Plant Physiological Adaptations to the Massive Foreign Protein Synthesis Occurring in Recombinant Chloroplasts

摘要

Genetically engineered chloroplasts have an extraordinary capacity to accumulate recombinant proteins. We have investigated in tobacco (Nicotiana tabacum) the possible consequences of such additional products on several parameters of plant development and composition. Plastid transformants were analyzed that express abundantly either bacterial enzymes, alkaline phosphatase (PhoA-S and PhoA-L) and 4-hydroxyphenyl pyruvate dioxygenase (HPPD), or a green fluorescent protein (GFP). In leaves, the HPPD and GFP recombinant proteins are the major polypeptides and accumulate to higher levels than Rubisco. Nevertheless, these engineered metabolic sinks do not cause a measurable difference in growth rate or photosynthetic parameters. The total amino acid content of transgenic leaves is also not significantly affected, showing that plant cells have a limited protein biosynthetic capacity. Recombinant products are made at the expense of resident proteins. Rubisco, which constitutes the major leaf amino acid store, is the most clearly and strongly down-regulated plant protein. This reduction is even more dramatic under conditions of limited nitrogen supply, whereas recombinant proteins accumulate to even higher relative levels. These changes are regulated posttranscriptionally since transcript levels of resident plastid genes are not affected. Our results show that plants are able to produce massive amounts of recombinant proteins in chloroplasts without profound metabolic perturbation and that Rubisco, acting as a nitrogen buffer, is a key player in maintaining homeostasis and limiting pleiotropic effects. nnnn--------------------------------------------------------------------------------nThe genetic modification of the plastid genome was achieved in higher plants more than 15 years ago (Svab et al., 1990; Svab and Maliga, 1993). This recombinant technology presents distinctive features that are very attractive from a biotechnological perspective. The most attractive of these features is the potential for extremely high expression of the transgene in plastid transformants, up to 70% of total soluble proteins (tsp) in leaves for an antibacterial lysin (Oey et al., 2009). A variety of pharmaceutical proteins have also been produced at very high levels in transgenic chloroplasts (Daniell et al., 2004; Daniell, 2006). The other characteristics of the technology have been extensively reviewed recently (Maliga, 2004; Bock, 2007; Verma and Daniell, 2007; Dubald et al., 2008) and concern (1) the targeted insertion of the transgenes, (2) the possibility to engineer more easily complex pathways using polycistronic vectors, (3) the apparent absence of epigenetic regulation, and (4) the natural confinement of the transgenes as a result of the almost exclusive maternal inheritance of these organelles. nChloroplasts have an extraordinary capacity to synthesize and accumulate foreign proteins. Curiously, very little attention has been devoted to evaluate and analyze the consequences on the plant physiology of this significant metabolic burden. In most reports, which include an insecticidal toxin expressed at 46% tsp (De Cosa et al., 2001), or a GFP expressed at 38% tsp (Yabuta et al., 2008), no obvious phenotypic defect, such as growth retardation, has been observed in plastid transformants. When phenotypic modifications were noted, these were directly linked to the specific properties of the expressed transgenes (Tregoning et al., 2003; Magee et al., 2004; Ruiz and Daniell, 2005; Chakrabarti et al., 2006; Hasunuma et al., 2008; Tissot et al., 2008). Only in the case of lysin was the hyperexpression of the recombinant protein reported to limit plant development by exhausting the protein synthetic capacity of chloroplasts (Oey et al., 2009). A number of issues are therefore still very unclear: (1) are the recombinant proteins produced on top of the resident proteins, meaning that plants naturally have the capacity to make significantly more proteins, at least if a sink is provided? Otherwise, (2) are they produced at the expense of all, or of only some, resident proteins, implying that these resident proteins are normally synthesized in excess? And (3), how are resources allocated between resident and recombinant proteins when the cell budget is reduced, in particular when there is a limitation in nitrogen supply? nnPlastid transformants expressing recombinant proteins at a high level provide a unique material to address these fundamental questions. To draw generic conclusions, we have for the first time, to our knowledge, studied in parallel transgenic tobacco (Nicotiana tabacum) lines expressing recombinant proteins of completely different nature: (1) a hydroxyphenyl pyruvate dioxygenase (HPPD) from Pseudomonas (Dufourmantel et al., 2007), which participates in plants in the synthesis of plastoquinones and is the target of various herbicides (Matringe et al., 2005); (2) an alkaline phosphatase from Escherichia coli (Bally et al., 2008) with no known substrate in chloroplasts, targeted to the thylakoids (PhoA-L) or expressed at a lower level in the stroma (PhoA-S); and (3) a GFP, with no enzymatic function, accumulating strongly in the stroma. We have started investigating the impact that massive transgene expression in chloroplasts may have on plant development, photosynthesis, leaf proteome, chloroplast transcriptome, and amino acid composition.