JXB Advance Access originally published online on February 3, 2008
Journal of Experimental Botany 2008 59(7):1909-1921; doi:10.1093/jxb/erm311
RESEARCH PAPER |
Construction of a tobacco master line to improve Rubisco engineering in chloroplasts
Molecular Plant Physiology Group, Research School of Biological Sciences, Australian National University, Canberra, Australian Capital Territory 0200, Australia
* To whom correspondence should be addressed. E-mail: spencer.whitney{at}anu.edu.au
Received 21 September 2007; Revised 26 October 2007 Accepted 12 November 2007
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and methodsResults and discussionSupplementary dataReferences |
|---|
The inability to assemble Rubisco from any photosynthetic eukaryote within Escherichia coli has hampered structure–function studies of higher plant Rubisco. Precise genetic manipulation of the tobacco chloroplast genome (plastome) by homologous recombination has facilitated the successful production of transplastomic lines that have either mutated the Rubisco large subunit (L) gene, rbcL, or replaced it with foreign variants. Here the capacity of a new tobacco transplastomic line, cmtrL, to augment future Rubisco engineering studies is demonstrated. Initially the rbcL was replaced with the selectable marker gene, aadA, and an artificial codon-modified cmrbcM gene that codes for the structurally novel Rubisco dimer (L2,
100 kDa) from Rhodosprillum rubrum. To obtain cmtrL, the aadA was excised by transiently introducing a T-DNA encoding CRE recombinase biolistically. Selection using aadA enabled transplantation of mutated and wild-type tobacco Rubisco genes into the cmtrL plastome with an efficiency that was 3- to 10-fold higher than comparable transformations into wild-type tobacco. Transformants producing the re-introduced form I tobacco Rubisco variants (hexadecamers comprising eight L and eight small subunits, 520 kDa) were identified by non-denaturing PAGE with fully segregated homoplasmic lines (where no L2 Rubisco was produced) obtained within 6–9 weeks after transformation which enabled their Rubisco kinetics to be quickly examined. Here the usefulness of cmtrL in more readily examining the production, folding, and assembly capabilities of both mutated tobacco and foreign form I Rubisco subunits in tobacco plastids is discussed, and the feasibility of quickly assessing the kinetic properties of those that functionally assemble is demonstrated. Key words: CO2 fixation, mutagenesis, photosynthesis, plastome transformation, Rubisco
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and methodsResults and discussionSupplementary dataReferences |
|---|
Significant advances have been made over 50 years into various aspects of Rubisco biology (Portis and Parry, 2007), with considerable efforts still pursuing a more comprehensive appreciation of the enzymes complex catalytic chemistry and identifying what features afford the natural kinetic variability found amongst the different Rubiscos, in the hope of engineering improved variants into crop plants (Andrews and Whitney, 2003; Raines, 2006; Tcherkez et al., 2006; Parry et al., 2007). The inability to assemble Rubisco readily from any eukaryote in Escherichia coli (Gatenby et al., 1987; Cloney et al., 1993) or cyanobacteria (Whitney et al., 2001) had thwarted mutagenic studies of the higher plant Rubisco holoenzyme until the development of techniques to transform the chloroplast genome (plastome) of some plants efficiently. While the range of species whose plastome can be transformed continues to expand (Koop et al., 2008), the transplantation of foreign or modified Rubisco variants into higher plant plastids has so far been exclusively performed in tobacco where the transformation efficiency is highest. Efforts are therefore focused on using tobacco as a model higher plant species to appreciate more fully the requirements and limitations to modifying the endogenous Rubisco into higher plant plastids as well as transplanting in foreign variants (Andrews and Whitney, 2003).
For the Rubisco large (L) subunit gene (rbcL), located in the plastome, genetic manipulation by plastome transformation in tobacco is a routine but protracted process. Generally at least 6 months is required to obtain transplastomic T0 tobacco lines with a completely segregated population of transformed plastomes (homoplasmic) and be of suitable maturity to measure the Rubisco kinetics and undertake physiological analyses, such as measuring its photosynthetic properties by whole leaf gas exchange and chlorophyll fluorescence. Previous efforts have been able to mutate the hexadecameric form I Rubisco in tobacco (Whitney et al., 1999) and replace it either totally, with the form II dimer (L2) from the bacterium Rhodospirillum rubrum (Whitney and Andrews, 2001a, 2003), or partially, by replacing rbcL with the sunflower gene (rbcLS) to produce a hybrid form I Rubisco comprising eight sunflower L subunits and eight cytosolically synthesized tobacco small (S) subunits (Kanevski et al., 1999). However, success was not forthcoming in producing a comparable hybrid Rubisco comprising cyanobacterial L subunits, or with expressing Rubisco L and S subunits from two non-green algae whose folding and assembly requirements were not met by the tobacco chloroplast chaperone complexes (Whitney et al., 2001). An appropriate means for engineering the native (or foreign) S subunit genes (RbcS) in higher plants also remains an elusive challenge due to the multiple RbcS copies that are located in the nucleus in higher plants which essentially precludes RbcS from targeted mutagenic or replacement strategies. Multiple attempts to assemble appreciable levels of plastid-synthesized tobacco S subunits showed an apparent necessity for very high levels of plastomic rbcS mRNA expression and reduced levels of the native RbcS message (Whitney and Andrews, 2001b; Zhang et al., 2002; Dhingra et al., 2004).
The present focus is on addressing five primary issues related to Rubisco engineering in higher plant chloroplasts. The first is identifying ways to control the level of Rubisco expression in chloroplasts with some degree of predictability by examining the influence of different 5' and 3' regulatory sequences and codon usage. Two other issues involve examining what classes of foreign Rubiscos can assemble in higher plant chloroplasts without chaperone engineering and whether the L subunits of other plant species can functionally assemble with tobacco S subunits without catalytic detriment, as has been found for the sunflower L–tobacco S hybrid Rubisco (Sharwood et al., 2008). The roles of the multiple L subunit post-translational modifications on their expression, folding, and assembly also remain to be established (Houtz and Portis, 2003), as does ascertaining to what extent foreign S subunits co-transplanted with their cognate foreign L subunit can outcompete the tobacco S subunits for functional assembly.
In this study, the development of a master tobacco line called cmtrL, that may be effective at addressing many of these issues, is demonstrated. The transplastomic line expresses R. rubrum Rubisco whose gene lacks sequence homology with re-introduced Rubisco genes, which beneficially excluded undesired, alternative recombination events and accelerated obtaining homoplasmic lines. The distinct structure (L2) and antigenicity of R. rubrum Rubisco in this master line from that of form I (L8S8) Rubiscos enabled the subunit expression, functional assembly, and catalytic prowess of re-introduced mutated and wild-type tobacco Rubisco to be assessed within 2 months following transformation.
|
Materials and methods |
|---|
|
TopAbstractIntroductionMaterials and methodsResults and discussionSupplementary dataReferences |
|---|
Plasmid construction, sequencing, and transformation
All plasmids and PCR products were sequenced using BigDye terminator sequencing (Applied Biosystems) on an ABI 3730 sequencer (Biomolecular Resource Facility, JCSMR, ANU). Plasmid pcmtrLA contains 1126 bp and 1176 bp of flanking tobacco plastome sequence that directs the homologous replacement of rbcL and most of its 3'-untranslated sequence with a codon-modified R. rubrum rbcM gene (cmrbcM) and an inversely oriented aadA selective marker gene (GenBank accession no. AY827488 [GenBank] ). The cmrbcM gene was assembled by primer extension overlap PCR using 80-mer primers that overlapped with adjacent primers by 15 nucleotides as described (Whitney and Sharwood, 2007). The codon bias of cmrbcM was modified to mimic tobacco rbcL (Supplementary Fig. S1 available at JXB online). Plastome transforming plasmids directing the re-insertion of the tobacco rbcL were derived from pLEV1 (L-subunit Encoding Vector; Whitney et al., 1999). Nucleotide 163 in the rbcL coding region was deleted following erroneous digestion and re-ligation with SacII to give pLEV
L. pL335V codes a leucine to valine mutation in rbcL at codon 335 and contains an adjacent silent mutation that introduces a unique EcoRV site (Whitney et al., 1999). The H6Ub-RbcS gene codes for a fusion protein comprising the 6xhistidine-tagged ubiquitin sequence from pHue (Baker et al., 2005) and the tobacco RbcS sequence from pRVSSu (Whitney and Andrews, 2001b). Sequence comprising 29 bp of the intergenic sequence (ISPt) immediately 5' to the rbcS gene in the Phaeodactylum tricornutum rbcLS operon (that is highly expressed in tobacco plastids; Whitney et al., 2001) was cloned immediately 5' to the H6Ub-RbcS gene and the ISPt-H6Ub-RbcS sequence inserted 3' to the rbcL stop codon in pLEV1 to give plasmid pLEVUbS. The plasmids were transformed into E. coli XL1-Blue cells, grown at 25 °C in LB media containing ampicillin (0.2 mg ml–1), purified using the Wizard mini-prep kit (Promega), and eluted in water then concentrated to 1 µg µl–1 by freeze drying. A 10 µg aliquot of freshly prepared plasmid was coated onto tungsten for biolistic transformation as described (Svab and Maliga, 1993). pcmtrLA was transformed into the plastome of tobacco (Nicotiana tabacum L. cv. Petit Havana [N,N]), producing cmtrLA lines. The aadA gene was excised following biolistic transformation of homoplasmic T0 tissue with the T-DNA plasmid pKO27 that codes for CRE recombinase containing an N-terminal fusion for the pea Rubisco S subunit plastid-targeting sequence (Corneille et al., 2001). Homoplasmic aadA-free cmtrL lines were obtained (see below), their flowers successively backcrossed with wild-type pollen, and the T2 progeny of cmtrL line 1 transformed with pLEV1, pLEVL, pL335V, and pLEVUbS.
Propagation of material in tissue culture
Eight cmtrLA transplastomic lines from 10 spectinomycin-resistant (specR) plants that grew on selective medium [agar-solidified Murashige–Skoog salts containing 3% (w/v) sucrose, 0.5 mg ml–1 spectinomycin, and hormones (Svab and Maliga, 1993)] were identified by PCR (see below). Leaf tissue from two cmtrLA lines (1 and 3) was carried through an additional round of regeneration on selective media and homoplasmicity was confirmed by DNA blot analysis (see below). Replicate plants for both cmtrLA lines were regenerated aseptically in Magenta pots (Sigma) on MS medium (selective medium without hormones or spectinomycin) and were either grown to maturity in soil under high CO2 pressures [1% (v/v) CO2 in air] as described (Whitney and Andrews, 2001a), or the sterile leaves of cmtrLA line 1 were transformed with pKO27 (Corneille et al., 2001) using the biolistic method (Svab and Maliga, 1993). After 2 d on MS medium the leaves were dissected (0.5 cm2) and propagated in kanamycin-selective medium (selective medium containing 15 µg ml–1 kanamycin in place of spectinomycin). The first six plantlets to develop were transferred to MSspec medium (MS medium containing 0.5 mg ml–1 spectinomycin) and after 3 weeks two cmtrL lines were identified by PCR and both grown to maturity in soil under high CO2 pressures.
Four to six weeks after propagation on selective medium, the first six SpecR plants from cmtrL leaves transformed with pLEV1, pL335V, pLEVUbS, and pLEVL were transferred to MSspec medium and transplastomic lines identified using non-denaturing PAGE and DNA blots (see below). Genomic DNA was isolated from all SpecR plants that developed up to 8 weeks post-bombardment, and the transplastomic lines were identified by PCR using primers LSA (Whitney et al., 1999) and LSE (Whitney and Andrews, 2001a).
Leaf sampling
Samples from a leaf from 6–8 cm high plants growing in MSspec medium at 7 weeks post-bombardment were taken for DNA (0.5 cm2) and RNA (1 cm2) blot analyses, and the remainder (2–10 cm2) were used for PAGE, Rubisco purification, and kinetic analyses. The harvested leaf sections were rapidly frozen in liquid nitrogen and stored at –80 °C.
Protein extraction
Leaf discs were extracted in 0.5–1.0 ml of ice-cold extraction buffer [50 mM HEPES-NaOH, pH 8.0, 1 mM EDTA, 2 mM dithiothreitol (DTT), 1% (v/v) plant protease inhibitor cocktail (Sigma-Aldrich), and 1% (w/v) polyvinylpolypyrrolidone (PVPP)] in a Wheaton glass homogenizer and the lysate centrifuged (35 000 g, 10 mins, 4 °C). Aliquots of the supernatant were either assayed relative to bovine serum albumin (BSA) for protein content by dye binding (Coomassie Plus kit, Pierce), treated with an equal volume of SDS buffer [125 mM TRIS-Cl, pH 6.8, 4% (w/v) SDS, 20% (v/v) glycerol, 150 mM 2-mercaptoethanol, 0.01% (w/v) bromophenol blue], diluted 2-fold with non-denaturing PAGE buffer [30% (v/v) glycerol, 0.001% (w/v) bromophenol blue], or incubated with 10–20 mM NaHCO3 and MgCl2 at 25 °C for 10 min before incubating for 5 min with or without 100 µM [14C]carboxypentitol-P2 before chromatography though a Superdex 200HR 10/30 column (see below).
Protein expression in the XL1-Blue E. coli cells used to prepare the transforming plasmid was examined by harvesting 5 ml of the cell cultures by centrifugation (5000 g, 5 min, 20 °C), suspending the cell pellet in extraction buffer (no PVPP), then lysing by passage through a French pressure cell (140 MPa). An aliquot of lysate sample was mixed with an equal volume of SDS buffer (total cellular protein), and the remaining sample was centrifuged (38 000 g, 15 min, 4 °C) and aliquots of the supernatant assayed for soluble protein content and treated with a equal volume of SDS buffer (soluble cellular protein).
PAGE and immunoblot analysis
Protein samples were separated by SDS–PAGE (BIS-TRIS-buffered 4–12% NuPAGE gels, Invitrogen), blotted onto nitrocellulose, probed with appropriate polyclonal antiserum raised in rabbits to pure spinach Rubisco, tobacco Rubisco S subunits, or R. rubrum Rubisco, and the immunoreactive bands visualized using AttoPhos (Promega) as described previously (Whitney et al., 2001). Proteins were separated by non-denaturing PAGE at 4 °C using 4–12% TRIS-glycine gels (Invitrogen) in 60 mM TRIS, 191 mM glycine buffer at 100 V for 4 h. Protein bands were visualized using Gelcode Blue reagent (Pierce).
Rubisco content, chromatography, and kinetics
Radiolabelled [14C]carboxypentitol-P2 (an isomeric mixture of carboxyarabinitol-P2 and carboxyribitol-P2) was synthesized as described (Pierce et al., 1980) and used to measure Rubisco content following pre-activation of Rubisco with 10–200 mM each of NaHCO3 and MgCl2 as described (Ruuska et al., 1998; Whitney and Andrews, 2001a). Soluble leaf protein was chromatographed through a Superdex 200HR 10/30 column equilibrated with specificity buffer (30 mM triethanolamine pH 8.3, 30 mM Mg acetate; Kane et al., 1994) using an ÄKTA explorer system (APBiotech). Fractions (0.3 ml) containing Rubisco were identified following SDS–PAGE by immunoblot analysis using antibodies to spinach or R. rubrum Rubisco and by scintillation counting of the samples incubated with [14C]carboxypentitol-P2. The three fractions only containing form I Rubisco from non-carboxypentitol-P2-treated samples were pooled (100–150 pmol L subunit sites) and used to measure CO2/O2 specificity at 25 °C as described (Kane et al., 1994) after equilibrating with an atmosphere containing 500 ppm CO2 in O2 controlled using three Wösthoff precision gas-mixing pumps.
Rubisco from LEVUbS leaves was purified by IMAC (immobilized metal affinity chromatography). Pieces of leaf material of 2–8 cm2 were homogenized in 1–2 ml of ice-cold affinity extraction buffer [50 mM TRIS-Cl, pH 8, 0.3 M NaCl, 10 mM imidazole, 1 mM phenylmethylsulphonyl fluoride (PMSF), 5% (v/v) glycerol, 1% (w/v) PVPP], centrifuged at 35 000 g for 2 min at 2 °C, and the soluble protein mixed by rotation with 0.2 ml (bed volume) of NiNTA-agarose (Qiagen) for 8 min at 4 °C. The resin was collected into a Wizard SV-miniprep syringe-interlocking column and then washed via syringe with two 50 bed volumes of affinity extraction buffer (no PVPP or PMSF). Bound protein was eluted in 0.2 ml of elution buffer (50 mM TRIS-Cl, pH 8, 0.3 M NaCl, 200 mM imidazole). The entire process took <20 min. The H6Ub tag was specifically removed from the tagged S subunits in the IMAC-purified L8S8 by incubating with ubiquitin–protease (USP) for 30 min at 37 °C as described (Baker, 2005). Samples were taken for SDS–PAGE, non-denaturing PAGE, and immunoblot analyses, and Rubisco yield was quantified by [14C]carboxyarabinitol-P2 binding (see above).
Radiolabelled 14CO2 fixation assays were used to measure maximal carboxylase activities (Vcmax) and the Michaelis constant for CO2 at ambient pO2 (Kc21%O2) at 25 °C, pH 8.0 as described (Whitney and Sharwood, 2007). The assays were initiated by the addition of soluble leaf protein extracts from homoplasmic plants rapidly isolated in extraction buffer containing 15 mM NaHCO3 and 15 mM MgCl2 that had been incubated at 25 °C for 10 min. The assays were performed in triplicate with Vcmax calculated by dividing the ribulose-P2-dependent CO2 fixation rate in assays containing 15 mM NaH14CO3 (specific activity 2000 cpm nmol–1 CO2) by the concentration of Rubisco active sites measured by [14C]carboxyarabinitol-P2 binding. Ribulose-P2 was synthesized according to Kane et al. (1998).
Leaf DNA and RNA blot analyses
Total leaf DNA and RNA were extracted, separated by agarose gel electrophoresis, and blotted onto membranes as described previously (Whitney et al., 1999; Whitney and Andrews, 2001b). Hybridization was performed at 55 °C using buffer supplied in the AlkPhos Direct Labelling Kit (GE Healthcare). The DNA blots were probed with the 1176 bp RFR or 1856 bp rbcL1 probe (Whitney and Andrews, 2001a) and the RNA blots probed with the 991 bp rbcL2 DNA fragment amplified from pLEV1 using primers LSA and LSB (Whitney et al., 1999) or the 336 bp RbcS DNA fragment excised from pLEVUbS. The probes were labelled with [
-32P]dATP using the MegaPrime Labelling Kit (GE Healthcare), and the autoradiograph signals were detected on storage phosphor screens and imaged using a BioRad Pharos Imaging System.
|
Results and discussion |
|---|
|
TopAbstractIntroductionMaterials and methodsResults and discussionSupplementary dataReferences |
|---|
Alternative recombination events can frustrate Rubisco engineering
Genetic manipulation of higher plant plastomes, which typically target the insertion or modification of a gene of interest, is accompanied by the introduction of a selectable marker gene (smg) coding for antibiotic resistance (Maliga, 2004). Transformation occurs by flanking the heterologous sequence with an appropriate flanking plastome sequence to direct the insertion of the genetic modifications with surgical precision via homologous recombination (Svab and Maliga, 1993). Previous attempts to mutate codon 335 in the tobacco rbcL gene (substituting Leu with Val, L335V; Whitney, 19994) or to replace rbcL with the gene from sunflower (rbcLS; Kanevski et al., 1999) were hampered by the preservation of sequence homology in the transforming plasmids between the mutated sequences and the commonly used aadA smg (which confers resistance to the antibiotics streptomycin and spectinomycin; specR). Generating the L335V mutant proved particularly difficult as the intermediary region of homologous sequence was larger, with most transplastomic lines incorporating the aadA gene and not the mutation. Only one L335V line (from eight independent transformants) was shown by PCR and DNA blot analyses to incorporate the rbcL mutation. However, during subsequent growth on selective medium towards a fully segregated transformed plastome population (homoplasmicity), most of the regenerating vegetative tissue underwent secondary recombination events with non-transformed plastome copies. This second population of transformed plastomes that incorporated the aadA gene but not the rbcL mutation were exclusively selected for. This selective pressure against retaining the plastome copies containing the rbcL mutations might have resulted from the kinetic impairment to Rubisco imparted by the L335V mutation (Whitney et al., 1999).
Another unwanted recombination event also occurred when replacing rbcL with rbcLS. As both genes retain large regions of sequence homology, a transformant producing a recombinant rbcL–LS variant containing the first 233 nucleotide of rbcL was produced (Kanevski et al., 1999). In contrast, total replacement of rbcL with the R. rubrum rbcM occurred without artefact (Whitney and Andrews, 2001a) as both genes, and the intervening sequence between rbcM and aadA, shared little sequence homology. The growth rate, phenotype, and leaf development of the resultant tobacco–rubrum plants in tissue culture were comparable with those of wild-type tobacco, which it was rationalized made them highly suited for efficiently re-transforming with mutated tobacco, and even foreign, rbcL variants.
Transformation efficiency is highest using the aadA gene
The initial study objective was to re-introduce variant rbcL genes into the existing tobacco–rubrum lines using an alternative smg to the aadA already stably in place (Andrews and Whitney, 2003). Towards this goal, the transformation efficiency of two alternative smgs was compared using the versatile transforming plasmid pRV112a (Zoubenko et al., 1994). The aadA coding sequence in pRV112a was replaced with the spinach badh gene (encoding betaine aldehyde dehydrogenase) or the entire aadA cassette replaced by the Prrn-atpB 5'UTR-neo-TrbcL sequence from plasmid pNK30 (Kuroda and Maliga, 2001). Bombardment of pRV112a-badh into 20 tobacco leaf sections yielded >100 plantlets that were resistant to 10 mM betaine aldehyde, none of which were plastome transformants (data not shown). In contrast, two plastome transformants were isolated from 20 tobacco leaf sections bombarded with pRV112a-neo after selecting on medium containing 50–100 µg ml–1 kanamycin, while eight specR plastome transformants were obtained following transformation of 10 leaves with pRV112A (data not shown). Clearly transformation efficiency was highest using spectinomycin selection, consistent with its preferential use in plastome transformation research (Maliga, 2004; Koop et al., 2008).
Construction of marker-less cmtrL tobacco–rubrum lines
As proposed in Andrews and Whitney (2003), a two-step approach was undertaken to engineer a photosynthetic tobacco master line that lacked sequence homology to form I Rubisco genes and enabled selection of transformants using the specR-coding aadA gene. To achieve this, new tobacco–rubrum plants were first generated using aadA selection, and then aadA was excised to allow its re-use. To complete the first step, four new cmtrLA tobacco–rubrum lines were selected (from five bombardments) where the tobacco rbcL coding sequence (excluding the first 42 nucleotides; Whitney and Andrews, 2001a) and 269 bp of its 3'-untranslated region (UTR) were replaced by 2960 bp of sequence containing a codon-modified rbcM gene (cmrbcM, Supplementary Fig. S1 at JXB online), 222 bp of the tobacco psbA 3'-UTR sequence, and an inversely oriented Prrn–aadA–Trps16 cassette that was flanked by directly oriented 34 bp loxP sites (Fig. 1A). The codon bias of the synthetic cmrbcM matched that of rbcL but maintained the same low level of sequence homology shown between rbcM and rbcL (Supplementary Fig. S1 at JXB online).
|
The second step of excising aadA from a cmtrLA line was achieved by the P1 bacteriophage site-specific CRE recombinase using a novel single-step protocol. Two aadA-free (marker-less) cmtrL lines were isolated following transient expression of CRE in cmtrLA leaves after biolistically introducing the T-DNA plasmid, pKO27, that encodes the cre gene fused in-frame with the 5' sequence coding for a pea Rubisco S subunit plastid-targeting sequence with expression regulated by the P2' Agrobacterium promoter (Corneille et al., 2001). Stable or transient introduction of pKO27, and other CRE-expressing T-DNA variants, by agroinfiltration has been highly successful in yielding marker-free transplastomic lines (Corneille et al., 2001; Hajdukiewicz et al., 2001; Lutz et al., 2006). Notably, agroinfiltration can result in both stable transformation with the T-DNA and transient expression of cre. Plants in which CRE was only transiently produced have to be identified and if cre has integrated into the nucleus the gene has to be segregated away in the seed progeny. This complication was avoided by using a simpler approach that biolistically introduced pKO27 into three sterile leaves of a homoplasmic T0 cmtrLA plant (line 1) that had been grown in MS medium following two rounds of regeneration in selective medium. As plasmid DNA can persist for >2 weeks in bombarded leaf tissue (Ye et al., 2003), the shot leaves were regenerated on selective medium containing 15 µg ml–1 kanamycin to provide some selection pressure for growth of cells containing pKO27. Plantlets emerged after 4 weeks from the segmented shot leaves that had become bleached. The first six sufficiently developed plantlets were transferred to growth in 0.75 l pots on MS medium. After 3 weeks, the plants were 4–7 cm in height, had developed roots, and were phenotypically indistinguishable from cmtrLA lines being grown under identical conditions for transfer to growth in soil. DNA from the youngest leaf was extracted, and two marker-less cmtrL lines were identified using PCR (Fig. 1B), with DNA blots indicating that both lines were already homoplasmic with no cmtrLA (or wild-type tobacco) plastome copies detected (Fig. 1C). The homoplasmic cmtrL1 and cmtrL2 lines were transferred to soil, grown to maturity under high CO2 pressures, and their flowers pollinated with wild-type pollen. All the cmtrL1 and cmtrL2 T1 progeny (>103 T0 seed from each line screened) were sensitive to growth on sucrose-free MS medium containing spectinomycin (0.5 mg ml–1) or kanamycin (0.1 mg ml–1), indicating that neither line carried a plastid aadA gene and were unlikely to contain stably incorporated nuclear neo copies and the T-DNA-associated cre gene. To confirm this, PCR analyses using leaf DNA from 10 T1 progeny from each cmtrL line and primers specific for cre were unable to amplify a product, indicating that both were cre free (data not shown).
Testing the ability to transform rbcL variants into cmtrL leaves
Transforming plasmids based on pLEV1 (Whitney et al., 1999; Whitney and Andrews, 2003) were used to re-introduce either the native tobacco rbcL or the L335V-coding mutated rbcL gene (pL335V), or to co-introduce rbcL with a tobacco RbcS gene that contained a sequence coding for an N-terminal 6xhistidine-tagged ubiquitin fusion (pLEVUbS, Fig. 2A). As shown previously, proteins whose genes are regulated by the tobacco rbcL promoter and 5'-UTR are constitutively expressed in E. coli, which can lead to the insertion of transposons in the 5'-UTR or coding sequence if the recombinant protein is detrimental to bacterial growth (Whitney and Andrews, 2001a). Therefore, in addition to sequencing, the integrity of the transforming plasmids shot into tobacco was examined by SDS–PAGE and immunoblot analyses to confirm expression of the Rubisco subunits (Fig. 2B). Consistent with that shown previously (Whitney and Sharwood, 2007), large amounts of the tobacco L subunits were produced in the E. coli transformed with pLEV1, pL335V, and pLEVUbS, and these were almost entirely insoluble as the folding and assembly requirements of higher plant Rubiscos are not met by the E. coli chaperone complexes (Gatenby et al., 1987). The small amount of soluble L detected appeared to be complexed with chaperonin complexes when separated by non-denaturing PAGE (data not shown). Approximately 10% of the ubiquitin–S subunit fusion produced was soluble, as seen previously for the native tobacco S subunit expressed in E. coli (Whitney and Andrews, 2001b; Whitney and Sharwood, 2007).
|
As transcription of the nuclear RbcS genes is heightened in tobacco–rubrum but the unassembled S subunit is degraded (Whitney and Andrews, 2003), it was reasoned that successful re-introduction of their cognate L subunits would readily facilitate production of form I hexadecamers (L8S8). As L8S8 Rubiscos are both structurally and antigenically distinct from the R. rubrum L2 enzyme, the feasibility of detecting, selectively purifying, and measuring the CO2/O2 specificity of the form I enzyme from heteroplasmic leaf tissue as soon as possible following biolistic transformation was examined. As summarized in Fig. 3A, specR plantlets began to develop 4–6 weeks after transformation, of which the first six were transferred directly into MSspec medium to promote normal vegetative development (Fig. 3B). After 2–3 weeks, the plants were 4–8 cm in height and had developed roots. Soluble protein from a leaf was analysed by non-denaturing PAGE for production of the R. rubrum L2 enzyme alone (non-transformed), both L2 and L8S8 Rubisco (heteroplasmic transformant), or just the L8S8 enzyme (homoplasmic transformant) (Fig. 3C). Transformants and the extent of their segregation were subsequently confirmed by DNA blot analysis of total genomic DNA isolated from the same leaf (Fig. 3D). Curiously the L8S8 enzyme in the LEVUbS lines appeared to resolve as normal-sized and larger oligomeric complexes, suggesting that the plastid-synthesized H6UbS subunits had successfully assembled into some of the hexadecamers (see below).
|
Blots of total RNA extracted from the same leaf were probed with rbcL DNA and showed that the LEV1 and L335V lines produced an abundant monocistronic (rbcL) transcript and an
80% less abundant bicistronic (rbcL–aadA) transcript as shown previously for other LEV1 transformants (Whitney and Andrews, 2003) (Fig. 3E). As the aadA gene is promoter-less and does not produce a monocistronic mRNA, production of sufficient rbcL–aadA transcript is requisite for the specR phenotype. In the LEVUbS transformants, two rbcL transcripts were made, an abundant bicistronic rbcL–UbRbcS mRNA that exceeded the amount of nuclear RbcS mRNA by >6-fold, and a 70–80% less abundant tricistronic rbcL–UbrbcS–aadA mRNA whose translation is necessary for spectinomycin selection (Fig. 3E). From the non-denaturing PAGE (Fig. 3C) and DNA blots (Fig. 3D), it was evident that five LEV1, three L335V, and five LEVUbS of the six specR plants analysed were transplastomic lines, with all except the LEV1#2 and L335V#3 lines already being homoplasmic. As the plants already had well-developed roots, they were immediately transferred to growth in soil without the need for repeated propagation on selective medium, thus reducing the time required to obtain mature transplastomic plants. Twelve weeks after beginning the transformation process, all three L335V and three of the LEVUbS transformants had been grown to maturity in soil and were beginning to set seed. This contrasts with the 8 months that was required previously to propagate the single L335V transformant to the same level of development (Whitney et al., 1999).
The transformation efficiency of cmtrL leaves is high
The production of transplastmic lines was more efficient using the leaves of cmtrL plants than wild-type tobacco leaves. To examine the transformation efficiency of cmtrL, the genomic DNA from all the specR plants that developed up to 8 weeks post-plasmid bombardment was analysed by PCR using one primer specific to rbcL (LSA) and one that anneals downstream of the aadA insertion site (LSE, Fig. 2A). In bona fide LEV1 and L335V transformants, this produced a 2.5 kb amplification product and a 3.2 kb product in transplastomic LEVUbS lines (data not shown). Comparable high transformation efficiencies (2–3 transformants per bombardment) were found in generating LEV1, L335V, and pLEVUbS transformants using cmtrL leaves (Table 1). Notably, the transforming plasmid had been prepared using a commercial quick spin mini-prep kit from 1 ml of the E. coli culture used for SDS–PAGE and immunoblot analysis (Fig. 2B). This contrasts with previous efforts using highly purified transforming plasmid prepared using a polyethylene glycol (PEG) extraction protocol (Whitney et al., 1999) which afforded transformation efficiencies 3- to 10-fold lower using wild-type tobacco leaves than those obtained using cmtrL1 (Table 1).
|
Transplastomic Rubisco null tobacco are still produced at a high transformation frequency
The high transformation efficiency in cmtrL leaves was not pervasively influenced by the re-introduction of functional L8S8 enzyme. A frameshift mutation was introduced into the rbcL coding region at the SacII site to give plasmid pLEV
L (Fig. 2A). The mutation introduced a stop codon after 57 codons, shortening the rbcL open reading frame to a 6 kDa L subunit whose production in transformed pLEVL-XL1-Blue-E. coli cells was not detected (Fig. 2B). Comparable with what was seen previously in non-photosynthetic Rubisco null tobacco (Kanevski and Maliga, 1994), the leaves of the transplastomic LEVL lines were chlorotic and highly susceptible to structural damage when handled (Fig. 3B). Despite there being no selective advantage in eliminating the R. rubrum Rubisco, all six of the original specR LEVL lines screened were transformants that, as expected, did not produce L8S8 Rubisco (Fig. 3C), and five lines were already homoplasmic (Fig. 3D). Like LEV1 and L335V, the LEVL lines produced comparable relative levels of rbcL and rbcL–aadA mRNAs (Fig. 3E) and analogous high transformation efficiencies (Table 1).
Rubisco purification
The variation in the size of R. rubrum L2 Rubisco (100 kDa) and L8S8 Rubisco (520 kDa) enabled purification of the latter by size exclusion chromatography. The capacity of a Superdex 200HR 10/30 size exclusion column to resolve L8S8 and L2 Rubisco was examined by incubating CO2-Mg2+-activated soluble leaf protein from heteroplamic lines producing both Rubiscos with the tight binding inhibitor [14C]carboxyarabinitol-P2. Both scintillation counting and immunoblot analyses showed the L8S8 and L2 Rubisco peaks were not totally resolved to baseline (Fig. 4A). This was in contrast to previous findings (S Whitney and H Ashida, unpublished data), and subsequent separation of protein standards indicated that the resolving capacity of the ageing Superdex 200HR 10/30 column was suboptimal. Nevertheless, fractions containing only the L8S8 Rubisco were obtainable (shaded region in Fig. 4A), which were pooled and used to measure Sc/o (see below).
|
The capacity of plastid-synthesized H6UbS subunits to assemble in the variant L8S8 enzymes produced in LEVUbS (Fig. 3B, see above) was examined using IMAC. A H6Ub ubiquitin fusion was chosen because (i) it can be specifically removed by Ub protease and not leave any residual N-terminal sequence on the fused target protein (Baker, 1996; Baker et al., 2005); (ii) chloroplasts do not appear to accumulate Ub proteases (Vierstra, 1996); and (iii) the appended Ub sequence can increase the stability of protein fusions due to the chaperonin-like effect of Ub on protein folding (Baker, 1996; Staub et al., 2000). SDS–PAGE and immunoblot analysis of the LEVUbS soluble leaf protein confirmed that small amounts of H6UbS subunits were produced and showed that tobacco L and S subunits were selectively purified with H6UbS by IMAC (Fig. 4B). On non-denaturing PAGE the ‘IMAC-pure’ protein (not digested with Ub protease, UbP) resolved as an enzyme complex larger than L8S8, suggesting that it incorporated one or more H6UbS subunits. Once the H6Ub sequence was removed by UbP treatment, the mobility of the ‘pure’ enzyme through the gel resembled that of the wild-type L8S8 enzyme (Fig. 4B).
While the H6Ub fusion strategy was successful in specifically isolating L8S8 tobacco Rubisco, its versatility was compromised by the low yield of hexadecamers that incorporated a plastid-synthesized H6UbS subunit. From [14C]carboxyarabinitol-P2 binding analyses it was determined that 1.5% of the total Rubisco (5 pmol active sites) extracted from 2 cm2 of a LEVUbS leaf was recovered by rapid IMAC purification (total processing time <15 min). The poor incorporation of plastid-synthesized H6UbS subunits into hexadecamers (despite the abundant UbRbcS mRNA pool size; Fig. 3E) correlates with that observed previously where similar levels (1.2% of the assembled S subunit pool) of plastid-synthesized, C-terminal histidine-tagged S subunits (SH7) were assembled into L8S8 enzyme even though the plastid pool of RbcS mRNA exceeded the pool of endogenous nuclear RbcS message >10-fold (Whitney and Andrews, 2001b). Even in anti-RbcS tobacco, when the pool of cytosolic S subunits was reduced >80%, the production of plastid-synthesized S subunits had no influence on Rubisco levels (Zhang et al., 2002) until the RbcS gene was incorporated within the trnI–trnA intergenic region of the 16S rDNA operon within the tobacco plastome inverted repeat region which is ascribed as being a transcriptional hot spot (Dhingra et al., 2004). Whether improvements in incorporation of H6UbS subunits into Rubisco hexadecamers can be attained by transforming its gene into the trnI–trnA intergenic region, or by reducing RbcS levels in the cmtrL line by antisense or RNAi techniques, remains to be examined.
The yield of Rubisco isolated from LEVUbS leaves by IMAC appeared compromised by cleavage of the H6Ub fusion during purification. On immunoblots of the IMAC-purified Rubisco, the proportion of mature S subunits exceeded the H6UbS signal >20-fold (Fig. 4B). This contrasts with previous IMAC-purified L8S8 complexes from the transplastomic lines producing plastid-synthesized SH7 subunits where the cytosolic-synthesized S:SH7 ratio was 7:1 (Whitney and Andrews, 2001b). As wild-type tobacco Rubisco (or any other soluble leaf protein) does not bind to IMAC (Whitney and Andrews, 2001b), the higher S:H6UbS stoichiometry in the IMAC-purified Rubisco from LEVUbS leaves indicates cleavage of the H6Ub sequence during the purification process. Notably, similar processing of the H6Ub sequence has been observed during the purification of recombinant H6Ub fusion proteins from E. coli (Baker et al., 2005) and tobacco chloroplasts (Staub et al., 2000). Clearly this premature processing of H6UbS coupled with the apparent preferential assembly of cytosolically synthesized S subunits (or poor translation of H6UbS) reduces the effectiveness of the H6UbS fusion strategy for purifying tobacco Rubisco from small amounts of leaf tissue and questions the merit of its use in future Rubisco transplantation studies.
Kinetic screening
Leaf material from LEV1 and L335V plants growing in MSspec medium (Fig. 3C) was used to measure quickly the kinetic parameters that are used to model the ribulose-P2-saturated CO2 assimilation rate of C3 photosynthesis (Farquhar et al., 1980; Andrews and Whitney, 2003; Parry et al., 2007). CO2/O2 specificity (Sc/o) was measured using Superdex 200HR 10/30-purified L8S8 Rubisco from heteroplasmic L335V (line 3) and LEV1 (line 2) leaf tissue (Fig. 4A). The measured Sc/o values matched that previously measured using highly purified enzyme from homoplasmic lines (Table 2). Likewise, the maximal carboxylation rates (Vcmax) and Michaelis constants for CO2 in the presence of air levels of 20.6% (v/v) O2 (Kc21%O2) measured using freshly isolated soluble leaf protein from homoplasmic lines (L335V line 4, LEV1 line 5) also matched those measured previously (Table 2; Whitney et al., 1999).
|
Features of cmtrL that make it a good recipient for form I Rubisco transformations
While comprehensively assessing the versatility of cmtrL lines for transplanting non-tobacco Rubiscos into tobacco chloroplasts remains to be examined there is optimism that the high transformation efficiency of cmtrL and the unique structure and antigenic specificity of the R. rubrum enzyme make it a more suitable recipient than wild-type tobacco for screening the production, folding, and assembly capabilities of both foreign and mutated tobacco form I Rubiscos. The ease with which numerous L335V transformants were readily obtained using cmtrL leaves contrasted with previous attempts where significant experimental effort was required to obtain one transformant (Whitney et al., 1999). Evidently the ease of obtaining the L335V, LEV1, LEVUbS, and even LEV
L transformants correlates with the poor sequence similarity between the cmrbcM gene in cmtrL and the introduced rbcL (and H6Ub-RbcS) sequences (Supplementary Fig. S1 at JXB online). The quick segregation of transformed plastome populations enabled prompt analysis of the L8S8 Rubisco kinetics. In most cases the transformants were already homoplasmic 6–9 weeks after bombarding the cmtrL leaves with transforming plasmids, which contrasts with previous efforts to generate homoplasmic LEV1 lines from transformed wild-type tobacco leaves that necessitated multiple rounds of regeneration on selective media (Whitney et al., 1999). Once homoplasmic, reliable, and reproducible measurements of Vcmax and Kc21%O2 can be made using leaf protein extract (Kubien et al., 2003; Ghannoum et al., 2005; Sharwood et al., 2008), while Sc/o determination using the method of Kane et al. (1994) requires at least partial purification (Whitney and Andrews, 1998) which, as shown here, can be sufficiently achieved by Superdex 200HR 10/30 chromatography (Table 2).
While the re-introduced tobacco L8S8 Rubiscos examined in this study were abundantly expressed, producing foreign L8S8 Rubisco in tobacco plastids may be more problematic. As evidenced in previous Rubisco transplantation studies, reductions in transcript abundance (Whitney and Andrews, 2001a), perturbations to translational processing (Whitney and Andrews, 2003), and problems with the folding requirements of introduced L subunits and their assembly with tobacco S subunits (Kanevski et al., 1999; Sharwood et al., 2008) significantly constrain the production of foreign Rubiscos in tobacco plastids, sometimes even precluding assembly of functional enzyme (Whitney et al., 2001). The high transformation efficiency and prompt isolation of Rubisco null LEVL lines suggest producing transformants where the introduced foreign Rubisco subunits do not, or poorly, assemble into functional Rubisco should not be problematic. Indeed, an indication of assembly problems should be evident by the production of pale green leaves during growth in MSspec medium (Fig. 3B). The detection of foreign Rubisco subunits is also simplified by the antigenic distinctiveness of the R. rubrum Rubisco relative to form I Rubiscos which, as shown previously (Whitney et al., 2001), can enable the level of assembled and unassembled introduced Rubisco subunits to be examined rapidly using immunoblot analyses. As shown in Fig. 3C, the assembly of L8S8 complexes can also be quickly ascertained using non-denaturing PAGE which additionally provides a means to assess whether the plants are homoplasmic according to whether or not the R. rubrum L2 enzyme is present.
Alternative screening strategies to consider
Like cmtrL, transplantating foreign or modified Rubiscos into higher plant plastids may be equally successful using a marker-less Rubisco null (rbcL) line as a transformation recipient. Similar to the LEVL lines produced here, previous Rubisco null tobacco produce bleached leaves that required grafting onto wild-type stock to produce seed (Kanevski and Maliga, 1994). While the production of a marker-less rbcL line has not been reported, it would, in theory, make a suitable candidate for transforming in mutated or foreign Rubisco genes. Generating an rbcL knockout using an alternative selectable marker gene, such as neo, would enable subsequent transformations using aadA selection without needing to remove neo. It might also be possible to omit antibiotic selection altogether with a rbcL line and use the restoration of green pigmentation as a phenotypic indicator of functional Rubisco production. Unlike the cmtrL leaves that are photosynthetic and have a wild-type-like phenotype that makes them amenable to biolistic transformation, the sensitivity of rbcL leaves to permanent tissue damage following even minor handling (as evidenced with LEVL leaves) may prevent their plastomes from being transformed by this procedure. This potential limitation might be circumvented by generating rbcL protoplast suspensions and transforming their plastomes by PEG treatment (Koop et al., 2008). An alternative strategy is to transform wild-type tobacco leaves and introduce sufficient silent nucleotide mutations into the candidate Rubisco rbcL to prevent unwanted recombination events occurring with tobacco rbcL. Notably the cost involved in making multiple changes or re-synthesizing an entire gene may curb the general applicability of this approach.
Future Rubisco transplantation applications
Defining the cmtrL plants as ‘master lines’ stems from the simplicity with which transplastomic tobacco expressing variant rbcL genes can be generated using cmtrL which provides unique opportunities to examine more readily the structure–function relationships of tobacco Rubisco. The capacity now exists to scrutinize the importance and function of the L subunit post-translational modifications (Houtz and Portis, 2003) and manipulate those residues that are thought to regulate strongly, and specify, the interactivity of Rubisco with its helper protein, Rubisco activase (Larson et al., 1997; Ott et al., 2000). The cmtrL lines should simplify efforts to examine whether the ability of catalytically normal cyanobacterial Rubisco complexes comprised of S subunit–L subunit (SL) fusions (Whitney and Sharwood, 2007) can be emulated in tobacco chloroplasts using tobacco SL fusion peptides. Mimicking many of the practical mutations already examined in form I Rubiscos from cyanobacteria and Chlamydomonas (Kellogg and Juliano, 1997; Parry et al., 2003) should also benefit the mechanistic understanding of higher plant Rubisco catalysis. Indeed, replicating the cyanobacterial L335V mutation (Leu332 in the Synechococcus PCC6301 L-subunit; Lee et al., 1993) in tobacco Rubisco unveiled novel aspects of the higher plant catalytic mechanism (Pearce and Andrews, 2003). The large database of available Rubisco sequences (>3200; Kapralov and Filatov, 2007) and the resolution of >20 high resolution Rubisco crystal structures (Andersson and Taylor, 2003) should greatly assist in identifying candidate residues for mutation, though undoubtedly such an undertaking would greatly benefit from more comprehensive kinetic surveys of sequenced Rubiscos.
The cmtrL plants also provide a means to improve our understanding of rbcL regulation and L subunit production. For plastid genes, protein expression is primarily regulated post-transcriptionally (Somanchi and Mayfield, 1999) and involves features in the 5'- and 3'-UTRs of the mRNA which are requisite for the binding of auxillary trans-acting factors (many nucleus encoded) that directly or indirectly influence transcript stability; mRNA processing, mRNA editing; and translation of the mRNA (Monde et al., 2000; Nickelsen, 2003; Bollenbach et al., 2004). By transforming into cmtrL it should now be possible to augment the previous study of Shiina et al. (1998) and use Rubisco as the reporter protein to pinpoint sequence elements in the rbcL regulatory sequences that are pertinent for transcription and translation, and possibly identify auxillary factors that are involved in regulating these processes.
As highlighted with the LEVUbS transformants, an appropriate means for the engineering changes to the native (or foreign) S subunit in tobacco remains an elusive challenge. The pervasive influence of the S subunit on the kinetics of form I Rubisco suggests engineering kinetic improvements may necessitate complementary engineering of both the L and S subunits (Spreitzer, 2003). The paucity of plastid-synthesized S subunits incorporated into Rubisco hexadecamers may arise from problems with their translation in the stroma or result from limitations in their post-translational folding and assembly with the L subunits that hinder them in effectively competing with the cytosolically synthesized S subunits for assembly (Whitney and Andrews, 2001b; Andrews and Whitney, 2003). Clearly a concerted effort is required to address whether this limitation to Rubisco engineering in tobacco can be surmounted by reducing/eliminating the nuclear RbcS mRNA levels in cmtrL by antisense/RNAi techniques and/or modifying the transformation strategy and introducing the tailored rbcL and rbcS genes either together or separately elsewhere in the plastome, such as the reported transcriptional hot spot in the trnI–trnA intergenic region (Dhingra et al., 2004).
Examining issues related to assembly of foreign Rubiscos in higher plant plastids is likely to be simplified using a cmtrL master line. The inability of tobacco plastids to meet the folding and assembly requirements of the ‘red’-type Rubisco subunits from the non-green algae Phaeodactylum tricornutum (diatom) and Galdieria sulphuraia (red alga) has deterred the pursuit of such endeavours with other ‘red’-type Rubiscos. However, the recent finding that a single amino acid substitution improved the assembly of Synechococcus PCC6301 Rubisco in E. coli >5-fold (Greene et al., 2007) suggests there is merit to exploiting the high transformation and L8S8 screening capabilities of cmtrL to survey more thoroughly the capabilities of other ‘red’-type Rubisco subunits to assemble functionally in tobacco chloroplasts and examine whether or not the tobacco S subunits are excluded from assembly. Broadening this survey to include replacement of cmrbcM with other higher plant rbcL genes without, and in conjunction with, one of their cognate RbcS gene copies is also paramount for, first, ascertaining to what extent the sustained kinetics and stability of the hybrid Rubisco comprising sunflower L and tobacco S subunits (Sharwood et al., 2008) are emulated in hybrid enzymes comprising other higher plant L subunits and, secondly, to examine the capacity of the complementary S subunit to outcompete the tobacco S subunit for assembly. Of further benefit to performing the studies in tobacco plastids is the capacity to examine in vivo using light transient gas exchange measurements (Hammond et al., 1998) the capacity of (or requirement for) the tobacco Rubisco activase to regulate the transplanted Rubisco.
|
Supplementary data |
|---|
|
TopAbstractIntroductionMaterials and methodsResults and discussionSupplementary dataReferences |
|---|
Supplementary data mentioned in this article are available at JXB online.
|
Acknowledgements |
|---|
This research was supported by an Australian Research Council Discovery grant (DP0450564) awarded to SW. We thank Pal Maliga for supplying the pKO27 and pNK30 plasmids and thank Jeff Wilson for his photography.
|
Abbreviations |
|---|
carboxyarabinitol-P2, carboxyarabinitol-1,5-bisphoshate; cmrbcM, artificial codon-modified gene coding for Rubisco from R. rubrum; cmtrL, marker-free transplastomic tobacco–rubrum plants; IMAC, immobilized metal affinity chromatography; ribulose-P2, ribulose-1,5-bisphoshate; ribulose-P2, carboxylase/oxygenase; UTR, untranslated region.
|
References |
|---|
|
TopAbstractIntroductionMaterials and methodsResults and discussionSupplementary dataReferences |
|---|
Andersson I, Taylor TC. Structural framework for catalysis and regulation in ribulose-1,5-bisphosphate carboxylase/oxygenase. Archives of Biochemistry and Biophysics (2003) 414:130–140.[CrossRef][Web of Science][Medline]
Andrews TJ, Whitney SM. Manipulating ribulose bisphosphate carboxylase/oxygenase in the chloroplasts of higher plants. Archives of Biochemistry and Biophysics (2003) 414:159–169.[Web of Science][Medline]
Baker RT. Protein expression using ubiquitin fusion and cleavage. Current Opinion in Biotechnology (1996) 7:541–546.[CrossRef][Web of Science][Medline]
Baker RT, Catanzariti AM, Karunasekara Y, Soboleva TA, Sharwood R, Whitney S, Board PG. Using deubiquitylating enzymes as research tools. In: Ubiquitin and protein degradation, Part A 398—Deshaies RJ, ed. (2005) San Diego, CA: Elsevier Academic Press. 540–554.
Bollenbach TJ, Schuster G, Stern DB. Cooperation of endo- and exoribonucleases in chloroplast mRNA turnover. Progress in Nucleic Acid Research and Molecular Biology (2004) 78:305–337.[Web of Science][Medline]
Cloney LP, Bekkaoui DR, Hemmingson SM. Co-expression of plastid chaperonin genes and a synthetic plant Rubisco operon in Escherichia coli. Plant Molecular Biology (1993) 23:1285–1290.[CrossRef][Web of Science][Medline]
Corneille S, Lutz K, Svab Z, Maliga P. Efficient elimination of selectable marker genes from the plastid genome by the CRE–lox site-specific recombination system. The Plant Journal (2001) 27:171–178.[CrossRef][Web of Science][Medline]
Dhingra A, Portis AR, Daniell H. Enhanced translation of a chloroplast-expressed RbcS gene restores the subunit levels and photosynthesis in nuclear RbcS antisense plants. Proceedings of the National Academy of Sciences, USA (2004) 101:6315–6320.
Farquhar GD, von Caemmerer S, Berry JA. A biochemical model of photosynthetic CO2 assimilation in leaves of C3 species. Planta (1980) 149:78–90.[CrossRef][Web of Science]
Gatenby AA, van der Vies SM, Rothstein SJ. Co-expression of both the maize large and wheat small subunit genes of ribulose-bisphosphate carboxylase in Escherichia coli. European Journal of Biochemistry (1987) 168:227–231.[Web of Science][Medline]
Ghannoum O, Evans JR, Chow WS, Andrews TJ, Conroy JP, von Caemmerer S. Faster rubisco is the key to superior nitrogen-use efficiency in NADP-malic enzyme relative to NAD-malic enzyme C-4 grasses. Plant Physiology (2005) 137:638–650.
Greene DN, Whitney SM, Matsumura I. Artificially evolved Synechococcus PCC6301 Rubisco variants exhibit improvements in folding and catalytic efficiency. Biochemical Journal (2007) 404:517–524.[CrossRef][Web of Science][Medline]
Hajdukiewicz PTJ, Gilbertson L, Staub JM. Multiple pathways for Cre/lox-mediated recombination in plastids. The Plant Journal (2001) 27:161–170.[CrossRef][Web of Science][Medline]
Hammond ET, Andrews TJ, Mott KA, Woodrow IE. Regulation of Rubisco activation in antisense plants of tobacco containing reduced levels of Rubisco activase. The Plant Journal (1998) 14:101–110.[CrossRef][Web of Science][Medline]
Houtz RL, Portis AR Jr. The life of ribulose 1,5-bisphosphate carboxylase/oxygenase—posttranslational facts and mysteries. Archives of Biochemistry and Biophysics (2003) 414:150–158.[Web of Science][Medline]
Kane HJ, Viil J, Entsch B, Paul K, Morell MK, Andrews TJ. An improved method for measuring the CO2/O2 specificity of ribulosebisphosphate carboxylase-oxygenase. Australian Journal of Plant Physiology (1994) 21:449–461.[Web of Science]
Kane HJ, Wilkin JM, Portis AR, Andrews TJ. Potent inhibition of ribulose-bisphosphate carboxylase by an oxidized impurity in ribulose-1,5-bisphosphate. Plant Physiology (1998) 117:1059–1069.
Kanevski I, Maliga P. Relocation of the plastid rbcL gene to the nucleus yields functional ribulose-1,5-bisphosphate carboxylase in tobacco chloroplasts. Proceedings of the National Academy of Sciences, USA (1994) 91:1969–1973.
Kanevski I, Maliga P, Rhoades DF, Gutteridge S. Plastome engineering of ribulose-1,5-bisphosphate carboxylase/oxygenase in tobacco to form a sunflower large subunit and tobacco small subunit hybrid. Plant Physiology (1999) 119:133–141.
Kapralov MV, Filatov DA. Widespread positive selection in the photosynthetic Rubisco enzyme—art. no. 73. BMC Evolutionary Biology (2007) 7:73.[CrossRef][Medline]
Kellogg EA, Juliano ND. The structure and function of RuBisCO and their implications for systematic studies. American Journal of Botany (1997) 84:413–428.[Abstract]
Koop H-U, Herz S, Golds TJ, Nickelsen J. The genetic transformation of plastids. Topics in Current Genetics (2008) 19:457–510.
Kubien DS, von Caemmerer S, Furbank RT, Sage RF. C4 photosynthesis at low temperature. A study using transgenic plants with reduced amounts of Rubisco. Plant Physiology (2003) 132:1577–1585.
Kuroda H, Maliga P. Sequences downstream of the translation initiation codon are important determinants of translation efficiency in chloroplasts. Plant Physiology (2001) 125:430–436.
Larson EM, O'Brien CM, Zhu GH, Spreitzer RJ, Portis AR Jr. Specificity for activase is changed by a Pro-89 to Arg substitution in the large subunit of ribulose-1,5-bisphosphate carboxylase/oxygenase. Journal of Biological Chemistry (1997) 272:17033–17037.
Lee EH, Harpel MR, Chen Y-R, Hartman FC. Perturbation of reaction-intermediate partitioning by a site-directed mutant of ribulose-bisphosphate carboxylase/oxygenase. Journal of Biological Chemistry (1993) 268:26583–26591.
Lutz KA, Bosacchi MH, Maliga P. Plastid marker-gene excision by transiently expressed CRE recombinase. The Plant Journal (2006) 45:447–456.[CrossRef][Web of Science][Medline]
Maliga P. Plastid transformation in higher plants. Annual Review of Plant Biology (2004) 55:289–313.[CrossRef][Medline]
Monde RA, Greene JC, Stern DB. The sequence and secondary structure of the 3'-UTR affect 3'-end maturation, RNA accumulation, and translation in tobacco chloroplasts. Plant Molecular Biology (2000) 44:529–542.[CrossRef][Web of Science][Medline]
Nickelsen J. Chloroplast RNA-binding proteins. Current Genetics (2003) 43:392–399.[CrossRef][Web of Science][Medline]
Ott CM, Smith BD, Portis AR, Spreitzer RJ. Activase region on chloroplast ribulose-1,5-bisphosphate carboxylase/oxygenase—nonconservative substitution in the large subunit alters species specificity of protein interaction. Journal of Biological Chemistry (2000) 275:26241–26244.
Parry MAJ, Andralojc PJ, Mitchell RAC, Madgwick PJ, Keys AJ. Manipulation of Rubisco: the amount, activity, function and regulation. Journal of Experimental Botany (2003) 54:1321–1333.
Parry MAJ, Madgwick PJ, Carvalho JFC, Andralojc PJ. Prospects from increasing photosynthesis by overcoming the limitations of Rubisco. Journal of Agricultural Science (2007) 145:31–43.[CrossRef][Web of Science]
Pearce FG, Andrews TJ. The relationship between side reactions and slow inhibition of ribulose-bisphosphate carboxylase revealed by a loop 6 mutant of the tobacco enzyme. Journal of Biological Chemistry (2003) 278:32526–32536.
Pierce J, Tolbert NE, Barker R. Interaction of ribulosebisphosphate carboxylase/oxygenase with transition-state analogues. Biochemistry (1980) 19:934–942.[CrossRef][Web of Science][Medline]
Portis AR, Parry MAJ. Discoveries in Rubisco (ribulose 1,5-bisphosphate carboxylase/oxygenase): a historical perspective. Photosynthesis Research (2007) 94:121–143.[CrossRef][Web of Science][Medline]
Raines CA. Transgenic approaches to manipulate the environmental responses of the C-3 carbon fixation cycle. Plant, Cell and Environment (2006) 29:331–339.[CrossRef][Medline]
Ruuska S, Andrews TJ, Badger MR, Hudson GS, Laisk A, Price GD, von Caemmerer S. The interplay between limiting processes in C-3 photosynthesis studied by rapid-response gas exchange using transgenic tobacco impaired in photosynthesis. Australian Journal of Plant Physiology (1998) 25:859–870.[Web of Science]
Sharwood RE, von Caemmerer S, Maliga P, Whitney SM. The catalytic properties of hybrid Rubisco comprising tobacco small and sunflower large subunits mirror the kinetically equivalent source Rubiscos and can support tobacco growth. Plant Physiology (2008) in press. Epub ahead of print.
Shiina T, Allison L, Maliga P. rbcL Transcript levels in tobacco plastids are independent of light: reduced dark transcription rate is compensated by increased mRNA stability. The Plant Cell (1998) 10:1713–1722.
Shinozaki K, Ohme M, Tanaka M, et al. The complete nucleotide sequence of the tobacco chloroplast genome: its gene organization and expression. EMBO Journal (1986) 5:2043–2049.[Web of Science][Medline]
Somanchi A, Mayfield SP. Nuclear–chloroplast signalling. Current Opinion in Plant Biology (1999) 2:404–409.[CrossRef][Web of Science][Medline]
Spreitzer RJ. Role of the small subunit in ribulose-1,5-bisphosphate carboxylase/oxygenase. Archives of Biochemistry and Biophysics (2003) 414:141–149.[CrossRef][Web of Science][Medline]
Staub JM, Garcia B, Graves J, et al. High-yield production of a human therapeutic protein in tobacco chloroplasts. Nature Biotechnology (2000) 18:333–338.[CrossRef][Web of Science][Medline]
Svab Z, Maliga P. High-frequency plastid transformation in tobacco by selection for a chimeric aadA gene. Proceedings of the National Academy of Sciences, USA (1993) 90:913–917.
Tcherkez GGB, Farquhar GD, Andrews TJ. Despite slow catalysis and confused substrate specificity, all ribulose bisphosphate carboxylases may be nearly perfectly optimized. Proceedings of the National Academy of Sciences, USA (2006) 103:7246–7251.
Vierstra RD. Proteolysis in plants—mechanisms and functions. Plant Molecular Biology (1996) 32:275–302.[CrossRef][Web of Science][Medline]
Whitney SM, Andrews TJ. The CO2/O2 specificity of single-subunit ribulose-bisphosphate carboxylase from the dinoflagellate, Amphidinium carterae. Australian Journal of Plant Physiology (1998) 25:131–138.[Web of Science]
Whitney SM, Andrews TJ. Plastome-encoded bacterial ribulose-1,5-bisphosphate carboxylase/oxygenase (RubisCO) supports photosynthesis and growth in tobacco. Proceedings of the National Academy of Sciences, USA (2001a) 98:14738–14743.
Whitney SM, Andrews TJ. The gene for the ribulose-1,5-bisphosphate carboxylase/oxygenase (Rubisco) small subunit relocated to the plastid genome of tobacco directs the synthesis of small subunits that assemble into Rubisco. The Plant Cell (2001b) 13:193–205.
Whitney SM, Andrews TJ. Photosynthesis and growth of tobacco with a substituted bacterial Rubisco mirror the properties of the introduced enzyme. Plant Physiology (2003) 133:287–294.
Whitney SM, Baldet P, Hudson GS, Andrews TJ. Form I Rubiscos from non-green algae are expressed abundantly but not assembled in tobacco chloroplasts. The Plant Journal (2001) 26:535–547.[CrossRef][Web of Science][Medline]
Whitney SM, Sharwood RE. Linked Rubisco subunits can assemble into functional oligomers without impeding catalytic performance. Journal of Biological Chemistry (2007) 282:3809–3818.
Whitney SM, von Caemmerer S, Hudson GS, Andrews TJ. Directed mutation of the Rubisco large subunit of tobacco influences photorespiration and growth. Plant Physiology (1999) 121:579–588.
Ye GN, Colburn SM, Xu CW, Hajdukiewicz P, Staub JM. Persistence of unselected transgenic DNA during a plastid transformation and segregation approach to herbicide resistance. Plant Physiology (2003) 133:402–410.
Zhang XH, Ewy RG, Widholm JM, Portis AR Jr. Complementation of the nuclear antisense rbcS-induced photosynthesis deficiency by introducing an rbcS gene into the tobacco plastid genome. Plant and Cell Physiology (2002) 43:1302–1313.
Zoubenko OV, Allison LA, Svab Z, Maliga P. Efficient targeting of foreign genes into the tobacco plastid genome. Nucleic Acids Research (1994) 22:3819–3824.
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  P. Barone, X.-H. Zhang, and J. M. Widholm Tobacco plastid transformation using the feedback-insensitive anthranilate synthase [{alpha}]-subunit of tobacco (ASA2) as a new selectable marker J. Exp. Bot., July 1, 2009; 60(11): 3195 - 3202. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. M. Whitney, H. J. Kane, R. L. Houtz, and R. E. Sharwood Rubisco Oligomers Composed of Linked Small and Large Subunits Assemble in Tobacco Plastids and Have Higher Affinities for CO2 and O2 Plant Physiology, April 1, 2009; 149(4): 1887 - 1895. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||