SARS-CoV-2 omicron variants are susceptible in vitro to Artemisia annua hot water extracts

Abstract

Ethnopharmacological relevance

Artemisia annua L. has >2000 yr of history in treating fever a symptom common to many infectious diseases including viruses. The plant is widely used as a tea infusion in many areas of the globe to thwart many infectious diseases.

Aim of the study

The SARS-CoV-2 (COVID-19) virus continues to infect millions while rapidly evolving new variants that are more transmissible and evade vaccine-elicited antibodies, e.g., omicron and its subvariants. Having shown potency against all previously tested variants, A. annua L. extracts were further tested against highly infectious omicron and its recent subvariants.

Materials and methods

Using Vero E6 cells, we measured the in vitro efficacy (IC₅₀) of stored (frozen) dried-leaf hot-water A. annua L. extracts of four cultivars (A3, BUR, MED, and SAM) against SARS-CoV-2 variants: original WA1 (WT), BA.1 (omicron), BA.2, BA.2.12.1, and BA.4. End point virus titers of infectivity in cv. BUR-treated human lung A459 cells overexpressing hu-ACE2 were determined for both WA1 and BA.4 viruses.

Results

When normalized to the artemisinin (ART) or leaf dry weight (DW) equivalent of the extract, the IC₅₀ values ranged from 0.5 to 16.5 μM ART and from 20 to 106 μg DW. IC₅₀ values were within limits of assay variation of our earlier studies. End-point titers confirmed a dose-response inhibition in ACE2 overexpressing human lung cells to the BUR cultivar. Cell viability losses were not measurable at leaf dry weights ≤50 μg for any cultivar extract.

Conclusions

A. annua hot-water extracts (tea infusions) continue to show efficacy against SARS-CoV-2 and its rapidly evolving variants and deserve greater attention as a possible cost-effective therapeutic.

Graphical abstract

Abbreviations:

ART

artemisinin

CC₅₀

concentration of drug(s) that killed 50% of cells

DL

drug-likeness

GCMS

gas chromatography mass spectrometry

IC₅₀

concentration of drug(s) that inhibited virus by 50%

MOI

multiplicity of infection

OB

oral bioavailability

PPQ

piperaquine

TCID

tissue culture infectious dose

VOC

variant of concern

1. Introduction

Artemisia annua L. has a long ethnobotanical history of use especially in treating fevers resulting from infectious diseases like malaria (Hsu, 2006; Tu, 2016). Fever is also a typical symptom of viral infections. Although human viruses are likely very ancient, even predating cellular origins (Krupovic et al., 2019), they have only recently been recognized as infectious agents in humans with the first identified in 1901, the YFV flavivirus that causes yellow fever (Woolhouse et al., 2008). People in Kenya and Uganda have now also been documented using A. annua tea as a treatment for viral infection of HIV/AIDS (Hirt et al., 2008; Lubbe et al., 2012; Willcox et al., 2011). Previously, we and others showed that extracts of dried leaves of many cultivars of the medicinal plants, A. annua, which produces the antimalarial sesquiterpene lactone, artemisinin (ART; Fig. 1 ), and A. afra Jacq. ex Willd., a related perennial species lacking artemisinin, prevented SARS-CoV-2 replication in vitro (Nair et al., 2021, 2022; Nie et al., 2021; Zhou et al., 2021). Although ART has some anti-SARS-CoV-2 activity, we also showed that antiviral efficacy was inversely correlated to ART content (Nair et al., 2021).

Fig. 1.

Artemisia annua L. and artemisinin.

The novel coronavirus, SARS-CoV-2 (a.k.a., COVID-19), with its rapidly evolving variants continues to plague the global population with >650 million cases resulting overall in >6.6 million deaths (https://coronavirus.jhu.edu/map.html, accessed June 26, 2022). The past year the omicron (BA.1) variant of concern (VOC) emerged along with a number of subvariants, especially BA.4 and BA.5 (Tegally et al., 2022). These are highly transmissible (Omicron RO ≥ 10, Delta RO = 7 (Burki, 2022)) infecting even vaccinated individuals, albeit with less severe outcomes (https://www.healthdata.org/covid/COVID-19-vaccine-efficacy-summary, accessed June 26, 2022). Omicron (BA.1) isolates have shown resistance to neutralization by antibodies in patients who have had COVID-19 or been vaccinated and even boosted with one of the widely used vaccines (Cao et al., 2022; Iketani et al., 2022; Liu et al., 2022). Recently, variants, BA.2.12.1 and BA.4/5 were shown to be 1.8 and 4.2 times, respectively, more resistant to sera from individuals who were vaccinated and boosted (Wang et al., 2022). Additionally, recent clinical case studies showed that vaccinated and boosted individuals who took a course of Paxlovid™ have shown relapse and relapsed individuals accidentally infect family members (Charness et al., 2022). This presents an even more pressing need for an expanded diversification of therapeutics, which may also serve as prophylactics in a population setting.

Although a number of different drugs have been trialed (Sakamuru et al., 2022) there are few approved small-molecule drugs available to treat COVID-19. The antiviral drug Paxlovid™ was recently approved as a per os combination drug of nirmatrelvir (or PF-07321332) with ritonavir, developed by Pfizer having good anti-SARS-CoV-2 efficacy and relatively few adverse drug reactions (Hung et al., 2022; Lamb, 2022). The viral protease inhibitor, nirmatrelvir (Hung et al., 2022), works with ritonavir. The latter inhibits CYP3A4 to improve the pharmacokinetics of nirmatrelvir (Owen et al., 2021). Despite this success, access to the drug may be limited (https://www.nature.com/articles/d41586-022-00919-5, accessed June 26, 2022) because generic production, while affordable in developed countries is unaffordable to many worldwide (https://www.reuters.com/business/healthcare-pharmaceuticals/generic-drugmakers-sell-pfizers-paxlovid-25-or-less-low-income-countries-2022-05-12/, accessed June 26, 2022). Thus, there remains a need for more cost-effective antiCOVID-19 therapeutics to treat the global population.

Here we report in vitro efficacy for four of the seven originally studied A. annua L. cultivars against omicron (BA.1) and three of its subvariants: BA.2, BA.2.12.1, and BA.4.

2. Materials and methods

2.1. Plant material, extract preparations, and artemisinin analyses

Hot-water extracts (tea infusions) were previously prepared from vegetative stage dried leaves of Artemisia annua L. (SAM, MASS 00317314; BUR, LG0019527; A3, Anamed; MED, KL/015/6407) In brief: 10 g dried leaves/L were boiled in water for 10 min, solids removed via sieving, then 0.22 μm filter-sterilized and stored at −80 °C for this study and as detailed in Nair et al. (2021). ART analyses of tea infusions were by gas chromatography mass spectrometry (GSMS) as previously described (Martini et al., 2020) using an Agilent 7890 A GC; MS, Agilent 5975 C; column, Agilent HP-5MS (30 m × 0.25 mm × 0.25 _m) column and He carrier gas at 1 mL/min; injection volume, 1 _μL in splitless mode; ion source temperature, 230 °C; inlet, 150 °C; transfer line, 280 °C; oven temperature, 125 °C held for 1 min, then increased to 240 °C at 5 °C/min, and then increased to 300 °C at 30 °C/min. The natural log transformation of the sum of the three peaks correlating to the ART standard (two thermal breakdown peaks and one intact molecule peak) was used to generate a standard curve. The natural log of the total peak area (Y) is plotted against the natural log of the concentration (X) and a linear equation is obtained (Y = mX + b, where m is slope and b is y-intercept). The concentration of ART in unknown samples is calculated from that equation and ART contents were detailed in (Nair et al., 2021): ART in μg/mL was: 42.5 for A3; 20.1 for BUR; 59.4 for MED; and 149.4 for SAM. TLC fingerprints and GCMS chromatograms showing the three ART peaks of the tea infusions of each of the four tested cultivars are available in Supplemental Materials Figs. S1–S3.

2.2. Vero E6 culture and infection

Cultivation of Vero E6 cells (ATCC CRL-1586) was in Essential Minimal Eagle's Medium (EMEM) containing penicillin-streptomycin (1 × 100 U/mL) and 10% fetal calf serum, and viral infections were performed as previously described (Nair et al., 2021). SARS-CoV-2 isolates (Table 1 ) were sourced from BEI Resources (www.beiresources.org). To determine their tissue culture infectious dose (TCID), viruses were titrated after propagation in Vero E6 cells, aliquoted, and frozen at −80 °C until later use. Multiplicity of infection (MOI) was 0.1 as used in other studies (Liu et al., 2020; Nair et al., 2021, 2022).

Table 1.

SARS-CoV-2 isolates used in this study.

SARS-CoV-2 isolate	BEI Resource Catalogue number
USA/WA12020	NR-52281 SARS-Related Coronavirus 2, Isolate USA-WA1/2020
Omicron BA.1	NR-56475 SARS-Related Coronavirus 2, Isolate hCoV-19/USA/HI-CDC-4359259-001/2021
Omicron BA.2	NR-56520 SARS-Related Coronavirus 2, Isolate hCoV-19/USA/CO-CDPHE-2102544747/2021
Omicron BA.2.12.1	NR-56781 SARS-Related Coronavirus 2, Isolate hCoV-19/USA/NY-MSHSPSP-PV56475/2022
Omicron BA.4	NR-56806 SARS-Related Coronavirus-2, Isolate hCoV-19/USA/MD-HP30386/2022

2.3. Drug inhibition assays of SARS-CoV-2 and cell viability

Dilutions of extracts were incubated for 1 h in 96-well tissue culture plates having a monolayer of Vero E6 cells seeded the prior day at 20,000 cells/well. SARS-CoV-2 virus was added to each well 1 h after extract addition to a final MOI of 0.1. Cells were cultured for 3 days in 5% CO₂ at 37 °C and then scored for cytopathic effects as previously detailed (Liu et al., 2020) and values converted into percent of control. Drug concentrations were log transformed and the concentration of drug(s) that inhibited virus by 50% (i.e., IC₅₀), and the concentration of drug(s) that killed 50% of cells (i.e., CC₅₀; viability), were log-transformed and determined via nonlinear logistic regressions of log (inhibitor) versus response-variable dose-response functions (four parameters) constrained to a zero-bottom asymptote by statistical analysis. We already reported viability of Vero E6 cells post extract treatment (Nair et al., 2021) for the same extracts. To normalize the IC₅₀ values for the new variants tested or the WT and variants tested previously, dry mass of leaves and total ART contents measured in the Artemisia extracts as previously described (Nair et al., 2021).

2.4. Human lung cell line culture and viral infection

Human lung tissue derived cells, line A549 (ATCC CCL-185), engineered to overexpress the human angiotensin converting enzyme (ACE2) by stable transfection of a hu-ACE2 expressing lentiviral construct under puromycin selection (Ikhlas et al., 2022) were cultured in Essential Minimal Eagle's medium (EMEM) with 10% fetal calf serum (FCS). These A549 cells stably expressing hu-ACE2 were plated overnight in 24 well plates to form a monolayer. The monolayer was treated with two doses of tea extract (cv BUR) on the next day. One dose (50 μg DW) was higher and the other lower (3.125 μg) than the IC50 of the tea against the WA1 virus for this cultivar. Virus (either USA/WA1 or Omicron BA.4) was added to the treated cells within 15 min post treatment. Plates were incubated at 37 °C/5% CO2 for 48 h to allow the virus to replicate. At 48 h, the infected/treated cells were subjected to end-point virus titration using Vero-ACE2/TMPRSS2 cells. Briefly, the supernatant was removed, and cells washed three times with phosphate-buffered saline (PBS) to remove any unbound virus and leftover residual extracts. Each washed monolayer of cells in the well was treated with 150 μL of 0.25% Trypsin-EDTA (Corning) and incubated for 3 min at 37 °C/5% CO₂. Following neutralization of the trypsin with EMEM containing 10% FCS, the cells were collected from the wells into a tube, centrifuged at 300×g for 5 min to pellet them. The pellet was recovered and resuspended in 250 μl of fresh EMEM+10% FCS. A 1:1 dilution in trypan blue was used to count the cell number per sample using the automated Biorad TC20 cell counter. Cells were then diluted in EMEM to set up an endpoint titration in 96 well plates starting at 10,000 cells/well. Twelve 3-fold dilutions of the cells were overlaid on Vero cells expressing ACE2 and TMPRSS2 (Vero/ACE2/TMPRSS2), which are extremely sensitive to virus replication and show vivid cytopathic effects. Cells were incubated at 37 °C/5%CO₂ for 72 h prior to determining the viral endpoint titer from each of the treated concentrations of the extract with eight replicates of titration from each concentration. A549-ACE2 cells that did not receive any extract and infected with the virus at same MOI were used as controls to determine the maximal endpoint titer achievable in the Vero-ACE2-TMPRSS2 cells.

2.5. Chemicals and reagents

Reagents were procured from Sigma-Aldrich (St. Louis, MO). Renilla-Glo was from Promega (E2720). EMEM (Cat # 30–2003) and XTT reagent (Cat # 30-1011 k) were from ATCC.

2.6. Statistical analyses

The anti-SARS-CoV-2 analyses were done at least in triplicate. Plant hot water extracts had n ≥ 6 independent assays as previously described (Nair et al., 2021). IC₅₀ values were calculated using GraphPad Prism V9.4. The endpoint titer of each replicate was scored and plotted using GraphPad Prism V9.4 and statistics using t-test (Wilcoxon matched-pairs signed rank test) was performed between the groups to determine the p-value. The percentage neutralization was calculated and plotted using GraphPad Prism V9.4 software.

3. Results and discussion

Hot-water extracts of four cultivars of A. annua inhibited the recently evolved omicron and its three tested subvariants of SARS-CoV-2 with IC₅₀ values calculated and normalized to the ART content of each tested tea infusion ranging from 0.5 to 16.5 μM ART (Fig. 2 ; the lower the IC₅₀, the more potent the drug/extract). When the IC₅₀ values were instead normalized to the dry mass of the extracted A. annua leaves, values ranged from 20 to 106 μg (Fig. 3 ). Although the values for these new variants are for the most part slightly higher than the IC₅₀ values reported for variants previously reported (Nair et al., 2021, 2022) and all are summarized in Table 2 , they fell within limits of assay variation. As already reported for extracts used in this study, there was no measurable loss of cell viability at a dry weight of ≤50 μg for any cultivar extract (Nair et al., 2021). Others have reported in vitro efficacy of A. annua (Nie et al., 2021; Zhou et al., 2021) and A. afra (Nie et al., 2021) extracts against earlier variants of SARS-CoV-2; however, to our knowledge there are no reports showing efficacy against omicron or its variants.

Fig. 2.

SARS-CoV-2 variant inhibition by four cultivars of A. annua L. hot water extracts normalized to their artemisinin (ART) content and compared to WT. WT, USA/WA1; omicron and its variants: BA.1, BA.2, BA.2.12.1, and BA.4 at a multiplicity of infection (MOI) of 0.1 in Vero E6 cells. Data are plotted from an average of three replicates from each of two experiments ± SEM.

Fig. 3.

SARS-CoV-2 variant inhibition by four cultivars of A. annua L. hot water extracts normalized to their A. annua leaf dry mass (DW) and compared to WT. WT, USA/WA1; omicron and its variants: BA.1, BA.2, BA.2.12.1, and BA.4 at a multiplicity of infection (MOI) of 0.1 in Vero E6 cells. Data are plotted from an average of three replicates from each of two experiments ± SEM.

Table 2.

Comparative IC₅₀ values of A. annua L. hot-water extracts (10 g/L) against all current and previously tested strains of SARS-CoV-2 based on either artemisinin content or leaf dry weight (DW).

Cultivar	Potency normalized to artemisinin content
	IC50 μM artemisinin
	WA1a (WT)	B.1.1.7a (alpha)	B.1.351a (beta)	P.1b (gamma)	B.1.617.1b (kappa)	B.1.617.2b (delta)	WA1 (WT)	BA.1 (omicron)	BA.2	BA.2.12.1	BA.4
SAM	3.4	4.9	8.4	7.9	7.0	7.0	4.9	8.2	7.6	5.3	16.5
A3	0.8	1.1	2.0	1.9	1.9	2.1	1.2	1.8	1.0	1.5	2.6
BUR	0.4	0.3	0.8	1.2	1.1	1.2	0.7	1.2	0.6	0.5	1.1
MED	2.9	2.0	3.6	2.9	2.5	4.8	4.2	4.7	2.1	3.9	6.4
Cultivar	Potency normalized to dry mass of leaves used in tea infusion
	IC50μg leaf DW
	WA1a	B.1.1.7a	B.1.351a	P.1b	B.1.617.1b	B.1.617.2b	WA1	BA.1	BA.2	BA.2.12.1	BA.4
SAM	21.5	31.3	53.7	50.7	45.0	45.1	31.4	52.5	48.9	34.1	106.0
A3	15.7	22.1	39.6	38.2	37.0	42.4	23.9	36.7	20.0	30.7	50.9
BUR	15.1	11.0	32.5	50.1	44.7	49.8	27.5	48.9	23.6	22.7	45.2
MED	41.7	28.2	51.5	41.0	37.0	67.7	59.7	67.0	30.0	56.0	90.6

IC₅₀s are values where virus is 50% inhibited. Data are an average of three replicates from each of two experiments.

^aValues taken from Nair et al. (2021).

^bvalues taken from Nair et al. (2022).

End point titer was measured for BUR, the most potent of the four tested cultivars, and results (Fig. 4 ) confirmed our previous dose-response studies (Nair et al., 2021, 2022) showing a dose-dependent response. The highest dose in the study had minimal viral load and indicated that initial viral loads at 50 μg/mL were almost minimal (>1000-fold lower) compared to that of the virus controls, even after the extract was removed. To quantify this further, we measured the frequency of infectivity for each group by calculating the average endpoint titers of the 8 replicates and averages are shown in Table 3 . The frequency of infectivity was inversely proportional to the dry weight equivalent of the added leaf extract thereby allowing us to conclude that the cv BUR extracts do have an inhibitory effect on the replication of SARS-CoV-2 in human A549-ACE2 cells.

Fig. 4.

Endpoint titer of cv. BUR hot water extract vs. USA/WA1 and omicron BA.4. Titer is shown as relative endpoint infectivity vs. untreated infected controls and compared at three dosages based on the equivalent dry weight of leaf mass of the applied A. annua extract. N = 8; bars are ± SEM for 3 replicated experiments.

Table 3.

Comparative averages of endpoint titers and corresponding frequency of infection for USA/WA 1 and Omicron BA.4.

Leaf dry mass (μg)	USA/WA1		Omicron BA.4
	Endpoint titer (Mean ± SEM)	Frequency of infection = 1/Endpt titer	Endpoint titer (Mean ± SEM)	Frequency of infection = 1/Endpt titer
50	1759.26 ± 473.43	0.0006	432.10 ± 101.70	0.0023
3.13	68.59 ± 16.40	0.0146	106.31 ± 41.46	0.0095
Controls	1.35 ± 0.50	0.7387	6.86 ± 2.06	0.1458

The observed dose-response was also proportional to the viral isolate tested; cv. BUR showed slightly less than two-fold increase in the IC₅₀ value against Omicron BA.4 as compared to the USA/WA1 original isolate when tested in the Vero E6 cells (Fig. 2). This difference is reflected in the current endpoint titer assay using human A549 lung cells, where the sensitivity of the extracts dropped by four-fold against Omicron BA.4 (endpoint titer 432 ± 101) compared to the WA1 (1759 ± 473) for the highest 50 μg/ml dose and about 1.5-fold against Omicron BA.4 (endpoint titer 106 ± 41.5) versus WA1 (68.5 ± 16.4) for the 3.125 μg/ml dose.

Using VSV-based spike pseudovirus we previously showed that inhibition was not at virus entry into cells, (Nair et al., 2021). Those results were reconfirmed here wherein none of the VSV-based pseudotyped viruses (WA1/WT, Omicron BA.1.1 or Omicron BA.4) were inhibited at any of the doses tested from 100 μg dry weight of the leaves at 5-fold dilutions (Supplemental Fig. S4). Calculated IC₅₀s of all tea extracts were greater than the highest dose tested indicating that these extracts do not have inhibitory activity at the virus entrance into the cell. As an assay control, serum from a patient who has broad neutralization responses was tested showing expected data that the more recently evolved Omicron variants BA.1 or BA.4 are more resilient to serum antibodies (Fig. S4) than the ancestral WA1 virus. Given that the drop in potency to the extracts (especially cv BUR) is far lower, future mechanistic studies, should focus on interaction of the Artemisia extracts at post-entry steps to inhibit viral replication.

Although ART IC₅₀ as well as leaf dry mass IC₅₀ values are shown in this study, we previously reported that potency was inversely related to ART concentration (Nair et al., 2021) and others showed that A. afra, a species lacking ART, was also highly effective in vitro against SARS-CoV-2 (Nie et al., 2021). Nevertheless, ART has some anti-SARS-CoV-2 activity as we and others showed (Cao et al., 2020; Gendrot et al., 2020a, 2020b; Hu et al., 2021; Nair et al., 2021; Nie et al., 2021; Zhou et al., 2021). Although there have been some clinical studies using ART, it was used as a combination therapy. For example, in a small non-randomized controlled trial where patients were treated with ART-piperaquine (ART-PPQ) or placebo the mean time for recovery when there was no longer PCR-detectable virus was 10.6 d for ART-PPQ treated patients vs. 19.3 d for those receiving placebo (Li et al., 2021). All patients treated with ART-PPQ were virus-free after 21 d compared to 36 d for placebo. In another small trial patients had faster recovery vs. placebo in those who used ArtemiC, an oral spray containing ART, curcumin, frankincense, and vitamin C (Hellou et al., 2022). To our knowledge, however, there are no reports of clinical trials using only A. annua or its extracts.

Because we and others (Nair et al., 2021; Nie et al., 2021) showed that ART is not the most likely anti-SARS-CoV-2 therapeutic phytochemical in A. annua extracts, questions remain regarding the identity of these non-ART phytochemicals. To resolve that question, several groups have used in silico approaches (Shahhamzehei et al., 2022; Tang et al., 2022). Tang et al. screened the Traditional Chinese Medicines for systems Pharmacology Databased and Analysis Platform to identify all phytochemicals reportedly in A. annua then ranked them according to oral bioavailability and drug likeness (OB and DL, respectively). That list was narrowed to 19 compounds within their OB and DL limits of ≥30% and 0.18, respectively. They concluded that many on the list of 19 compounds had anti-inflammatory, immune regulatory, and therapeutic properties. Among the top therapeutic candidates were luteolin and isorhamnetin. Using a ZINC library the Efferth lab also screened an in silico library of >39,000 natural product compounds including some from plants with known antiviral activity and narrowed their hits to 33 likely compounds (Shahhamzehei et al., 2022). Of the top 12, three, isorhamnetin, luteolin, and rosmarinic acid, are present in A. annua and when tested in vitro had IC₅₀ values of 8.42, 11.81, and 9.43 μM, respectively, against the main protease in SARS-CoV-2, 3CL^pro, a chymotrypsin-like protease involved in viral replication. Along with reports of anti-SARS-CoV-2 activity of other A. annua phytochemicals, e.g., quercetin and myricetin against NTPase/helicase (Russo et al., 2020; Solnier and Fladerer, 2021), many other small molecules, especially flavonoids, are showing antiviral potential and likely work in combination (synergistically) in these extracts to achieve the therapeutic response.

While artemisinin is always of interest when studying the antimicrobial efficacy of A. annua, it is not likely the main anti-SARS-CoV-2 molecule. Artemisinin has some antiviral activity, however, it appears to be antagonistic to the overall anti-SARS-CoV-2 activity of the hot water extracts (Nair et al., 2021). Of eight different tested cultivars of A. annua containing different amounts of artemisinin, there was an inverse relationship between artemisinin content and extract potency; IC₅₀ values decreased in potency with increasing artemisinin content (Nair et al., 2021). Further evidence for non-artemisinin anti-SARS-CoV-2 activity in Artemisia sp. was provided by Nie et al. who showed that A. afra extracts lacking artemisinin were equally potent against the virus as A. annua (Nie et al., 2021). Clearly, future studies are needed to isolate, identify and validate the activity of nonartemisinin putative anti-SARS-CoV-2 therapeutic compounds in A. annua and in combination with artemisinin.

While authors are not promoting any one type of delivery, in low- and middle-income countries a traditional tea infusion may be appropriate, culturally acceptable, cost-effective, and perhaps the only delivery mode available, especially for people in rural areas. For those in the more developed parts of the world, an encapsulated form of powdered dried leaves is likely more reasonable. Using ART as a marker molecule, therapeutically relevant concentrations post-consumption can be delivered from tea infusions (Räth et al., 2004). They showed A. annua tea infusions containing 94.5 mg ART/delivered dose/subject delivered high levels of ART into the serum of healthy human subjects with a C_max of 240 μg/L post tea consumption. Delivery as powdered Artemisia leaves per os, e.g., in a capsule is also possible. Evidence in rats showed that compared to ingestion of pure ART, serum levels of ART 1 h post oral gavage were 7.4 and 18.7 time greater in males and females, respectively, from A. annua than from pure ART, a result consistent among 6 tested organs including lungs (Desrosiers et al., 2020). Delivery as powdered Artemisia leaves in a capsule tested in one human had serum levels of A. annua-delivered ART at 7.04 μg/mL post consumption of 3 g of powdered and encapsulated A. annua leaves (Nair et al., 2021). Furthermore, again using ART as a marker molecule, TLC data of MeOH and DCM extracts of leaves, vs. a hot water extract of a tea infusion, show that tea infusions have an equivalent amount of extracted artemisinin as DCM and MeOH extracts (Supplemental TLC data Fig. S3).

4. Conclusions

Hot-water (tea infusion) extracts of A. annua continue to show activity against SARS-CoV-2 and the newest VOCs including omicron and three of its highly transmissible subvariants. Although the specific phytochemicals have not yet been identified, there are a number of possible candidates emerging in the literature. Validation of A. annua extracts against SARS-CoV-2 VOCs in a rodent model are needed as a next step towards human trials. Nevertheless, this plant is safe to use, and we urge testing in clinical trials sooner rather than later. WHO announced in 2021 that through its COVID-19 Solidarity Therapeutics Plus Trial that it included intravenous artesunate as one of three repurposed drugs to treat COVID-19 (Kupferschmidt, 2021). Results are not anticipated until 2023, (last accessed July 11, 2022, https://www.isrctn.com/ISRCTN18066414). Regardless of outcome and based on the continuing efficacy of A. annua extracts against all tested variants (10 to date), we again urge the WHO to consider including encapsulated dried leaf A. annua as a separate arm in their trial. While there is no evidence that use of A. annua induces ART resistance (Elfawal et al., 2015), the plant could be crucial in helping many in the world where access to vaccines and standard therapeutics is logistically challenging.

Funding

Award Number NIH-2R15AT008277-02 to PJW from the National Center for Complementary and Integrative Health funded phytochemical analyses of the plant material used in this study. The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Center for Complementary and Integrative Health or the National Institutes of Health.

Institutional review board statement

Not applicable.

Informed consent statement

Not applicable.

CRediT authorship contribution statement

M.S. Nair: Conceptualization, Methodology, Formal analysis, Writing – review & editing, Supervision. Y. Huang: Methodology, Formal analysis. M. Wang: Methodology. P.J. Weathers: Conceptualization, Writing – original draft, preparation, Writing – review & editing.

Declaration of competing interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Handling Editor: Dr. Thomas Efferth

Appendix A. Supplementary data

The following are the Supplementary data to this article:

Data availability

Data will be made available on request.