While this is an issue that must be resolved for field growth, it is also a potential opportunity for gene containment since high PHB produ-cing seeds inadvertently lost in the field would not survive. An appropriate inducible gene switch system, such as the ecdysone system demonstrated for inducible overproduc-tion of abscisic acid in Arabidopsis seeds [29] and inducible PHB production in Arabidopsis leaves [30], could possibly be used to generate viable seedlings for field planting.
Progress has also been made with increasing the pro-duction of PHB in organelles other than plastids in C3 plants by manipulation of enzymes central to the metab-olism of acetyl-CoA in the cytosol [31] and peroxisomes [32]. PHB production in the cytosol of Arabidopsis results in a dwarf phenotype [33], with a reduction of nearly 90%in fresh weight compared to wild type plants [31]. A similar phenotype has also been observed in Arabidopsis plants with lower levels of ATP citrate lyase (ACL) activity [34], an ATP-dependent cytosolic enzyme that generates acetyl-CoA and oxaloacetate from citrate and CoA. Recent work has shown that overexpression of ACL in Arabidopsis cytosolic PHB producers increased rosette size, fresh weight, leaf size, and inflorescence height compared to parental lines without an ACL transgene [31]. While only a small increase in PHB per unit dry weight was observed with ACL expression and amounts of polymer produced per unit dry weight were still very low (0.15 mg/mg DW), the larger healthier plants increased total plant PHB levels by 4.5–6.9 fold. Inter-estingly, the production of PHB alone is enough to increase ACL activity expressed from the endogenous copy of the gene, likely through post-translational modi-fications that correlate with sink strength.
Peroxisomal production of PHB has previously been examined as a possible route for high level polymer production since acetyl-CoA is generated within the organelle from the b-oxidation of fatty acids (Figure 2b). Initial work to engineer peroxisomal pro-duction of polymer in Arabidopsis produced a PHBV copolymer that was primarily PHB (>95%) at levels up to 0.11% DW but yielded a shorter hypocotyl length of etiolated transgenic Arabidopsis seedlings [35]. This observation suggested that competition for acetyl-CoA between PHB synthesis and endogenous peroxisomal metabolic pathways, such as the glyoxylate cycle and gluconeogenesis, may limit production of polymer [35]. Citrate synthase converts peroxisomal acetyl-CoA to citrate within the glyoxylate cycle and thus is an appro-priate target for maximizing carbon flow to PHB. Recent studies have reduced peroxisomal citrate synthase activity by targeted microRNA knockdown in plants engineered for peroxisomal PHB production [32]. Growth of these plants in soil produced approximately three fold more PHB than parental lines grown under the same conditions yielding up to 1.7% DW PHB [32].