Chloroplast genome transformation of medicinal plant Artemisia annua

Artemisia annua produces lifesaving antimalarial drug artemisinin. The biosynthetic pathway of artemisinin is intercalated with chloroplasts of A. annua. Therefore, it will be imperative to optimize its chloroplast transformation for increasing the yield of artemisinin, which has not been achieved to desired levels via nuclear genome transformation (Ikram and Simonsen, 2017). The low yield has limited its current global demand for the artemisinin-based combination therapies (ACTs). For optimizing the chloroplast transformation, species-specific flanking sequences were amplified from the total genomic DNA of A. annua. Vector was named as CP-AA (LF:Prrn16S-g10-aadA-3'UTR-Trps::Ppsba-5'UTRpsba-eGFP-3UTRpsbA:RF) (Fig. 1a), (GenBank Accession # MT096403) was designed in silico and synthesized by vendor (BioBasic Inc., Canada). It was integrated between the intergenic spacer region of rrn16S-trnI and trnA–rrn23S via homologous recombination, without impacting the functionality of other genes (Jin and Daniell, 2015). The aadA (aminoglycoside adenylyltransferase) gene conferring resistance to spectinomycin was codon optimized (GenBank Accession # MT096403). Its expression was regulated by 5' ribosome-binding site region of the bacteriophage T7 gene10 (Olins et al., 1988) and 3'UTR rps16 from tobacco (NC_001879). The second ORF encodes the enhanced green fluorescence protein eGFP (Zhang et al., 1996) driven by light-regulated PEP promoter psbA, along with 5' and 3' UTRs of psbA of tobacco (NC_001879). The eGFP reporter gene was used for early detection of transgenic cultures and optimization of chloroplast transformation (Fig. 1c). Seeds were germinated as described (Alam and Abdin, 2011), and cultures were maintained in light conditions (16h light:8h dark), with 40-50 μmol m-2 s-1 intensity and temp 25 ± 2°C. The pinnately compound, sessile leaves (Fig. 1m), excised from aseptically grown 3-4 weeks old A. annua, were bombarded with 4 µg CP-AA DNA containing ~ 1.35x1011 copies of the plasmid coated on 0.6 µm gold particles (Seashell Technology, La Jolla, CA, USA) using a biolistic DNA delivery system as reported (Kumar and Daniell, 2004; Malhotra et al., 2016). After 48 hours of dark incubation, bombarded leaves were excised into small pieces and placed on selection medium (Fig. 1n). Several independent transgenic shoot primordia were observed after repetitive subculture on every two weeks for 2–3 months on MS medium containing plant growth regulators (1.25 mg/L BAP and 0.05 mg/L NAA) and 500 mg/L spectinomycin (Fig. 1o-p). These primordia were progressively proliferated into distinct shoots after 20 days on MS medium supplemented with reduced spectinomycin (250 mg/L) (Fig. 1q). Transgenic shoots were multiplied and elongated in 3–4 months (subculture after 2 weeks intervals) on MS medium when spectinomycin was reduced to one fourth (125 mg/L) (Fig. 1r). Rooting was induced only in elongated shoots after 5–7 weeks on MS medium containing 125 mg/L spectinomycin and 0.01 mg/L NAA (Fig. 1s-t). Four independent transplastomic plants with proper root were recovered. These plantlets after analysing for transgenes integration by PCR, Southern blot and eGFP protein expression by Western blot analysis were transferred to greenhouse. The transplastomic plants were further clonally propagated as reported (Wetzstein et al., 2018). Transplastomic plants thrive well in the greenhouse, and growth was comparable to WT plants (Fig. 1u). The early-stage transgenic green primordia screening on selection medium (Fig. 1o) was confirmed by eGFP expression (Fig. 1c). The site-specific integration of CP-AA vector cassette into the chloroplast genome was confirmed by primers AA-LF and AA-RF (Fig. 1b) that landed outside the flanking regions (LF-RF), yielding ~4.7Kb PCR amplicon in all transgenic shoot primordia in first round of selection (Fig. 1e). After repeated subculture of transgenic primordia on antibiotic selection, homoplastomic shoots were produced yielding PCR amplicon 4.7 Kb while heteroplastomic shoots also contained a faint signal of 1.8 Kb (Fig. 1f). The gene-specific PCR primers F-aadA and R-aadA; F-eGFP and R-eGFP (Fig. 1b) yielded amplicons ~0.8 Kb (Fig. 1g) and ~ 0.6 Kb (Fig. 1h), respectively, confirming integration of transgenes aadA and eGFP into the chloroplast genome. The homoplastomy status of integrated transgenes was confirmed by Southern blotting. Genomic DNA of transgenics T1-T4 and WT was digested with BamHI. The probe (0.65 Kb) was excised from CP-AA vector for Southern blot hybridization as shown (Fig. 1a). The presence of only ~ 3.5 Kb signal in transgenic plants (T1-T4) and absence of WT chloroplast background (0.9 Kb) has confirmed the homoplastomic status of plants (Fig. 1i). The fluorescent eGFP was used for a preliminary screening to distinguish the transgenics from WT, under the Nikon-A1 confocal microscopy. The eGFP micrographs of transgenics (Fig. 1c) showed auto red fluorescence of chlorophyll (excitation: 565 nm), (left), a green fluorescent micrograph expressing the eGFP (excitation: 485 nm) (middle), and merged micrograph of first two showed the colocalization of GFP expression in chloroplasts (right). Further, localization of eGFP in chloroplast genome was also confirmed by FP-425 primer (land 425 base pair outside the left flank) and R-eGFP primer bind to eGFP gene that yielding ~ 3.2 Kb PCR amplicon (Fig. 1d). For eGFP analysis, total 40 µg protein from transplastomic and WT plants was electrophoresed on 12% SDS-Polyacrylamide gel and stained with Coomassie Brilliant Dyes, yielding a 28 kDa signal only in transgenics lanes (Fig. 1k). The eGFP expression was confirmed by Western blot using a polyclonal GFP antibody conjugated with HRP. About 28 kDa signal was observed in all the four transgenic lines (T1, T2, T3 and T4). No signal was detected in WT lane (Fig. 1l). The maximum transformation efficiency (16.6%) was observed when explants were bombarded using 1100 psi rupture disc at 6 cm distance from microcarrier (Fig. 1j). No transgenics were recovered using rupture discs below 1100 psi or above 1350 psi. The use of 1800 psi rupture disc was deleterious to bombarded explants, causing an internal damage to tissues, leaching phenolic compounds, and cultures died in two weeks after browning. The optimization of chloroplast transformation in agronomically important recalcitrant species is a major challenge due to poor tissue culture response and inhibitory impact of antibiotics on the shoot and root induction in transplastomic tissues (Bock, 2015). In potato, only 3 lines were regenerated out of 104 plates subjected to bombardment, in Arabidopsis, only two transplastomic plants were obtained out of 201 bombarded leaves. In tomato, 3 transplastomic lines were recovered out of 30 leaf samples. However, obtaining a lesser number of chloroplast transgenic should not be a concern as chloroplast transformation is highly site specific via homologous recombination. All transplastomics as well as their clonally propagated plants produced a similar level of artemisinin when compared to WT plants, which indicated that there was no deleterious impact of aadA and eGFP transgenes on plants. In brief, we have optimized a stable chloroplast genome transformation in medicinal plant A. annua, which could be explored further for expressing the non-glycosylated proteins of artemisinin biosynthetic pathway to produce a higher quantity of artemisinin (Malhotra et al., 2016) to meet its global demand for ACTs. This work was supported by the Department of Biotechnology, Government of India through grant no. ND/DBT/18/017 to SK and UGC-RGN fellowship for PhD provided to CK. We are thankful to Dr. Karan Malhotra for helping in chloroplast transformation protocol and Dr. Srinivasan R., ICGEB, New Delhi for helping in Southern blot analysis. Thanks to Dr. Shalini Agarwal, DST-INSPIRE Faculty, Jawaharlal Nehru University, New Delhi, India for providing GFP polyclonal antibody. The authors declare that they have no competing or conflicting interests. Department of Biotechnology, Government of India through grant no. ND/DBT/18/017 to SK and UGC fellowship to CK. CK carried out the transformation optimization experiments and molecular analysis on transplastomic plants. MZA assisted in interpreting the results, reviewed the study. SK conceptualized the study and supervised the whole experimental research, data and analysis of results. CK and SK interpreted results and wrote the final manuscript. Not applicable. All authors read the manuscript and approved to publish. All data generated or analysed during this study are included in this article and whole vector description, and DNA sequence information is uploaded to public domain with GenBank Accession # MT096403.