Selectable marker-free transgenic orange plants recovered under non-selective conditions and through PCR analysis of all regenerants

Selectable marker (SM) genes have been considered necessary to achieve acceptable rates in the generation of transgenic plants. Genes encoding antibiotic or herbicide resistance are widely used for this purpose. In most cases, once transgenic plants have been regenerated, permanence of SM genes in the plant genome is no longer necessary, and it becomes a matter of public concern. Moreover, the removal of SM genes from transgenic plants could facilitate gene stacking through successive transformations, particularly when the availability of these markers is rather limited for most crop plants. In the genus Citrus, with highly heterozygotic species of long generation cycles, methods implying the segregation and removal of marker transgenes in the progeny are not feasible. Here, we have evaluated the direct production of SM-free citrus plants under non-selective conditions, using a “clean” binary vector carrying only the transgene of interest, and through the recovery of transformants by polymerase chain reaction (PCR) analysis of all regenerated shoots. The response of two different citrus genotypes, Carrizo citrange (intergeneric hybrid of C. sinensis L. Osb. X Poncirus trifoliata L. Raf.) and Pineapple sweet orange (C. sinensis L. Osb.), was evaluated. Our results indicate that, in this system, the competence between transgenic and non-transgenic cells is the main factor determining final transgenic regeneration frequencies. For Carrizo citrange, no transgenic plant could be recovered. For Pineapple sweet orange, marker-free transformation efficiency was 1.7%, paving the way for the viable production of orange transformants carrying only the transgene(s) of interest.     

Evaluation of selection strategies alternative to <Emphasis Type="Italic">nptII</Emphasis> in genetic transformation of citrus

The neomycin phosphotransferase (nptII) selection system has proved successful in citrus transformation; however, it may be recommendable to replace it given the pressure exerted against antibiotic-resistance selectable marker genes in transgenic plants. The present work investigates three different selection alternatives, comparing them to nptII selection in two citrus genotypes, Carrizo citrange and Pineapple sweet orange. The first method used the beta-glucuronidase (uidA) reporter marker gene for selection; the second attempted to generate marker-free plants by transforming explants with a multi-auto-transformation (MAT) vector, combining an inducible R/RS-specific recombination system with transgenic-shoot selection through expression of isopentenyl transferase (ipt) and indoleacetamide hydrolase/tryptophan monooxygenase (iaaM/H) marker genes; while the third exploited the phosphomannose isomerase (PMI)/mannose conditional positive selection system. Firstly, GUS screening of all regenerated shoots in kanamycin-free medium gave 4.3% transformation efficiency for both genotypes. Secondly, workable transformation efficiencies were also achieved with the MAT system, 7.2% for citrange and 6.7% for sweet orange. This system affords an additional advantage as it enables selectable marker genes to be used during the in vitro culture phase and later removed from the transgenic plants by inducible recombination and site-specific excision. Thirdly, the highest transformation rates were obtained with the PMI/mannose system, 30% for citrange and 13% for sweet orange, which indicates that this marker is also an excellent candidate for citrus transformation.     

Genetic transformation of lime (Citrus aurantifolia Swing.): factors affecting transformation and regeneration

We have previously developed procedures for the efficient production of sweet orange (Citrus sinensis L. Osbeck) and Carrizo citrange (C. sinensis L. Osbeck×Poncirus trifoliata L. Raf.) transgenic plants using an Agrobacterium tumefaciens-mediated transformation and shoot tip grafting in vitro regeneration system. We now report on the optimization of the cocultivation, regeneration and selection conditions for efficient and reliable production of transgenic lime (C. aurantifolia Swing.) plants. Improved transformation frequencies were obtained by cocultivating the explants with Agrobacterium on feeder plates. Optimum regeneration of transgenic shoots was obtained by exposing the explants to darkness for 2 weeks and by using kanamycin at 100 mg/l as selective agent. Attempts to use geneticin as selection antibiotic were not successful. Shoot tip grafting of regenerated shoots on Troyer citrange seedlings resulted in 100% successful production of transgenic plants. The presence and expression of the transferred genes in the regenerated plants was verified by β-glucuronidase histochemical and fluorimetric assays, neomycin phosphotransferase ELISA assays, PCR and Southern analyses. Received: 10 December 1996 / Revision received: 10 February 1997 / Accepted: 25 February 1997     

Regeneration of transgenic citrus plants from the trimmed shoot/root region of etiolated seedlings

Transformation and high efficient regeneration of transgenic plants from the trimmed etiolated shoot/root region (TESRR) of Anliucheng sweet orange [Citrus sinensis (L.) Osb.] seedling was reported. A visual green fluorescent protein (GFP) marker gene was introduced to evaluate transformation efficiency by using the explants from TESRR and epicotyls. The transformation protocol was: infection 20 min, co-culture 3 d, selection culture 30 d, and rooting 15 d. Out of a total of 288 sprouted shoots obtained from TESRR, 34 shoots (11.8 %) yielded GFP expression. In contrast, only 2 (3.0 %) of the 67 sprouted shoots from epicotyl transformation yielded GFP expression. In all plants showing the green fluorescence an expected 500 bp GFP fragment was proved by PCR analysis. Southern blot analysis further confirmed the integration of GFP gene into citrus genome. Transgenic plantlets were obtained within 80 d using the TESRR, compared within 150 d by using epicotyls.     

Combining a regeneration-promoting <Emphasis Type="Italic">ipt</Emphasis> gene and site-specific recombination allows a more efficient apricot transformation and the elimination of marker genes

The presence of marker genes conferring antibiotic resistance in transgenic plants represents a serious obstacle for their public acceptance and future commercialization. In addition, their elimination may allow gene stacking by the same selection strategy. In apricot, selection using the selectable marker gene nptII, that confers resistance to aminoglycoside antibiotics, is relatively effective. An attractive alternative is offered by the MAT system (multi-auto-transformation), which combines the ipt gene for positive selection with the recombinase system R/RS for removal of marker genes from transgenic cells after transformation. Transformation with an MAT vector has been attempted in the apricot cultivar ‘Helena’. Regeneration from infected leaves with Agrobacterium harboring a plasmid containing the ipt gene was significantly higher than that from non-transformed controls in a non-selective medium. In addition, transformation efficiencies were much higher than those previously reported using antibiotic selection, probably due to the integration of the regeneration-promoting ipt gene. However, the lack of an ipt expression-induced differential phenotype in apricot made difficult in detecting the marker genes excision and plants had to be evaluated at different times. PCR analysis showed that cassette excision start occurring after 6 months approximately and 1 year in culture was necessary for complete elimination of the cassette in all the transgenic lines. Excision was confirmed by Southern blot analysis. We report here for the first time in a temperate fruit tree that the MAT vector system improves regeneration and transformation efficiency and would allow complete elimination of marker genes from transgenic apricot plants by site-specific recombination.     

Improved Transformation Efficiency in Citrus by Plasmolysis Treatment

Procedures for high efficiency production of transgenic citrus plants using an Agrobacterium tumefaciens system with plasmolysis treatment were developed. Longitudinally cut epicotyl segments of eight different citrus species [’Milam’ Rough lemon (Citrus jambhiri Lush), ‘Volkamer’ lemon (Citrus volkameriana L), Rangpur lime (Citrus limonia L), ‘Hamlin’ sweet orange (Citrus sinensis L Osbeck), ‘Duncan’ grapefruit (’Citrus paradisi’ Macf), Sour orange (Citrus aurantium L), ‘Cleopatra’ mandarin (Citrus reticulata Blanco) and Carrizo citrange (Citrus sinensis L Osbeck x Poncirus trifoliata L Raf) ] were plasmolyzed in different concentrations of sucrose and maltose [0, 3, 6, 8, 9, 10, 12 % (w/v) ] prior to Agrobacterium inoculation. Plasmolyzed epicotyl explants were cocultivated with either the hypervirulent Agrobacterium tumefaciens strain, the EHA-101 (harboring a binary vector pGA482GG) or Agl-1 (carrying pCAMBIA1303 vector). Both binary vectors contained neomycin phosphotransferase II (NPT II) and β-glucuronidase (GUS) genes. The binary vector, pCAMBIA1303 also contained a fused mGFP5 gene at the 3’ end of GUS gene as a reporter. Epicotyl explants of Rangpur lime, Rough and ‘Volkamer’ lemons plasmolyzed in 9–12 % maltose showed transient GUS gene expression comprising up to 95 % of the cut surface of explants, while Carrizo citrange showed 80 % expression when they were plasmolyzed in 6–10 % sucrose. On the other hand, epicotyl explants of ‘Hamlin’ sweet orange, Grapefruit, Sour orange and ‘Cleopatra’ mandarin showed transient GUS expession in 80–90 % of explants with 6–10 % sucrose. Basal portions of the regenerated putative transgenic shoots harvested from the cut surface of epicotyl explants within 2–3 months, were assayed for GUS, and apical portions were shoot-tip grafted in vivo for the production of whole plants. The transformation efficiencies in different species obtained are the highest so far reported for citrus.     

Green fluorescent protein as a screenable marker to increase the efficiency of generating transgenic woody fruit plants

 The green fluorescent protein (GFP) from Aequorea victoria has been introduced into three different citrus genotypes [Citrus aurantium L., C. aurantifolia (Christm.) Swing. and C. sinensis L. Osbeck×Poncirus trifoliata (L.) Raf.] which are considered recalcitrant to transformation, mainly due to low transformation frequencies and to the regeneration of escape shoots at high frequencies from the Agrobacterium-inoculated explants. High-level GFP expression was detected in transgenic cells, tissues and plants. Using GFP as a vital marker has allowed us to localize the sites of transgene expression in specific cells, always occurring in callus tissue formed from the cambium of the cut ends of explants. Whereas green fluorescent shoots regenerated in all cases from this callus, most escapes regenerated directly from explants with almost no callus formation. Thus, development of callus from cambium is a prerequisite for citrus transformation. Furthermore, in vivo monitoring of GFP expression permitted a rapid and easy discrimination of transgenic and escape shoots. The selection of transgenic shoots could be easily favored by eliminating the escapes and/or by performing shoot-tip grafting of the transgenic buds soon after their origin. GFP-expressing shoots have also been observed in citrus explants co-cultivated with Agrobacterium but cultured in a medium without the selective agent kanamycin. This opens the possibility to rescue the transgenic sectors and to regenerate transgenic plants without using selectable marker genes conferring antibiotic or herbicide resistance, which is currently a topic of much discussion for the commercialization of transgenic plants. Received: 28 October 1998 / Accepted: 28 November 1998     

An alternative method for the genetic transformation of sweet organce

Summary An alternative method for transforming sweet organe [Citrus sinensis (L.) Osbeck] has been developed. Plasmid DNA encoding the non-destructive selectable marker enhanced green fluorescent protein gene was introduced using polyethylene glycol into protoplasts of ‘Itaborai’ sweet organe isolated from an embryogenic nucellar-derived suspension culture. Following protoplast culture in liquid medium and transfer to solid medium, transformed calluses were identified via expression of the green fluorescent protein, physically separated from non-transformed tissue, and cultured on somatic embryogenesis induction medium. Transgenic plantlets were recovered from germinating somatic embryos and by in vitro rooting of shoots. To expedite transgenic plant recovery, regenerated shoots were also micrografted onto sour orange seedling rootstocks. Presence of the transgene in calluses and regenerated sweet organe plants was verified by gene amplification and Southern analyses. Potential advantages of this transformation system over the commonly used Agrobacterium methods for citrus are discussed.     

Influence of virus and virus-like agents on the development of citrus buds cultured in vitro

Tissue culture in vitro was used to determine the effect of six major citrus virus and virus-like agents. Nodal stem segments from inoculated Pineapple sweet orange (Citrus sinensis (L.) Osb.), Mexican lime (C. aurantifolia (Christm.) Swing.) and Arizona Etrog citron 861-Sl (C. medica L.) were cultured in vitro to induce shoots. Some virus and virus-like agents had a marked effect on bud development and further recovery of plantlets. The number and size of the shoots that developed from each bud were affected as a result of infection. The effect depended on the specific virus, the isolate and the host-disease combination. The possible implications of these results are discussed.     

Identification of closely related citrus cultivars with inter-simple sequence repeat markers

 Inter-simple sequence repeat (ISSR) markers generated by 22 primers were tested for their ability to distinguish among samples from 94 trees of 68 citrus cultivars. Within each of the six cultivar groups studied, most of these cultivars are so closely related that they are difficult to distinguish by other molecular-marker techniques. ISSR markers involve PCR amplification of DNA using a single primer composed of a microsatellite sequence anchored at the 3′ or 5′ end by 2–4 arbitrary, often degenerate, nucleotides. The amplification products were separated on non-denaturing polyacrylamide gels and detected by silver staining. ISSR banding profiles were very repeatable on duplicate samples. Different citrus species had very different fingerprint patterns. Within Citrus sinensis (L.) Osbeck and C. paradisi Macf., in which all cultivars have originated by the selection of mutants, ISSR markers distinguished 14 of 33 sweet orange and 1 of 7 grapefruit cultivars. Five of six lemon cultivars were discriminated by ISSR markers. Many differences were found among mandarin cultivars; however, all five satsuma cultivars analyzed had identical ISSR fingerprints. Four of five citrange cultivars were distinguishable, but ‘Troyer’ and ‘Carrizo’ had identical ISSR fingerprints. ‘Kuharske Carrizo’ citrange, which has better citrus nematode resistance than other ‘Carrizo’ citrange accessions, had unique ISSR fingerprints. Three ISSR markers that differentiated certain sweet orange cultivars were hybridized to Southern blots of sweet orange DNA digested with different restriction endonucleases. The sweet orange cultivars tested could be distinguished by these ISSR-derived RFLP markers. Moreover, one ISSR marker unique to ‘Ruby’ blood orange was observed in its progeny trees. Received: 9 September 1996 / Accepted: 4 April 1997     

Protoplast transformation and regeneration of transgenic Valencia sweet orange plants containing a juice quality-related pectin methylesterase gene

Valencia orange [Citrus sinensis (L.) Osbeck] is the leading commercial citrus species in the world for processed juice products; however, the presence of thermostable pectin methylesterase (TSPME) reduces its juice quality. A long-term strategy of this work is to eliminate or greatly reduce TSPME activity in Valencia orange. Previous work resulted in the isolation of a putative TSPME gene, CsPME4, associated with a thermostable protein fraction of Valencia orange juice. To begin research designed to overexpress CsPME4 to verify the thermostability of the protein product and/or to downregulate the gene, a sense gene cassette containing a gene-specific sequence from a putative TSPME cDNA and the enhanced green fluorescent protein (GFP) as a selectable marker was constructed (M2.1). In the work reported here, M2.1 plasmid DNA was transformed (polyethylene glycol-mediated) into protoplasts isolated from an embryogenic suspension culture of Valencia somaclone line B6-68, in an effort to obtain transgenic Valencia lines. A vigorous transformed line was identified via GFP expression, physically separated from non-transformed tissue, and cultured on somatic embryogenesis induction medium. One transgenic proembryo expressing GFP was recovered and multiple shoots were regenerated. The recovery of multiple transgenic plants was expedited by in vitro grafting. Polymerase chain reaction analysis revealed the presence of the PME gene in transgenic plants, and subsequent Southern blot analysis confirmed the presence of the eGFP gene. These transgenic plants show normal growth and minor morphological variation. The thermostability of PME in these plants will be assessed after flowering and fruit set. This is the first successful transfer of a target fruit-quality gene by protoplast transformation with recovery of transgenic plants in citrus. This method of transformation has the advantage over Agrobacterium-mediated transformation in that it requires no antibiotic-resistance genes.     

Field performance of transgenic citrus trees: Assessment of the long-term expression of uidA and nptII transgenes and its impact on relevant agronomic and phenotypic characteristics

ABSTRACT: BACKGROUND: The future of genetic transformation as a tool for the improvement of fruit trees depends on the development of proper systems for the assessment of unintended effects in field-grown GM lines. In this study, we used eight transgenic lines of two different citrus types (sweet orange and citrange) transformed with the marker genes beta-glucuronidase (uidA) and neomycin phosphotransferase II (nptII) as model systems to study for the first time in citrus the long-term stability of transgene expression and whether transgene-derived pleiotropic effects occur with regard to the morphology, development and fruit quality of orchard-grown GM citrus trees. RESULTS: The stability of the integration and expression of the transgenes was confirmed in 7-year-old, orchard-grown transgenic lines by Southern blot analysis and enzymatic assays (GUS and ELISA NPTII), respectively. Little seasonal variation was detected in the expression levels between plants of the same transgenic line in different organs and over the 3 years of analysis, confirming the absence of rearrangements and/or silencing of the transgenes after transferring the plants to field conditions. Comparisons between the GM citrus lines with their non-GM counterparts across the study years showed that the expression of these transgenes did not cause alterations of the main phenotypic and agronomic plant and fruit characteristics. However, when comparisons were performed between diploid and tetraploid transgenic citrange trees and/or between juvenile and mature transgenic sweet orange trees, significant and consistent differences were detected, indicating that factors other than their transgenic nature induced a much higher phenotypic variability. CONCLUSIONS: Our results indicate that transgene expression in GM citrus remains stable during long-term agricultural cultivation, without causing unexpected effects on crop characteristics. This study also shows that the transgenic citrus trees expressing the selectable marker genes that are most commonly used in citrus transformation were substantially equivalent to the non-transformed controls with regard to their overall agronomic performance, as based on the use of robust and powerful assessment techniques. Therefore, future studies of the possible pleiotropic effects induced by the integration and expression of transgenes in field-grown GM citrus may focus on the newly inserted trait(s) of biotechnological interest.     

Agrobacterium-mediated transformation of sweet orange and regeneration of transgenic plants

Summary Transgenic sweet orange (Citrus sinensis L. Osbeck) plants have been obtained by Agrobacterium tumefaciens-mediated gene transfer. An hypervirulent A. tumefaciens strain harboring a binary vector that contains the chimeric neomycin phosphotransferase II (NPT II) and ß-glucuronidase (GUS) genes was cocultivated with stem segments from in vivo grown seedlings. Shoots regenerated under kanamycin selection were harvested from the stem segments within 12 weeks. Shoot basal portions were assayed for GUS activity and the remaining portions were shoot tip grafted in vitro for production of plants. Integration of the GUS gene was confirmed by Southern analysis. This transformation procedure showed the highest transgenic plant production efficiency reported for Citrus.Abbreviations BA benzyladenine - CaMV cauliflowermosaic virus - GUS ß-glucuronidase - LB Luria Broth - MS Murashige and Skoog - NAA naphthalenacetic acid - NOS nopaline synthase - NPT II neomycin phosphotransferase II - PEG polyethylene glycol - RM rooting medium - SRM shoot regeneration medium     

Morphogenesis and tissue cultures of three citrus species

Studies were performed to define tissue culture techniques and culture conditions for morphogenesis, callus culture and plantlet culture of sweet orange (Citrus sinensis (L.) Osb.), citron (C. medica L.) and lime (C. aurantifolia) (Christm. Swing). The optimal concentrations of NAA to induce root formation on stem segments were 10 mg l^-1 for sweet orange and lime, and 3 mg l^-1 for citron. The optimal BA concentration for shoot and bud proliferation was 3 mg l^-1 for sweet orange and citron, and 1 mg l^-1 for lime. Callus initiation was accomplished in a culture medium containing 10 mg l^-1 NAA and 0.25 mg l^-1 BA. Callus was maintained by periodical subculture into the same medium supplemented with 10% (v:v) organge juice. In vitro plantlets of the three species were obtained by rooting of shoots developed from bud cultures, and of citron and lime by development of shoots from root cultures. The plants were successfully established on soil.     

Evaluation of parameters affecting <Emphasis Type="Italic">Agrobacterium</Emphasis>-mediated transformation of citrus

An improved protocol for genetic transformation of juvenile explants of Carrizo (Citrus sinensis Osb. × Poncirus trifoliata L. Raf.), Duncan (Citrus paradisi Macf.), Hamlin (Citrus sinensis (L.) Osbeck) and Mexican Lime (Citrus aurantifolia Swingle) cultivars using a vector containing a bifunctional egfp-nptII fusion gene is described. Several parameters were investigated to optimize genetic transformation of these four cultivars. It was determined that a short preincubation in hormone rich liquid medium and subculture of Agrobacterium for 3 h in YEP medium containing 100 μM acetosyringone were required for improvement of transformation efficiency. Co-cultivation duration as well as addition of acetosyringone to co-cultivation medium also played an important role in transformation efficiency as did OD₆₀₀ value of the Agrobacterium suspension used for transformation. We regenerated numerous EGFP expressing transgenic lines from all four cultivars. Based on these results, we conclude that genetic transformation of citrus is cultivar specific and optimization of conditions for maximum transgenic production are required for each individual cultivar.     

Production of Marker-free Transgenic Tobacco Plants by FLP/frt Recombination System

Selectable marker genes that usually encode antibiotic or herbicide resistances are widely used for the selection of transgenic plants, but they become unnecessary and undesirable after transformation selection. An important strategy to improve the transgenic plants' biosafety is to eliminate the marker genes after successful selection. In the FLP/frt site-specific system of 2-μm plasmid from Saccharomyces cerevisiae, the FLP enzyme efficiently catalyzes recombination between two directly repeated FLP recombination target (frt) sites, eliminating the sequence between them. By controlled expression of the FLP recombinase and specific allocation of the frt sites within transgenic constructs, the system can be applied to eliminate the marker genes after selection. Through a series of procedures, the plant FLP/frt site-specific recombination system was constructed, which included the frt-containing vector pCAMBIA1300-betA-frt-als-frt and the FLP expression vector pCAMBIA1300-hsp-FLP-hpt. The FLP recombinase gene was introduced into transgenic (betA-frt-als-frt) tobacco plants by re-transformation. In re-transgenic plants, after heat-shock treatment, the marker gene als flanked by two identical orientation frt sites could be excised by the inducible expression of FLP recombinase under the control of hsp promoter. Excision of the als gene was found in 41 % re-transgenic tobacco plants, which indicated that this system could make a great contribution obtaining the marker-free transgenic plants.     

Genetic transformation of three sweet orange cultivars from explants of adult plants

The difficulty in adult tissue genetic transformation in woody species is still an obstacle to be overcome, including in most sweet orange cultivars of the Brazilian citrus industry. This work reports that, after in vitro culture adjustments, transgenic adventitious buds of ‘Hamlin’, ‘Pêra’, and ‘Valencia’ sweet oranges (Citrus sinensis L. Osbeck) were recovered using adult material as explant source, in genetic transformation experiments via Agrobacterium tumefaciens. The transgenic buds were identified by the GUS histochemical analysis and confirmed by PCR analysis, which indicated the presence of an amplified fragment of 817 bp corresponding to the uidA gene sequence. The efficiencies of genetic transformation for ‘Hamlin’, ‘Pêra’, and ‘Valencia’ sweet orange cultivars were 2.5, 1.4, and 3.7%, respectively. Media supplemented with auxins and cytokinins during co-culture, and media with high concentrations of cytokinins (3 mg L⁻¹) during transgenic selection led to the transformation and, consequently, the regeneration of adequate number of adventitious buds for the three cultivars. The use of sonication during the explant disinfection was not effective to reduce endophytic contamination and reduced transformation efficiency.