EC) No 1048/2005 [11]. Therefore, there is a high need for the development of selective control alternatives with novel target sites to establish a biorational resistance management strategy based on all available information on the extent and nature of resistance in mosquitoes because vaccines have limited effectiveness in controlling dengue [12].
In this study, our aim was to assess whether the three lignans, asarinin, xanthoxylol-γ,γ-dimethylallylether (XDA) and sesamin, and the isobutylamide alkaloid pellitorine, extracted from the bark of Z. piperitum, had the toxicity to third-instar larvae from insecticide-susceptible C. pipiens pallens and Ae. aegypti, as well as wild colonies of C. pipiens pallens resistant to various insecticides [23]. The toxicity of the bark constituents was compared with that of the currently available mosquito larvicide temephos to assess their use as future commercial mosquito larvicides because it is registered as a larvicide for the control of mosquitoes in South Korea [34]. Also, the quantitative structure-activity relationship (QSAR) of the test compounds is discussed.

Instrumental analysis
The 1 H and 13 C nuclear magnetic resonance (NMR) spectra were recorded in CDCl 3 on Varian NMR system spectrometers (Varian, Palo Alto, CA, USA), using tetramethylsilane as an internal standard. The chemical shifts are given in δ (ppm). The ultraviolet (UV) spectra were obtained in methanol on a UVICON 933/934 spectrophotometer (Kontron, Milan, Italy) and the mass spectra on a GSX 400 spectrometer (Jeol, Tokyo, Japan). Silica gel 60 (0.063-0.2 mm) (Merck, Darmstadt, Germany) and Sephadex LH-20 (Sigma-Aldrich, St. Louis, MO, USA) were used for column chromatography. Merck precoated silica gel plates (Kieselgel 60 F 254 ) were used for analytical thin-layer chromatography (TLC). An Agilent 1200 series high-performance liquid chromatograph (Agilent, Santa Clara, CA, USA) was used to isolate the active constituents.

Materials
The organophosphorus (OP) insecticide temephos (97.3%) was purchased from Riedel (Seelze, Lower Saxony, Germany). Triton X-100 was purchased from Coseal (Seoul, South Korea). All of the other chemicals used in this study were of reagent-grade quality and are available commercially.

Mosquitoes
The stock cultures of C. pipiens pallens (susceptible KS-CP strain) and Ae. aegypti have been maintained in the laboratory without exposure to any known insecticide, as described previously [35]. Larvae from YS-CP colony of C. pipiens pallens, originally collected near rice paddy fields and cowsheds in Yusung (Daejeon, South Korea) in September 2010, showed extremely high levels of resistance to fenthion (resistance ratio (RR), 390) and deltamethrin (RR, 164) and moderate levels of resistance to cyfluthrin (RR, 14) and temephos (RR, 14) [23]. Adult mosquitoes were maintained on a 10% sucrose solution and blood fed on live mice. Larvae were reared in plastic trays (24 × 35 × 5 cm) containing 0.5 g of sterilised diet (40-mesh chick chow powder/yeast, 4/1 by weight). All stages were held at 27 ± 1°C, 65-75% relative humidity, and a 14:10 h light:dark cycle.

Extraction and isolation
Air-dried bark (550 g) of Z. piperitum was pulverised, extracted with methanol (3.3 L) two times at room temperature for 2 days, and filtered. The combined filtrate was concentrated to dryness by rotary evaporation at 40°C to yield approximately 70 g of a dark brownish sticky solid. The extract (20 g) was sequentially partitioned into hexane-(6.4 g), chloroform-(1.36 g), ethyl acetate-(0.46 g), and water-soluble (11.78 g) portions for the subsequent bioassays. This fractionation procedure was repeated three times. The organic solvent-soluble portions were concentrated under vacuum at 35°C, and the water-soluble portion was freeze-dried. To isolate the active constituents, 10-50 mg/l of each Z. piperitum bark-derived fraction was tested in a mortality bioassay, as described by Perumalsamy et al. [22].
The hexane-soluble fraction (19.2 g) was the most biologically active fraction (Table 1) and was chromatographed on a 5.5 × 70 cm silica gel (500 g) column by elution with a gradient of chloroform and methanol [(100:0 (2 l), 95:5 (1 l), 90:10 (2 l), 80:20 (1 l), 50:50 (1 l), and 0:100 (1.5 l) by volume] to provide 34 fractions (each approximately 250 ml) (Fig. 1). The column fractions were monitored by TLC on silica gel plates developed with a chloroform and methanol (9:1 by volume) mobile phase. Column fractions with similar R f values on the TLC plates were pooled. The spots were detected by spraying the plate with 4% H 2 SO 4 and then heating on a hot plate. Active fractions 11-17 (H3) were pooled and rechromatographed on a 5.5 × 70 cm silica gel (500 g) column by elution with a gradient of hexane and ethylacetate [(90:10 (1 l), 80:20 (1 l), and 0:100 (1 l) by volume] and finally with 1 l methanol to afford 16 fractions (each approximately 250 ml). The fractions were monitored by TLC on silica gel plates developed with a hexane and ethyl acetate (7:3 by volume) mobile phase. Active fractions 8-13 (H33) were pooled and crystallised during being dried by rotary evaporation at 35°C to yield compound one (H331). The residual portion (H332) was isolated by Sephadex LH-20 column chromatography using a mobile phase of methanol. Two active fractions 4-11 (H3322) and 12-21 (H3323) were obtained. The H3322 fraction (4.09 g) was rechromatographed on a 5.5 × 70 cm silica gel (120 g) column. Separation was achieved with a gradient of hexane and acetone [80:20 (2 l), 70:30 (1 l), 50:50 (1 l), and 0:100 (0.5 l) by volume] and finally with 1 l methanol to afford 25 fractions (each approximately 200 ml). Column fractions were monitored by TLC on silica gel plates developed with a hexane and ethyl acetate (4:6 by volume) mobile phase. Active fractions 1-7 (H33221) were obtained. Fraction H33221 was rechromatographed on a silica gel column using a gradient of chloroform and ethyl acetate [20:1 (0.3 l), 10:1 (0.2 l), 8:2 (0.2 l), 1:1 (0.1 l), and 0:10 (0.5 l) by volume] and finally with acetone (0.2 l) to afford five fractions (each approximately 200 ml). A preparative high-performance liquid chromatography (HPLC) was performed to separate the constituents from the active H332212 fraction. The column was a 3.9 mm i.d. × 300 mm bondaclone ten silica (Phenomenex, Torrance, CA, USA) using a mobile phase of chloroform and ethyl acetate (95:5 by volume) at a flow rate of 1 ml/min. Chromatographic separation was monitored using a UV detector at 264 nm. The two active constituents two and three were isolated at retention times of 8.05 and 10.03 min, respectively. For separation of a constituent from another active H3323 fraction (1.3 g), a preparative HPLC was performed. The column was a 21.2 mm i.d. × 250 mm Phenomenex Prodigy ODS with a mobile phase of acetonitrile and water (1:1 by volume) at a flow rate of 1 ml/min. Chromatographic separation was monitored at 287 nm. Finally, an active constituent four was isolated at a retention time of 5.35 min.

Bioassay
A mortality bioassay [36] was used to assess the toxicity of all compounds to third-instar larvae from the susceptible and wild mosquitoes. In brief, each compound in acetone was suspended in distilled water with Triton X-100 (20 μl/l). Groups of 20 mosquito larvae were separately put into paper cups (270 ml) containing each compound solution (250 ml). Temephos served as a positive control and was similarly formulated. Negative controls consisted of the acetone-Triton X-100 solution in distilled water. Based on the preliminary test results, the toxicity of each test compound and insecticide was determined with four to six concentrations ranging from 0.1 to 100 mg/l and 0.001 to 0.1 mg/l, respectively. All treatments were replicated three times using 20 larvae per replicate. Treated and control (acetone-Triton X-100 solution only) larvae were held under the same conditions as those used for colony maintenance without providing food. Larval mortalities were determined 24 h posttreatment. A larva was considered dead if it did not move when prodded with a fine wooden dowel [22].

Data analysis
Data were corrected for control mortality using Abbott's formula [37]. Concentration-mortality data were subjected to probit analysis [38]. A compound having LC 50 > 100 mg/l was ineffective as described by Kiran et al. [39]. The LC 50 values for each species and their treatments were significantly different from one another when their 95% confidence intervals did not overlap.

Bioassay-guided fractionation and isolation
The fractions obtained from the solvent partitioning of the methanol extract of the Z. piperitum bark were bioassayed toward third-instar larvae from insecticidesusceptible C. pipiens pallens (Table 1) and Ae. aegypti (Table 2). Significant differences in toxicity were observed among the fractions and were used to identify the peak activity fractions for the next step of purification. Based on the 24 h LC 50 values, the hexane-soluble fraction was the most toxic material, followed by the Fig. 1 Procedures to isolate the mosquito larvicidal constituents. The Zanthoxylum piperitum bark methanol extract was sequentially partitioned into hexane-, chloroform-, ethyl acetate-, and water-soluble portions. The hexane-soluble fraction was the most biologically active fraction, and HPLC was performed. Each fraction (10-50 mg/l) was tested in a mortality bioassay to isolate the active constituents from the fraction chloroform-soluble fraction. No toxicity was obtained using the ethyl acetate-or water-soluble fractions. Mortality in the acetone-Triton X-100-water-treated controls for any the species in this study was less than 2%.

Larvicidal activity of test compounds
The toxicity of the four isolated constituents to thirdinstar larvae from KS-CP strain of C. pipiens pallens was likewise compared with that of temephos, which was used as a positive control (Table 3). Responses varied according to compound examined. Based on the 24 h LC 50 values, XDA (0.27 mg/l) was the most toxic compound, followed by pellitorine (1.12 mg/l). These constituents were 45 and 187 times less toxic than temephos, respectively. LC 50 of sesamin was 14.28 mg/l. The toxicity of asarinin was the lowest of any of the compounds examined. Interestingly, the toxicity of all constituents was virtually identical toward both insecticide-susceptible and wild third-instar C. pipiens pallens larvae (Table 4), indicating a lack of crossresistance in the resistant larvae. Toward third-instar Ae. aegypti larvae (Table 5), XDA (LC 50 , 0.24 mg/l) was the most toxic compound, followed by pellitorine (LC 50 , 0.98 mg/l), as judged by the 24 h LC 50 values. These constituents were 27 and 109 times less toxic than temephos, respectively. LC 50 of sesamin and asarinin was 23.98 and 28.15 mg/l, respectively.
In the current study, we used a mortality bioassay to identify the larvicidal constituents from the Z. piperitum bark extracts. The active constituents were determined to be the furofuranoid lignans (-)-asarinin (1), (+)-XDA (2) and sesamin (4), and the isobutylamide alkaloid pellitorine (3). The interpretations of the proton and carbon signals of compounds 1, 2, 3, and 4 were largely consistent with those of Perumalsamy et al. [22], Biavatti et al. [52], Perumalsamy et al. [22] and Park et al. [44], and Ju et al. [53], respectively. XDA was isolated from Z. piperitum as a new larvicidal constituent. This compound was most toxic toward larvae of two vector mosquito species, although it was less toxic than temephos. Pellitorine was also highly toxic toward C. pipiens pallens and Ae. aegypti, as described previously [22,45]. In addition, these constituents were also effective toward C. pipiens pallens larvae resistant to various insecticides. The present finding indicates that Z. piperitum bark-derived preparations containing the active constituents, particularly XDA and pellitorine, hold promise for the development of novel, effective, naturally occurring mosquito larvicides even toward currently insecticide-resistant mosquito populations, because XDA (LC 50 0.24-0.27 mg/l for two mosquito species) and pellitorine (LC 50 0.98-1.12 mg/l for two mosquito species) meet the stage 3 criteria (LC 50 < 1 mg/l) set by Shaalan et al. [14]. The next step stage 4 involves the determination of effective field application rates of various formulations in simulated field trials and/or small-scale field trials [14].  QSARs of phytochemicals in many insects have been well noted. For example, Wang et al. [23] studied the toxicity of six linear furanocoumarins including imperatorin and six simple coumarins including osthole. They reported that the chemical structure and alkoxy substitution and length of the alkoxyl side chain at the C8 position are essential for imparting toxicity. Park et al. [44] reported that the larvicidal activity toward three vector mosquito species was much more pronounced in compounds such as guineensine, pipercide, and retrofractamide A with an isobutylamine moiety than in one such as piperine without this moiety among the methylenedioxyphenyl (MDP)-containing compounds. In addition, the isobutylamides with an MDP moiety was more active than the ones without an MDP moiety. The MDP moiety is thought to stabilise the chemical structure [54]. In the current study, XDA with an MDP moiety was more toxic than either asarinin or sesamin with two MDP moieties. In addition, sesamin was more toxic than asarinin, 7-epimer of sesamin. Our findings, along with previous studies, indicate that other factor(s) such as chemical structure, functional group, and isomerism, as well as hydrophobic (log P) and molecular refraction parameters, may play, in part, a role in determining the lignan toxicities to mosquito larvae, although the MDP moiety might contribute, to some extent, to the larvicidal effect.
An investigation of the modes of action and the resistance mechanisms of biolarvicides may contribute to the development of selective mosquito control alternatives with novel target sites. Major mechanisms of resistance to insecticides currently available to control mosquitoes are target site insensitivity that reduces sodium channel sensitivity to pyrethroid insecticides or sensitivity of acetylcholinesterase to OP and carbamate insecticides, as well as enhanced metabolism of various groups of insecticides [55,56]. Some phytochemicals were found to be highly effective toward insecticide-resistant mosquitoes [14,22,23], and they are likely to be useful in resistance management strategies. For example, imperatorin and osthole are effective toward larvae from wild C. pipiens pallens with extremely high to moderate levels of resistance to cyfluthrin, deltamethrin, fenthion, and temephos [22]. The current findings that the three furofuranoid lignans and the isobutylamide alkaloid described were of equal toxicity to both insecticide-susceptible and -resistant larvae of C. pipiens pallens suggest that the phytochemicals and the pyrethroid and OP insecticides do not share a common mode of action or elicit cross-resistance. Detailed tests are needed to understand fully the exact mode of action of the furofuranoid lignans and the isobutylamide alkaloid, although the octopaminergic and γ-aminobutyric acid receptors have been suggested as novel target sites for some monoterpenoid essential oil constituents in the American cockroach [57] and the cotton bollworm [20] and the fruit fly [21], respectively. It has also been reported that tannins and pellitorine primarily affect the midgut epithelium and secondarily affect the gastric caeca and the malpigian tubules in C. pipiens larvae [58] and Ae. aegypti larvae [59], respectively.