Experimental methods
Malaria, a vector-borne disease caused by Plasmodium falciparum, infects 500 million people annually, predominately in sub-Saharan Africa (Snow et al., Reference Snow, Guerra, Noor, Myint and Hay2005). Resistance of P. falciparum to the widely used antimalarial drug chloroquinone means that this control approach is now ineffective in most malarial regions (Talisuna et al., Reference Talisuna, Okello, Erhart, Coosemans and D'Alessandro2007). Accordingly, the World Health Organization (WHO) has recommended administration of artemisinin (ART) in combination with other antimalarial drugs (artemisinin-based combination therapy, ACT) in chloroquinone-resistant regions (WHO, 2000). Artemisinin is a sesquiterpene lactone endoperoxide extracted from the predominantly outbreeding medicinal plant Artemisia annua L. (2n = 18). Although ART biosynthesis has not yet been fully elucidated, it involves the cytosolic mevalonate and plastid-located 1-deoxyxylulose pathways (Covello et al., Reference Covello, Teoh, Polichuk, Reed and Nowark2007; Towler and Weathers, Reference Towler and Weathers2007). As ART cannot currently be economically synthesized using bulk fermentation or chemical synthesis (Van Noorden, Reference Van Noorden2010), extraction from plants remains the most economically feasible method. However, commercial supply of ART has historically fluctuated due to limitations in the quality of available plant material, climatic conditions and agricultural management. Concentration of ART in surveys of A. annua accessions is known to be predominantly < 1%, with the highest content reported being 1.4% (Woerdenbag et al., Reference Woerdenbag, Pras, Chan, Bang, Bos, Van Uden, Phan Van, Van Boi, Batterman and Lugt1994; Delabays et al., Reference Delabays, Simonnet and Gaudin2001; Kumar et al., Reference Kumar, Gupta, Singh, Bajpai, Gupta, Singh, Gupta, Ram, Shasany and Sharma2004; Khanuja et al., Reference Khanuja, Paul, Shasany, Gupta, Darokar, Gupta, Verma, Ram, Kumar, Lal, Bansal, Singh, Bhakuni and Tandon2005). As part of a wider programme aimed at enhancing ART concentration and supply (Davies et al., Reference Davies, Atkinson, Burns, Woolley, Hipps, Arroo, Dungey, Robinson, Brown, Flockart, Hill, Smith and Bentley2009), we report the screening of a collection of 70 A. annua plants sampled from 14 diverse (and genetically heterogeneous) germplasm accessions, identifying germplasm with ART concentrations >2.0%.
Fourteen A. annua accessions from a range of geographical locations were collated (Table 1). For each accession, five plants were germinated in a heated glasshouse, and transplanted into field nurseries (45 × 45 cm spacing, single individuals from each population arranged within each of five blocks, following a randomized block design) on 30th April 2010 (Cambridge, UK, 52.20°N, 0.13°E). Flowering time (date on which the first individual within each accession flowered) and overall vegetative vigour (1: very vigorous, 2: vigorous, 3: weak, 4: heterogeneous) were scored. Additionally, ART concentrations (% wt/dry wt) were determined for each individual at harvest: material harvested from longitudinal sections through each plant was collected on 19th September and dried at 40°C for 24 h in a fan-assisted oven. Solid-phase extraction from dried leaf material was performed by sonication with analytical grade hexane (Fisher Scientific Ltd, Loughborough, UK). Aliquots of the supernatants were evaporated under N2 gas flow and the dry extract reconstituted into mobile phase (1 ml) prior to injection. ART concentration was analysed using an Agilent 1100 HPLC equipped with a Phenomenex Luna C18 column (250 mm × 4.6 mm, particle size 5 μm, pore size 100 Å) and a UV DAD detector (215 nm), mobile phase = acetonitrile/water (60/40, v/v) at a mobile phase flow rate of 1 ml/min, column oven temperature: 25°C. Calibration of the HPLC method was determined using 10 mcg/ml of pure artemisinin standard (purity >98%; confirmed by melting range, found to be 150–152°C) every tenth sample during the determination of the investigated samples, returning coefficients of variation of 1%. Exemplar individuals from each accession are maintained at NIAB by vegetative propagation (Waseem et al., Reference Waseem, Jilani, Khan, Kiran and Khan2011 ).
Table 1 Artemisinin (ART) concentration, vegetative vigour and flowering time scores in the Artemisia annua accessions investigated
SD, standard deviation.
a Mean ART concentration and SD of five individual plants per accession are shown, as well as the individual line with the highest ART concentration for each accession. ART is measured as % of dry weight.
b Vegetative vigour: 1 (very strong growth), 2 (strong growth), 3 (weak growth), 4 (segregating for growth type).
c Flowering in the field: Y (flowering), N (no flowering).
All but two accessions (1001 and 1012) flowered when grown in the field under UK summer day lengths (maximum: 16 h 48 mins, 21st June; minimum: 12 h 28 mins, 19th September). Mean ART concentration within the 14 accessions was found to vary from 0.08% (accession 1014) to 2.03% (accession 1015) (Table 1), with calibration controls of pure artemisinin standards returning coefficients of variation of ~1%. Intra-accession standard deviations (SDs) ranged from 0.02 to 0.64, with individual 1015-4 displaying the highest ART concentration (2.22%). Indeed, all five individuals from accession 1015 displayed elevated levels of ART, resulting in a high mean (2.03% ART) and a low SD (0.19). The only other accession with mean ART concentration >1% was 1013 (mean 1.45, SD 0.64), including an individual with an ART concentration of 2.16 (1013-4). The highest-yielding accession (1015) was grown in the field the following year, included within a replicated block design, returning a mean ART concentration of 2.10, indicating consistently high ART yields across the two growth seasons investigated. Evaluation of overall growth type (Table 1) found just two accessions (1016, 1017) to possess very vigorous growth habit (score 1); both displayed relatively low mean ART concentrations ≤ 0.5 (Table 1). However, high-yielding ART lines 1013 and 1015 both possessed good (score 2) growth habits.
Discussion
Despite increasing deployment of ACT, and the recent development of genomic tools in A. annua (Graham et al., Reference Graham, Besser, Blumer, Branigan, Czechowski, Elias, Guterman, Harvey, Issac, Khan, larson, Li, Pawson, Penfield, Rae, Rathbone, Reid, Ross, Smallwood, Segura, Townsend, Vyras, Winzer and Bowles2010), comparatively little systematic evaluation has been reported in A. annua germplasm. Here, we screen a wide range of A. annua lines for ART concentration and agronomic suitability to European cultivation. The finding that the majority of accessions flowered under UK summer photoperiods indicates that the previously reported requirement for short-day photoperiods (~8 h) for floral initiation is not absolute, or that the critical minimum day length is higher than previously assumed. The competence to flower under UK conditions demonstrates that future breeding activities can be conducted without the need to artificially reduce photoperiod. Considerable variation in ART concentration was found within accessions. Previous studies show that ART concentration is a highly heritable trait (Delabays et al., Reference Delabays, Simonnet and Gaudin2001). However, due to the out-breeding nature of A. annua, considerable intra-accession genetic variability is predicted to be present. Accordingly, ART concentrations in A. annua are reported to vary wildly, ranging between 0.03 and 1.4% of dry weight. Here, we identify A. annua lines combining good growth habit (score 2) and competence to flower under long-day photoperiods, with ART concentrations in excess of 2.0%, providing enhanced chemovars from which improved parental lines and commercial F1 hybrid seeds can be developed. The establishment of improved breeding lines forms the basis from which future genetic and genomic studies can be best undertaken. Immortal lines for all accessions are maintained at NIAB (http://www.niab.com/) by vegetative propagation.
Acknowledgements
The present work was funded by Defra through the Renewable Materials LINK Programme (LK0822, ‘Developing and alternative UK industrial crop A. annua, for the extraction of artemisinin to treat multi-drug-resistant malaria’), with financial input from the Horticultural Development Company and the NIAB Trust.
