Bioassay-Guided Isolation of Iridoid Glucosides from Stenaria nigricans, Their Biting Deterrence against Aedes aegypti (Diptera: Culicidae), and Repellency Assessment against Imported Fire Ants (Hymenoptera: Formicidae)

In our natural product screening program, we screened natural products for their repellency and toxicity against insect vectors. Methanolic extract of aerial parts of Stenaria nigricans (Lam.), with no published chemistry, was tested for repellency against mosquitoes and imported hybrid fire ants. Methanolic extracts showed biting deterrence similar to DEET (N,N-diethyl-3-methylbenzamide) against Aedes aegypti L. Based on this activity, the crude extract was fractionated into chloroform, ethyl acetate, and methanol subfractions. The active methanolic subfraction was further fractionated into 13 subfractions. These fractions were tested for their biting deterrence against Ae. Aegypti. Active subfractions were further characterized to identify the compounds responsible for this activity. Four undescribed iridoid glucosides (1–4) and three previously reported compounds (5–7) were isolated from active subfractions and tested for their biting deterrent activity. Based on BDI values, compounds 2, 3, 6, and 7, with biting deterrence similar to DEET, showed the potential to be used as repellents against mosquitoes. In an in vitro digging bioassay, none of these compounds showed any repellency against hybrid imported fire ants at a dose of 125 µg/g. This is the first report of biting deterrence and repellency of S. nigricans extract and its pure compounds, iridoid glucosides against mosquitoes and imported fire ants. Further studies will be conducted to explore the repellent potential of these compounds in different formulations under field conditions.


Introduction
Mosquitoes are important in global public health because of their ability to transmit diseases. Mosquitoes are vectors that can cause human diseases such as malaria, dengue fever, yellow fever, and Chikungunya. In cases of high levels of transmission, epidemics can result in substantial human morbidity and mortality. In addition to other viruses, Aedes aegypti (L) and Ae. albopictus (Skuse) are vectors of the dengue and Zika viruses [1]. Dengue is one of the major vector-borne diseases that causes severe morbidity and mortality, affecting around 50-100 million people yearly [2]. Malaria, which presents a serious threat to global health, is a common disease vectored by the Anopheles spp. of mosquitoes [3], whereas Culex quinquefasciatus Say transmits the West Nile virus [4]. Imported fire ants (Hymenoptera: Formicidae) have a wide distribution in the world, and are pests of significant agricultural and medical importance. Solenopsis invicta Buren and S. richteri Forel are two fire ant species present in the United States. Extensive hybridization is reported to occur along the population boundaries between S. invicta and S. richteri in the Southern

Results and Discussion
Methanolic crude extract of S. nigricans aerial parts at 10 µg/cm 2 showed biting deterrence similar to DEET at 25 nmol/cm 2 in K & D bioassay (Figure 2a). To follow the activity, the methanolic crude extract was fractionated by vacuum liquid chromatography (VLC) over silica gel with chloroform, ethyl acetate, and methanol to yield three respective subfractions (A-C), which were tested for their biting deterrent activity (Figure 2b). Fraction C, which showed biting deterrence similar to DEET (Figure 2b), was further fractionated by silica gel column chromatography (CC) into 13 fractions (C1-C13). Fraction C4 turned out to be pure compound 1. Except for Fr. C5 (insufficient amount), the other 12 fractions (C1-C4 and C6-C13) were tested for their biting deterrence against Ae. aegypti females ( Figure 3). Frs. C2, C3, C6, C7, C8, and C13 at 10 µg/cm 2 showed biting deterrence activity similar to DEET ( Figure 3). Subsequently, the bioactive fractions were subjected to chromatographic techniques to find the active principles that resulted in the purification of compounds 2-7. The purified compounds were characterized by spectroscopic and spectrometric techniques (1D-and 2D-NMR and HRESIMS). Compounds 1-3 and 5-7 were tested for their biting deterrent activity at 25 nmol/cm 2 . Based on BDI values, compounds 2, 3, 6, and 7 showed biting deterrence similar to DEET ( Figure 4) and have the potential to be used as repellents against mosquitoes. In an in vitro digging bioassay, none of these compounds showed any repellency against hybrid imported fire ants at a dose of 125 µg/g. This is the first report of biting deterrence and repellency of S. nigricans extract and its pure compounds, iridoid glucosides, against mosquitoes and imported fire ants. Further studies should be conducted to explore the repellent potential of these compounds in different formulations under field conditions.

Results and Discussion
Methanolic crude extract of S. nigricans aerial parts at 10 µg/cm 2 showed biting deterrence similar to DEET at 25 nmol/cm 2 in K & D bioassay (Figure 2a). To follow the activity, the methanolic crude extract was fractionated by vacuum liquid chromatography (VLC) over silica gel with chloroform, ethyl acetate, and methanol to yield three respective subfractions (A-C), which were tested for their biting deterrent activity ( Figure 2b). Fraction C, which showed biting deterrence similar to DEET (Figure 2b), was further fractionated by silica gel column chromatography (CC) into 13 fractions (C1-C13). Fraction C4 turned out to be pure compound 1. Except for Fr. C5 (insufficient amount), the other 12 fractions (C1-C4 and C6-C13) were tested for their biting deterrence against Ae. aegypti females ( Figure 3). Frs. C2, C3, C6, C7, C8, and C13 at 10 µg/cm 2 showed biting deterrence activity similar to DEET ( Figure 3). Subsequently, the bioactive fractions were subjected to chromatographic techniques to find the active principles that resulted in the purification of compounds 2-7. The purified compounds were characterized by spectroscopic and spectrometric techniques (1D-and 2D-NMR and HRESIMS). Compounds 1-3 and 5-7 were tested for their biting deterrent activity at 25 nmol/cm 2 . Based on BDI values, compounds 2, 3, 6, and 7 showed biting deterrence similar to DEET ( Figure 4) and have the potential to be used as repellents against mosquitoes. In an in vitro digging bioassay, none of these compounds showed any repellency against hybrid imported fire ants at a dose of 125 µg/g. This is the first report of biting deterrence and repellency of S. nigricans extract and its pure compounds, iridoid glucosides, against mosquitoes and imported fire ants. Further studies should be conducted to explore the repellent potential of these compounds in different formulations under field conditions.      Stenigroside A (1) was obtained as an amorphous powder and its molecular formula, C21H28O11, was deduced from the [M+HCOOH-H] − ion peak at m/z 501.1626 (calculated for C22H29O13, 501.1608) in the HRESIMS, indicating an equivalence of eight double bonds. The 13 C-NMR spectrum of 1 displayed 21 resonances, of which ten were assignable to the iridoid skeleton, five to a 2-methylbutanoyl group, and six to a sugar unit ( Table 1) (Tables 1 and 2). The chemical shifts and coupling constant values related to the sugar moiety were typical of the β-glucopyranose unit [26] (Tables 1 and 2). The γ-lactone was supported by the HMBC correlation observed between H-6 (δH 5.46) and C-11 (δC 170.3). The placement of the sugar unit at C-1 and methyl butanoate moiety at C-10 was confirmed by HMBC correlations of the anomeric proton (δH 5.34) with C-1 (δC 93.1) and oxymethylene protons Ha/Hb-10 (δH 4.66 and 4.77) with carbonyl (δC 176.1), respectively. Based on these results and a detailed study of the NMR data, the structure of compound 1 was determined to be similar to asperuloside (5) [27], except the resonances of the acetyl group were absent, instead showing resonances for a 2-methylbutanoyl group in 1. The complete assignment of 1 H-and 13 C-NMR spectroscopic data (Tables 1 and 2) was done based on 1 H-1 H COSY couplings and HMBC correlations ( Figure 5). Based on NOESY correlations (Figure 6), the relative configuration of 1 was found to be the same as that of asperuloside (5) [28], i.e., cis arrangement of ring  (Tables 1 and 2). The chemical shifts and coupling constant values related to the sugar moiety were typical of the β-glucopyranose unit [26] (Tables 1 and 2). The γ-lactone was supported by the HMBC correlation observed between H-6 (δ H 5.46) and C-11 (δ C 170.3). The placement of the sugar unit at C-1 and methyl butanoate moiety at C-10 was confirmed by HMBC correlations of the anomeric proton (δ H 5.34) with C-1 (δ C 93.1) and oxymethylene protons Ha/Hb-10 (δ H 4.66 and 4.77) with carbonyl (δ C 176.1), respectively. Based on these results and a detailed study of the NMR data, the structure of compound 1 was determined to be similar to asperuloside (5) [27], except the resonances of the acetyl group were absent, instead showing resonances for a 2-methylbutanoyl group in 1. The complete assignment of 1 H-and 13 C-NMR spectroscopic data (Tables 1 and 2) was done based on 1 H-1 H COSY couplings and HMBC correlations ( Figure 5). Based on NOESY correlations (Figure 6), the relative configuration of 1 was found to be the same as that of asperuloside (5) [28], i.e., cis arrangement of ring junction protons (H-5, H-6, and H-9) and β axial orientation of glycosidic linkage. Ultimately, the structure of stenigroside A (1) was elucidated, as shown in Figure 1.  junction protons (H-5, H-6, and H-9) and β axial orientation of glycosidic linkage. Ultimately, the structure of stenigroside A (1) was elucidated, as shown in Figure 1.  junction protons (H-5, H-6, and H-9) and β axial orientation of glycosidic linkage. Ultimately, the structure of stenigroside A (1) was elucidated, as shown in Figure 1.    Figure 6, which consecutively supported the α-orientation of the methoxy group due to the NOESY correlation observed between H-4 and H-3. Thus, the structure of stenigroside C (3) was determined, as shown in Figure 1.
The molecular formula of stenigroside D (4), C 16 (Table 1) was found to be comparable with that of the scandoside [29], which contained a -COOH group, except for a significant deshielding of C-4 (+5.4 ppm) and C-11 (+3.8 ppm) and shielding of C-3 (−4.1 ppm) in 4 that verified the anionization of the C-4 carboxyl group. The anionization effect has been reported to cause a deshielding impact on 13 C-NMR resonances close to the negative charge [30,31]. The stereochemistry of 4, assigned by the NOESY experiment ( Figure 6), was the same as that of scandoside. Thus, the structure of stenigroside D (4) was elucidated, as shown in Figure 1.

General Procedures
IR spectra (frequency range 4000-500 cm −1 ) were recorded on an Agilent Technologies Cary 630 FTIR. Optical rotations were carried out at 28 • C chamber temperature on an AU-TOPOL II automatic polarimeter (Rudolph Research Analytical, Hackettstown, NJ, USA). UV spectroscopy was performed on a Thermo Scientific Evolution 201 UV-Vis spectrophotometer at ambient temperature. Mass data were determined on an Agilent Technologies 6230 ToF mass spectrometer. NMR spectra were recorded at 25 • C on a Bruker AU III 500 MHz NMR spectrometer using CD 3 OD or C 5 D 5 N. Chemical shifts were referenced to the residual solvent signals of methanol and pyridine. Column chromatography was performed using flash silica gel (40-63 µm, 60 Å, SiliCycle Inc., Quebec, QC, Canada) and Sephadex LH-20 (GE Healthcare, Chicago, IL, USA), with analytical grade solvents from Fisher Scientific. Thin layer chromatography (TLC) was performed on a silica gel F 254 aluminum sheet (20 cm × 20 cm, 200 µm, 60 Å, Sorbtech, Norcross, GA, USA) or a RP-18 silica F 254 aluminum sheet (20 cm × 20 cm, 150 µm, 60 Å, Sorbtech, Norcross, GA, USA). The spots on the silica card were visualized by spraying with 0.5% vanillin (Sigma, St. Louis, MO, USA) solution in concentrated H 2 SO 4 -EtOH (5:95), followed by heating (≈130 • C). DEET was purchased from Sigma-Aldrich (St Louis, MO, USA).
Adults of Ae. aegypti used in these studies were obtained from the laboratory colonies maintained at the Mosquito and Fly Research Unit at the Center for Medical, Agricultural and Veterinary Entomology, USDA-ARS, Gainesville, Florida. For biting deterrence and repellent bioassays, eggs were hatched and the larvae were reared to adults in the laboratory. They were maintained at 27 ± 2 • C and 60 ± 10% RH, with a photoperiod regimen of 12:12 h (L:D). Eight-to eighteen-day-old adult females were used in these bioassays. Hybrid fire ant workers used in these studies were obtained from the mounds located under natural field conditions at the University Field Station, University of Mississippi, 15 County Road 2078, Abbeville, Mississippi 38601.

Plant Material
The aerial parts of Stenaria nigricans (Specimen # MOBOT 270) were obtained from the Missouri Botanical Garden, Missouri, USA. A sample specimen (# 6824) was deposited in the repository at the National Center for Natural Products Research, University of Mississippi.

Extraction and Isolation
The plant powder (69 g) was extracted with methanol (3 L × 20 h) at room temperature, including sonication for 60 min. The crude extract (12.5 g) was obtained after removal of the solvent under reduced pressure at 45 • C, followed by lyophilization. The extract (11.9 g) was fractionated by vacuum liquid chromatography (VLC) over silica gel (30 cm

In Vitro K & D Biting Deterrent Bioassay
Bioassays were conducted using a six-celled in vitro Klun and Debboun (K&D) module bioassay system for quantitative evaluation of biting deterrence [35]. Briefly, the assay system consists of a six-well reservoir with each of the 3 cm × 4 cm wells containing 6 mL of feeding solution. The reservoirs were covered with a layer of collagen membrane (Devro, Sandy Run, Swansea, SC, USA). The test compounds were applied to six 4 cm × 5 cm marked areas of organdy cloth (G Street Fabrics, Rockville, MD, USA) and positioned over the collagen-covered CPDA-1+ATP solution [15]. A six-celled K&D module containing five female mosquitoes per cell was positioned over treated organdy, covering the six CPDA-1+ATP solution membrane wells, and trap doors were opened to expose the treatments to these females. The number of mosquitoes biting through cloth treatments in each cell was recorded after a 3 min exposure. The crude preparations were evaluated at dosages of 10 µg/cm 2 , and pure compounds including DEET were tested at a concentration of 25 nmol/cm 2 . Sets of 5 replications, each with 5 females per treatment, were conducted on 2-3 different days using a newly treated organdy and a new batch of females in each replication. Treatments were repeated 10 times.
Statistical Analyses. Proportion not biting (PNB) was calculated using the following formula: Total number of females biting Total number of females The K&D module bioassay system could only handle four treatments, along with negative and positive controls, in order to make direct comparisons among test samples and compensate for variation in overall response among replicates; biting deterrent activity was quantified as the biting deterrence index (BDI) [15]. The BDI was calculated using the following formula: where PNB i,j,k denotes the proportion of females not biting when exposed to test compound i for replication j and day k (I = 1-4, j = 1-5, k = 1-2), PNB c,j,k denotes the proportion of females not biting the solvent control "c" for replication j and day k (j = 1-5, k = 1-2), and PNB d,j,k denotes the proportion of females not biting in response to DEET "d"(positive control) for replication j and day k (j = 1-5, k = 1-2). This formula adjusts for inter-day variation in response, and incorporates information from the solvent control as well as the positive control. BDI values not significantly different from 1 are similar to DEET. Data were analyzed using SAS Proc ANOVA [single factor: test compound (fixed)] [36].
To determine whether confidence intervals included the values of 0 or 1 for treatments, Scheffe's multiple comparison procedure with the option of CLM was used in SAS.

In Vitro Digging Bioassay for Fire Ants
Ants were identified by analysis of venom alkaloid and cuticular hydrocarbon profiles, as described by Chen et al [37]. A digging bioassay was used to determine the repellency of the pure compounds against imported fire ants. The digging bioassay used in this study was described by Ali et al [38]. The sand of a uniform size of 1 mm was washed in de-ionized water and dried at 150 • C. An amount of 4 g of sand was treated in a 45 mL aluminum weighing dish (Fisher Scientific, 300 Industry Drive, Pittsburgh, PA, USA) at a volume of 400 µL. After the evaporation of ethanol, de-ionized water was added at a rate of 0.6 µL/g of sand to moisten the sand. The vials were filled with treated sand, whereas the sand in the control treatment was only treated with ethanol. The vials were then screwed to the caps attached to the bottom of the arena. Fifty hybrid fire ant workers were released in the center of the arena petri dish. The experiment was conducted at 25 ± 2 • C temperature and 50 ± 10% relative humidity. The pure compounds were screened at our standard screening dose of 125 µg/g. After 24 h, sand was collected back into aluminum dishes, dried at 150 • C for 1 h, and weighed.

Conclusions
Methanolic extract of aerial parts of Stenaria nigricans and its pure compounds showed biting deterrence against Aedes aegypti. Seven compounds, including four undescribed iridoid glucosides (1-4), were isolated through mosquito-biting deterrent bioassay-guided isolation. High biting deterrence of stenigroside B, stenigroside C, deacetylasperuloside, and daphylloside indicated a high potential of these natural compounds to be used as mosquito repellents. Further research, through intensive laboratory and field trials in in vivo bioassays and field trials, is needed to explore the potential of these natural products against mosquitoes. However, none of these compounds showed digging suppression against fire ants.