3.1 Isolation and purification of N-alkylamides from A. ciliata
Methanol extraction of A. ciliata flowers yielded ca. 30 g (15% by dry weight) of a brownish-yellow crude extract. HPLC of the crude showed the presence of 10 distinct peaks indicating as many compounds in the extract (Fig 1A). Retention time (RT) of the peaks ranged from 2.36 min to 17.24 min, of which the peak with an RT of 8.8 min was the most prominent one (Fig 1A). TLC-assisted purification of the crude by column chromatography yielded two compounds having RT of 5.1 min and RT 8.8 min, with utmost purity ~97% and a yield of 200 mg and 275 mg, respectively (Fig 1B and 1C). They were annotated as compound 1 (C1) and compound 2 (C2) respectively.
3.2 Structure determination of the isolated compounds
The results of FTIR, HRMS, 1H-NMR, and 13C-NMR of the purified compounds C1 and C2 are presented below.
Compound 1 (C1): Physically it was dark brown sticky compound. FTIR showed presence of functional groups with characteristic absorbance peaks at 3368 (NH), 2925, 2854 (C-H stretch), 1717, 1663 (C=O), 1456, 1370, 1316, 1268, 1077 (unsaturation) (Fig 2A). HRMS indicated a m/z of 230.1543 (M+H)+ for the compound. An introspection of the FTIR and HRMS data indicated that the molecule could have an empirical formula of C15H19NO with a molar mass of 229.1467 g/mol (Fig 3A). 1H-NMR (CDCl3, 400 MHz) data: 𝛿 7.18 (1H, dd, H3), 6.19 (1H, dd, H5), 6.02-6.08 (1H, m, H4), 5.82 (1H, d, H2), 5.66 (1H, br s, NH), 3.16 (2H, t, H1’), 2.38 (4H, d, C6 and C7), 1.98 (1H, s, H11), 1.74-1.83 (1H, sept, H2’), 0.92 (6H, d, H3’ and H4’) (Fig 4A, Table 1). 13C-NMR (CDCl3, 100 MHz) data: δ 166.35 (C1), 140.65 (C2), 139.27 (C3), 129.91 (C4), 123.22 (C5), 76.84 (C8), 68.37 (C9), 65.61 (C10), 65.19 (C11), 47.12 (C1’), 31.39 (C2’), 28.73 (C7), 20.25 (C3’), 20.25 (C4’), and 18.95 (C6) (Fig 4B, Table 1).
Compound 2 (C2): Physically it was a pale-yellow sticky compound. FTIR showed presence of functional groups with characteristic absorbance peaks at 3414 (NH), 2958, 2925 (stretch of C-H), 1657, 1627 (C=O), 1555, 1465, 1316, 1264, 998, 878, 781, 725 (unsaturation), 668 (Fig 2B). HRMS indicated a m/z of 248.2022 (M+H)+. An introspection of the FTIR and HRMS data indicated that the molecule could have an empirical formula of C16H25NO with a molar mass of 247.1936 (Fig 3B). 1H-NMR (CDCl3, 400 MHz) data: 𝛿 7.17 (1H, dd, H3), 6.30 (1H, ddt, H10), 6.15 (1H, dd, H4), 6.07 (1H, dt, H5), 5.96 (1H, br t, H9), 5.77 (1H, d, H2), 5.71 (1H, dq, C11), 5.65 (1H, br s, NH), 5.25 (1H, dt, H8), 3.15 (2H, t, H1’), 2.23-2.27 (4H, m, H6 and H7), 1.78 (3H, dd, H12), 1.78 (1H, sept, H2’), and 0.91 (6H, d, H3’ and H4’) (Fig 4C, Table 1). 13C-NMR (CDCl3, 100 MHz) data: δ 166.54 (C1), 142.17 (C2), 141.28 (C3), 129.97 (C4), 129.45 (C5), 128.79 (C8), 128.04 (C9), 126.86 (C10), 122.18 (C11), 47.08 (C1’), 33.13 (C2’), 28.73 (C6), 27.03 (C7), 20.25 (C3’), 20.25 (C4’), and 18.45 (C12) (Fig 4D, Table 1).
Analysis of the data from the three spectroscopic methods above and a comparison with the published reports confirmed the identity of C1 as 2E,4E-N-isobutyl undeca-2,4-diene-8,10-diynamide and and C2 as 2E,4E,8Z,10E -N-isobutyl dodeca-2,4,8,10-tetraenamide. For convenience, the C1 and C2 were abbreviated as NUD and NDT respectively.
3.3 Assessment of larvicidal activity of purified N-alkylamides
The larvicidal potency of NUD and NDT were tested on larvae of Ae. aegypti and Cx. quinquefasciatus. The larvae of both mosquitoes were found susceptible to the test compounds. The compounds tested and the methoprene (MET) standard induced mortality on the larvae in a dose-dependent manner (Supplementary Table 1 and 2). The LC50 and LC90 of compounds are presented in Supplementary Fig 1 and Table 2. NDT, a C12 alkene N-alkylamide, showed higher toxicity than the NUD, a C11 alkyne N-alkylamide (Supplementary Fig 1, Table 2). NUD and NDT showed a LC50 of 44.19 ppm and 18.28 ppm respectively for Ae. aegypti, while they were 30.89 ppm and 11.75 ppm, respectively, for Cx. quinquefasciatus larvae (Supplementary Fig 1, Table 2). LC90 for NUD and NDT were 90.17 ppm and 56.86 ppm respectively against Ae. aegypti, while they were was 86.46 ppm and 21.76 ppm, respectively, for Cx. quinquefasciatus (Supplementary Fig 1, Table 2). Both the compounds tested in this study in addition showed a potency exceeding that of MET standard with NDT being the most potent. NDT was 2 to 3 times more potent than NUD and 4 to 5 times that of MET (Table 2). Variation in the larvicidal potency could be attributed to the structural diversity in N-alkylamides resulting from the differences in chain length, type and degree of unsaturation (Figure 4B and 4D).
3.4 Analysis of Quantitative Structure Activity Relationship (QSAR)
The key physicochemical properties predicted using Swiss-ADME and Molinspiration tools indicated that both NUD and NDT did possess drug-like properties (Table 3). The difference in the larvicidal potency observed with test compounds was validated by QSAR analysis. Lipophilicity, expressed as log P value, is one of the physicochemical features that determine the solubility and membrane permeability of molecules and therefore used as a descriptor to evaluate drug-likeness under Lipinski’s rule of five. A correlation of the lipophilicity (log P) and LA (log 1/LC50) indicated that the log P value of MET fell outside the limit of Lipinski’s rule of five. For N-alkylamides, on the other hand, it was found within the specified limits (Fig 5, Table 3). It was intriguing to observe that MET had some deviation for drug-likeness under Lipinski’s rule of five. Swiss-ADME-mediated prediction of the biological targets for N-alkylamides revealed that both N-alkylamides, NUD and NDT, had some common binding partners such as family A G-protein coupled receptors (GPCRs), nuclear receptors and voltage-gated ion channels (Fig 6). A lower total polar surface area (TPSA), a determinant of the transport property related to intestinal absorption and blood-brain barrier, of the isolated N-alkylamides NUD and NDT in comparison to MET (Table 3) is indicative of high cellular permeability.
3.5 Determination of N-alkylamides content in different organs of A. ciliata
The calibration curves drawn for NUD and NDT isolated in the present study were linear with a high R2 value (Supplementary Fig 2, Table 5). HPLC of the test samples showed that the NUD and NDT content varied in the range of 2.06 mg/g to 68.55 mg/g by dry weight of the crude extract in different parts of A. ciliata. The compounds accumulated at its maximum in the flowers (19.01 and 68.55 mg/g, respectively) and turned out to be lowest in stem and leaf (Table 4). The LOD and LOQ of the HPLC method for NUD were 0.80 µg/ml and 2.51 µg/ml, respectively, while, these were 5.50 µg/ml and 16.70 µg/ml, respectively, for NDT (Table 5). The intra-day and inter-day precision of the method, expressed as percent relative standard deviation (RSD %), for estimation of NUD was 2.38 % and 2.71 %, respectively, whereas, the values of the same for NDT were 2.75 % and 3.37 % respectively (Table 5).