Flavonoid Versus Artemisinin Anti-malarial Activity in Artemisia annua Whole-Leaf Extracts

Abstract

Artemisinin, a sesquiterpene lactone produced by Artemisia annua glandular secretory trichomes, is the active ingredient in the most effective treatment for uncomplicated malaria caused by Plasmodium falciparum parasites. Other metabolites in A. annua or related species, particularly flavonoids, have been proposed to either act as antimalarials on their own or act synergistically with artemisinin to enhance antimalarial activity. We identified a mutation that disrupts the CHALCONE ISOMERASE 1 (CHI1) enzyme that is responsible for the second committed step of flavonoid biosynthesis. Detailed metabolite profiling revealed that chi1-1 lacks all major flavonoids but produces wild-type artemisinin levels, making this mutant a useful tool to test the antiplasmodial effects of flavonoids. We used whole-leaf extracts from chi1-1 and mutant lines impaired in artemisinin production in bioactivity in vitro assays against intraerythrocytic P. falciparum Dd2. We found that chi1-1 extracts did not differ from wild-type extracts in antiplasmodial efficacy nor initial rate of cytocidal action. Furthermore, extracts from the A. annua cyp71av1-1 mutant and RNAi lines impaired in amorpha-4,11-diene synthase gene expression, which are both severely compromised in artemisinin biosynthesis but unaffected in flavonoid metabolism, showed very low or no antiplasmodial activity. These results demonstrate that in vitro bioactivity against P. falciparum of flavonoids is negligible when compared to that of artemisinin.

Keywords: Artemisia annua; Plasmodium falciparum; artemisinin; chalcone isomerase; flavonoids; malaria.

FIGURE 1

Discovery and characterization of the chi1-1 mutation. (A) Artemisia annua CHI1, CHI2 and CHI3 expression in meristems, cotyledons, young leaf trichomes, young leaves, fully expanded leaves, mature leaves, stems, flowers, and no template control (NTC) were determined by semi-quantitative PCR. UBQ (Putative ubiquitin-like protein, GQ901904) was used as a loading control. (B) Gene schematic of CHI1 indicates the position of the chi1-1 mutation. (C) The A. annua CHI1 protein structure was modeled by I-TASSER (Yang et al., 2015) on the 10 most closely related structural analogs. The parts of the structure expected to be missing in the chi1-1 mutant are highlighted in yellow, naringenin (enzyme product) bound to CHI1 is shown in blue.

FIGURE 2

Effects of the chi1-1 mutation on the metabolite profile of Artemisia annua. Box and whisker plots showing levels of (A) four major flavonoids, putative naringenin chalcone and artemisinin as measured by UPLC-MS in young (leaves 1–5 as counted from the apical meristem), mature (leaves 11–13), old (three leaves above first senescing leaf) and dry (oven-dried) leaf material harvested from 12 to 14-week-old plants of the Artemis wild type (black), heterozygous (blue) and homozygous chi1-1 mutant (red) and (B) selected flavonoids, sesquiterpenes and monoterpenes in the extracts from dipped (dip) or ground leaf material for the wild type (black) and heterozygous (blue) and homozygous (red) chi1-1 mutant. Metabolite concentrations measured by GC- or UPLC-MS (A,B) are expressed as a proportion of the residual dry leaf material following extraction. Groups not sharing letters representing Tukey’s range test results indicate statistically significant differences (p < 0.05). Each box is represented by minimum of 20 (A) or by five (B) biological replicates. (C,D) Principal component analysis of 83 UPLC-MS identified peaks (C) and of 58 GC-MS identified peaks (D) from dipped and ground leaf material from wild type (black) and heterozygous (blue) and homozygous chi1-1 (red). Dip leaf extracts are represented by circles and ground leaf extracts by triangles. Principal component analysis was performed on log-scaled and mean-centered data.

FIGURE 3

Comparison of in vitro antiplasmodial activity of leaf extracts from Artemisia annua wild type, mutant and antisense lines with altered flavonoid and artemisinin content. (A,C) Log concentration-normalized response curves of Plasmodium falciparum parasites after 48 h of treatment with extracts used to determine the EC₅₀ (50% effective concentration of extract needed to inhibit growth of the P. falciparum parasites) of the indicated extracts. (B,D) Bioluminescent Relative Rate of Kill (BRRoK) assays to determine the initial (6 h) cytocidal action, compared to an untreated control after exposure to extracts of wild type, heterozygous and homozygous chi1-1, and the AMS silenced line (B) or heterozygous and homozygous cyp71av1-1 (D) at multiples of the EC₅₀ alongside dihydroartemisinin (DHA) > chloroquine (CQ) > mefloquine (MQ) > atovaquone (ATQ) benchmark controls. Error bars represent standard deviations from the means of three biological replicates.