Pharmacokinetics of ivermectin metabolites and their activity against Anopheles stephensi mosquitoes

Abstract

Background: Ivermectin (22,23-dihydroavermectin B_1a: H₂B_1a) is an endectocide used to treat worm infections and ectoparasites including lice and scabies mites. Furthermore, survival of malaria transmitting Anopheles mosquitoes is strongly decreased after feeding on humans recently treated with ivermectin. Currently, mass drug administration of ivermectin is under investigation as a potential novel malaria vector control tool to reduce Plasmodium transmission by mosquitoes. A "post-ivermectin effect" has also been reported, in which the survival of mosquitoes remains reduced even after ivermectin is no longer detectable in blood meals. In the present study, existing material from human clinical trials was analysed to understand the pharmacokinetics of ivermectin metabolites and feeding experiments were performed in Anopheles stephensi mosquitoes to assess whether ivermectin metabolites contribute to the mosquitocidal action of ivermectin and whether they may be responsible for the post-ivermectin effect.

Methods: Ivermectin was incubated in the presence of recombinant human cytochrome P₄₅₀ 3A4/5 (CYP 3A4/5) to produce ivermectin metabolites. In total, nine metabolites were purified by semi-preparative high-pressure liquid chromatography. The pharmacokinetics of the metabolites were assessed over three days in twelve healthy volunteers who received a single oral dose of 12 mg ivermectin. Blank whole blood was spiked with the isolated metabolites at levels matching the maximal blood concentration (C_max) observed in pharmacokinetics study samples. These samples were fed to An. stephensi mosquitoes, and their survival and vitality was recorded daily over 3 days.

Results: Human CYP3A4 metabolised ivermectin more rapidly than CYP3A5. Ivermectin metabolites M1-M8 were predominantly formed by CYP3A4, whereas metabolite M9 (hydroxy-H₂B_1a) was mainly produced by CYP3A5. Both desmethyl-H₂B_1a (M1) and hydroxy-H₂B_1a (M2) killed all mosquitoes within three days post-feeding, while administration of desmethyl, hydroxy-H₂B_1a (M4) reduced survival to 35% over an observation period of 3 days. Ivermectin metabolites that underwent deglycosylation or hydroxylation at spiroketal moiety were not active against An. stephensi at C_max levels. Interestingly, half-lives of M1 (54.2 ± 4.7 h) and M4 (57.5 ± 13.2 h) were considerably longer than that of the parent compound ivermectin (38.9 ± 20.8 h).

Conclusion: In conclusion, the ivermectin metabolites M1 and M2 contribute to the activity of ivermectin against An. stephensi mosquitoes and could be responsible for the "post-ivermectin effect".

Fig. 1

Ivermectin metabolism. Chemical structure of ivermectin and putative structure of the metabolites (M1–M9): M1: Desmethyl-H₂B_1a, M2: Hydroxy-H₂B_1a, M3: Hydroxy-H₂B_1a, M4: Desmethyl, hydroxy-H₂B_1a, M5: Hydroxy-H₂B_1a monosaccharide, M6: Desmethyl, hydroxy-H₂B_1a, M7: Hydroxy-H₂B_1a monosaccharide, M8: Dihydroxy-H₂B_1a, M9: Hydroxy-H₂B_1a. Ivermectin (10 µM) was incubated for 60 min in the presence of CYP3A4 (blue) or CYP3A5 (red) enzymes (supersomes). Incubations were performed in the presence (empty circle) or absence (filled circle) of ketoconazole (1 µM), a potent 3A inhibitor. Formation of the ivermectin metabolites is shown as mean peak area ratio (analyte peak area: ivermectin-d₂ area). The error bars correspond to the standard error of the mean. A representative chromatogram recorded 60 min post-incubation is depict for each metabolite transition. Ivermectin-d₂ (IVM-d₂) was added as point of comparison into each chromatogram

Fig. 2

Ivermectin metabolites pharmacokinetics in human. Mean blood level-time curves of ivermectin metabolites determined in pharmacokinetic (PK) study samples. A number of 12 individuals received a single oral dose of 12 mg ivermectin. The blood samples of the participants were analysed to derive the maximal blood level (C_max) from the metabolite peak intensity time plots. The error bars correspond to the standard deviation. The C_max (blue) and T0 (yellow) chromatograms of PK subject 6 are shown next to the blood level-time curves. The metabolite signals of spiked blank blood samples (T0 sample of PK subject 6) are colored in red. The arrow in the chromatogram indicates whether the amount of the spiked sample had to be increased (red) or decreased (blue) to reach the peak intensity of C_max samples. If a metabolite was not detected in PK samples, as for M9, the metabolite level was adjusted to the lower limit of detection of the method. In this case, blank blood fed to mosquitoes was spiked with the metabolite to obtain a signal intensity three times greater than the corresponding noise signal. It should be noted that the peak area of M4 in spiked blood samples was on average larger than the observed C_max levels. However, the M4 peak area of the spiked blank sample of subject 6 was smaller than the observed C_max. For this reason, the amount of M4 was reduced to obtain values that corresponded to the average C_max of all 12 study participants (Additional file 1: Table S5)

Fig. 3

Screening for ivermectin metabolites effect on mosquito mortality. Anopheles stephensi mosquitoes were treated with blank human blood, blood containing ivermectin (IVM, 50 ng/mL), or ivermectin metabolite (M1-M9). Data represent three independent replicates per compound, each containing an average of 28 mosquitoes/condition. A Mortality over time. Mean mosquito survival was assessed after 24 h, 48 h and 72 h. Error bars correspond to the standard error or the mean. B Percent survivorship of An. stephensi 72 h; after feeding on ivermectin or its metabolites (M1–M9). Significant differences in the means of two groups are marked, *** shows a p-value < 0.001 with a t-test

Fig. 4

Anopheles stephensi survival probability after imbibing blood meals containing ivermectin and ivermectin metabolites. Anopheles stephensi mosquitoes were given a blood meal that contained ivermectin (IVM) or its metabolites (M1–M9) levels corresponding to those interpolated from the IVM pharmacokinetic curve. The mosquitoes’ survival was monitored at 24 h, 48 h and 72 h post feeding. Ivermectin (H₂B_1a) metabolites: M1: Desmethyl-H₂B_1a, M2: Hydroxy-H₂B_1a, M3: Hydroxy-H₂B_1a, M4: Desmethyl, hydroxy-H₂B_1a, M5: Hydroxy-H₂B_1a monosaccharide, M6: Desmethyl, hydroxy-H₂B_1a, M7: Hydroxy-H₂B_1a monosaccharide, M8: Dihydroxy-H₂B_1a, M9: Hydroxy-H₂B_1a. Mosquitoes that exhibited significantly reduced survival (Log rank test) compared to control (blood meal contained only dimethyl sulfoxide) have a p-value < 0.0001

Fig. 5

Ivermectin metabolites effect on mosquito mortality: LC₅₀. Multiple amounts of ivermectin (IVM), desmethyl-H₂B_1a (M1) or hydroxy-H₂B_1a (M2) were fed to mosquitoes to determine the lethal concentration that killed 50% of the mosquitoes (LC₅₀) 72 h after treatment. The blood samples of the participants of the pharmacokinetics trial [25] were analysed to derive the maximal blood level (C_max) for ivermectin, M1 and M2 (Fig. 2). An administration of 12 mg ivermectin yielded a C_max of 50 ng/mL for ivermectin, therefore the LC₅₀ was evaluated using ivermectin concentrations ranging between 1 and 12.5 ng/mL. For the metabolites M1 and M2, C_max corresponds to the maximal intensity (peak area, counts) from the metabolite peak intensity time plots. Anopheles stephensi were treated with different dilutions of ivermectin (¹/₄ C_max, ¹/₆ C_max, ¹/₁₀ C_max, ¹/₂₀ C_max, ¹/₅₀ C_max), M1 (C_max, ¹/₂ C_max, ¹/₄ C_max, ¹/₆ C_max, ¹/₁₀ C_max) and M2 (C_max, ¹/₂ C_max, ¹/₃ C_max, ¹/₄ C_max, ¹/₅ C_max). Mean mosquito survival was assessed after 24, 48 and 72 h. Error bars correspond to the standard error of the mean