4. Discussion
Unlike other common essential oil crops such as peppermint (Mentha piperita) or lemon grass (Cymbopogon flexuosus (Nees ex Steud.)), in which the highest essential oil yield was achieved at 20 or 40 min distillation time ( Cannon et al., 2013), the essential oil yield of A. annua fresh biomass increased with the distillation time up to 240 min. This result may be due to differences in the structure of the glands that sequester the essential oils of peppermint, palmarosa (Cymbopogon martinii (Roxb.) J.F. Watson), and lemon grass, compared to A. annua. This finding is important, because most reports on A. annuaessential oil content seem to be based on much shorter distillation times, e.g., 120 min ( Ozguven et al., 2008 and Tzenkova et al., 2010). It is possible that previous reports based on shorter distillation time of A. annua may have underestimated the essential oil yield. However, overall essential oil yield in this study was similar to literature reports. For example, Malik et al. (2009) reported 0.28% of essential oil yield in A. annua fresh biomass subjected to hydrodistillation for 4 h. However, Héthelyi et al. (1995) reported essential oil content of flowering tops between 0.48 and 0.81%, but the authors did not describe the sample and the flowering tops probably did not include the stems (containing few essential oil glands).
This is the first report that attempts to explore the use of A. annua plant residue from distillation (PRD) as a source of artemisinin, and the first report on the antioxidant activity of PRD subjected to different distillation times. The results demonstrated that distillation using tap water at 100 °C, even for 1.25 min, led to the reduction of artemisinin in leaves from an average of 1.15% (undistilled control) to 0.18% ( Fig. 1). It is important to consider that the oil distillation was performed with plant material composed of stems and leaves and by using tap water, as to mimic industrial distillations. However, artemisinin was not degraded by refluxing leaves in boiling deionized, distilled, water for up to 20 min, and was reduced only from 0.7 to 0.5% at 40 min of boiling. This stability of artemisinin in hot water has not been previously reported. Unlike tap water, deionized, distilled water has no minerals, such as iron, that could react with artemisinin during boiling. Our controlled experiment in Soxhlet apparatus, where the leaves of an undistilled control were boiled with deionized, distilled water showed that artemisinin in the water phase was stable for at least 20 min. These findings indicate that distillation temperature alone for 1.25 min was not the only factor involved in the degradation of 84% of the artemisinin. Further research may be needed to elucidate this assumption. However, even if deionized, distilled water could be used in commercial distillation of A. annua, this technique would have to be improved to allow satisfactory extraction of the essential oil in 20 min, opening the possibility for the use of the PRD for posterior artemisinin extraction.
Regarding the antioxidant capacity of the PRD, our results demonstrated that PRD from 240 min of distillation (the longest time tested) still contained 75% of the antioxidant capacity of the undistilled control, as determined by the ORAC assay. This material represents tons of PRD resulting from essential oil distillation of large quantities of A. annua in commercial settings, and could potentially be incorporated into animal feed or be used as a source of antioxidants for nutraceuticals destined for human/animal consuption. Artemisia annua leaves have been previously shown to be a rich source of antioxidants, crude protein, crude fat, amino acids, and minerals including potassium, manganese, and copper ( Brisibe et al., 2009). In vitro studies also showed that A. annua leaves containing 1.4% artemisinin (g/100 g dw) had in vitro organic matter digestibility of 63.3%, compared to 68.5% in alfalfa (Medicago sativa), indicating that artemisinin had no negative impact on ruminal microflora and fermentation ( Turner and Ferreira, 2005). The leaves of this plant are highly nutritious, non-toxic, and contain three times more antioxidant capacity than conventional forages (e.g., Sericea lespedeza) and seven times more antioxidant capacity than alfalfa hay ( Ferreira, 2009), being of potential benefit for animal nutrition.
Regarding the anti-parasitic activity, in vivo trials using Boer and Boer cross goat artificially infected with the blood-sucking nematode Haemonchus contortus showed that infected animals fed 450 g A. annua/day for 6 days (1/3 of their daily diet) presented no signs of toxicity and kept the Haemonchus fecal egg counts (FEC) unchanged for 14 days. In contrast, goats infected with H. contortus and fed control diets had a 32% increase in FEC ( Hart et al., 2008). Also, 600–1000 mg kg−1 bw of ethanolic extracts of Brazilian and Chinese A. annua (1.26 and 2.7% artemisinin), or 200 mg kg−1 bw of artemisinin fed to gerbils artificially infected with H. contortus showed that both the plant and the drug resulted in no toxicity, although significant reduction in the number of adult worms was not achieved in such a system ( Squires et al., 2011). Although A. annua and artemisinin had a small, or no, impact on H. contortus infections in goats and gerbils, in vitro tests indicate that the plant extracts have good trematocidal and anti-cancer activity ( Singh et al., 2011)
Tests with petroleum ether and water for the extraction of artemisinin by refluxing, by pressurized liquid extraction (PLE), and by the traditional hot water tea procedure (infusion) showed that the use of water to extract artemisinin is the least efficient method, even when using PLE, which used water at 100 °C pressurized to 1500 psi (10.34 MPa) with nitrogen. In contrast, petroleum ether has been shown to be a more efficient solvent to extract artemisinin than water, although not the most efficient (Lapkin et al., 2006). The poor solubility of artemisinin in water only allows 0.048 mg mL−1 of artemisinin to be present in water at 37 °C, although our experience indicated that this solubility can be increased to 0.9 mg artemisinin/mL−1 of water in a hot tea situation (unpublished data). Unfortunately, petroleum ether and other organic solvents also extract several of the essential oil components, thus, limiting the oil distillation technology to water as the most feasible solvent, but destroying artemisinin in the plant material during the lengthy process. Fortunately, steam distillation only extracts 25% of the antioxidant components of the leaf, allowing the byproduct to be used as a source of antioxidants that can be applied immediately to animal feeding and, possibly, with applications in functional foods and other human health applications.
Acknowledgements
We thank Mr. Thomas Horgan of Mississippi State University for his help with the extraction of the essential oils. We greatly acknowledge the work efforts of both Mr. Harish Tummala (Woodraw Wilson High School, Beckley, WV) in helping to perform the experiments that validated our results, and of Mr. Barry Harter (USDA-ARS-AFSRC) for his valuable help with artemisinin and ORAC analyses during this collaborative project involving Mississippi State University, the USDA-ARS in Beaver-WV, and the University of Wyoming.