*2.3. Selection Methods and Regeneration Protocols for Transplastomic Cells and Plants*

Ideally, plant species commonly used for plastid transformation, such as tobacco, should have a well-developed tissue culture system and high regeneration potential. It is therefore generally considered that plastid transformation is confined to crops that fulfill these two criteria. Different plant species, or even different cultivars from the same species, require specific tissue culture conditions. Protocols for in vitro culture often take time to set up, including the optimization of the growth temperature, humidity, the composition of the culture medium, to name a few variables. Plastid transformation studies and their applications are thus still largely restricted to certain plant species like rice, tobacco, and lettuce.

Although plant tissue culture conditions vary extensively, the selection procedure for screening homoplasmic plants are quite similar and have not changed much over the past decades [60]. Several antibiotics have been used in plastid transformation [15,61]. For example, the expression of *Neomycin Phosphotransferase* (*nptII*) or *Aminoglycoside 3 -Phosphotransferase* (*aphA6*) confers resistance to kanamycin [31,62], while the expression of *Bialaphos Resistance* (*bar*) provides resistance to the herbicide glyphosate in transgenic plants. However, the gene *Streptomycin 3 -Adenylyltransferase* (*aadA*), a spectinomycin and streptomycin resistance gene, remains the most commonly used marker for plastid transformation [15,60]. Spectinomycin inhibits plastid protein translation by binding to chloroplast ribosomes [63,64]. At least in Chlamydomonas and tobacco, mutations in the plastid ribosomal subunit targeted by spectinomycin confer resistance to the antibiotic. In fact, the first report of plastid transformation in tobacco used a plastid DNA fragment carrying such a mutation to demonstrate transformation potential and act as a selectable marker at the same time [50].

Parker et al. reported that one spectinomycin tolerance strategy in Arabidopsis involves Acetyl-Coa Carboxylase 2 (ACC2) [65,66]. Indeed, seedlings cannot develop beyond the cotyledon stage under spectinomycin selection when they have a functional copy of the nuclear *ACC2* gene, thereby limiting selection efficiency in Arabidopsis plastid transformation. Based on these results, Yu et al. went on to use an *acc2* loss of function Arabidopsis mutant (SALK\_148966C) to test the effectiveness of particle bombardment with a plasmid encoding an aadA-GFP fusion [25]. Transformation efficiency increased around 100-fold in the *acc2* mutant background. Ruf et al. later used clustered regularly interspaced short palindromic repeats (CRISPR)/Cas9-mediated genome editing to inactivate the *ACC2* locus and create a recipient line for plastid transformation. Transformants in this *acc2* mutant background grew to maturity and produced seeds, as the loss of ACC2 function is not accompanied by visible growth phenotypes [67]. These reports may, therefore, open an avenue for routine engineering of the plastid genome in Arabidopsis, and provide valuable information for plastid transformation in other, more recalcitrant species.
