Improving in vitro culture and transformation conditions in Agrobacterium-mediated transformation of maize --Conti E, Lanzanova C, Baldoni E, Allegri L, Lupotto E Agrobacterium-mediated transformation of cereal crops is being extensively investigated because of the many advantages offered by this system of gene delivery over the widely adopted biolistic method. It is indeed now well attested that Agrobacterium-mediated transformation —now possible also in monocotiledonous species — results in a higher number of stable, low copy number, correctly expressing transgenic events, than the biolistic system. Agrobacterium-mediated transformation of maize, first based on the adoption of supervirulent strains carrying Ti-plasmid engineered with extra copies of the virB, virC and virG genes (Ishida et al., Nature Biotechnol 14:745-750, 1996), can now also be effective using standard binary vector systems and L-cysteine as a transformation enhancer (Frame et al, Plant Physiol 129:13-22, 2002).

Present work in our laboratory aims at extending the methodology of transformation mediated by Agrobacterium tumefaciens to maize genotypes of agronomical value. This includes the identification of reliable protocols aimed to enhance the in vitro responsiveness — related to the development of embryogenic callus cultures — and the frequency of transformation.

In a previous work (Lupotto et al., Maydica 44:211-218, 1999), some parameters for the establishment of a transformation system in maize have been explored in the maize A188 inbred line and in a series of Lo inbred lines of agronomical importance (Bertolini et al., Maydica 36:87-106, 1991). In these latter inbreds, the lack of in vitro responsiveness appeared to be the major bottleneck in obtaining transgenic events. As a first step toward our goal, a series of Lo inbred lines belonging to different heterotic groups (Hartings et al., Maize Coop. Newslet. 76:5-6, 2002) was chosen. Their in vitro response to the embryogenic callus induction (EC) and subsequent regenerative capability were evaluated in the Lo per se and in crosses between each Lo and the highly embryogenic maize model genotype A188 (Table 1). With the exception of Lo1010, the induction of E callus did not appear to be a limiting step; E calli were obtained in all cases, although at different rates depending on the genotype. As expected, the in vitro response of F2 immature embryos derived from each cross showed a significantly higher frequency of EC induction. The most important result was the case of Lo1010xA188, in which a complete switch from non-regenerative to regenerative callus culture was obtained. The number of completely regenerated plants varied largely according to the genotype; in addition, plants raised from each cross were not only higher in number, but also more vigorous, with a properly developed root system established in vitro — which is the most important limiting factor for transplantation into soil — showing a stronger phenotype. Evaluation of the experiment based on the number of completely regenerated plants allowed us to conduct transformations in at least four different crosses.

For Agrobacterium-mediated transformation the strain A.tumefaciens EHA105 was used. This was engineered with a binary pCAMBIA3301-based vector, containing the bar gene as selectable marker, and the gus gene as visual marker. The vector also contains the cassette ubi1-b32-nos3’ for obtaining the constitutive expression of the maize gene b32 (see above Lanzanova et al.) as a potential anti-fungal gene. Transformation was performed either with immature embryos just after explant or with embryos subjected to a pre-culture period of 5 days. This additional step was introduced with the aim of overcoming the limitation in efficiency due to the deleterious effects of L-cysteine and of the infection process on the subsequent EC induction. As summarised in Table 2, the results obtained strongly supported the adoption of the additional pre-culture. The effect of L-cysteine added during the co-culture period in the case of infection of immature embryos without preculture, varied with the genotype. Of the four crosses evaluated, two were completely inhibited by L-cysteine, while the others were only partially inhibited. In all cases, the absence of L-cysteine raised the rate of EC induction after infection. When a five day pre-culture period before Agrobacterium infection was adopted, the negative effect of L-cysteine on EC induction was drastically reduced; it was also evident that the rates of EC induction paralleled those without L-cysteine. Callus culture in general showed a better phenotype and higher rate of callus proliferation (Figure 1). Presence of the engineered cassette was detected in most of the regenerated plants, as shown by PCR analysis in Figure 2. This result is indicative of the efficacy of the protocol adopted, paving the way for a more consistent set of experiments on the selected maize genotypes. The work is developed within the framework of the EU-funded project SAFEMAIZE (ICA4-CT2000-30033) in FP5.

Table 1.
Genotype No. explanted embryos No. embryogenic calli obtained % EC No. regenerating calli % R No. plants obtained
Lo1096 x A188 334 266 79.6 89 33.4 76
Lo1095 x A188 253 211 83.3 118 55.9 38
Lo904 x A188 165 124 75.1 60 48.3 23
Lo1124 x A188 163 120 73.7 17 14.1 14
Lo1067 x A188 192 128 66.7 5 3.9 2
Lo1066o2x A188 196 108 55.1 31 28.7 29
Lo1010 x A188 180 131 72.7 11 8.3 9
Lo1010 200 23 11.5 3 13 0
Lo1095 204 134 65.7 96 71.6 6
Lo1096 192 152 79.1 0 0 0
Lo904 194 77 39.7 12 15.6 2
Lo1124 175 97 55.4 0 0 0


Table 2.
Genotype No. explanted embryos % EC obtained

+cys no cys

Infection: T0    
Lo1096 x A188 100 0 85.5
Lo1095 x A188 100 52.8 86.4
Lo904 x A188 100 40 82.6
Lo1124 x A188 100 0 53.7
Infection: T5    
Lo1096 x A188 100 76.8 84.8
Lo1095 x A188 100 87.5 67.9
Lo904 x A188 100 69.5 84.4
Lo1124 x A188 100 98.8 71.7

Figure 1. Embyogenic maize callus in regeneration medium, derived from F2 immature embryos of the cross Lo904 x A188, after Agrobacterium transformation

Figure 2. PCR analysis of T0 events of transformation with Agrobacterium EHA105(pCAMBIA3301/ubi1-b32). Lanes: 1 to 6 are independent T0 events: plants 1,2,4 and 5 show the expected band for ubi1-b32; 7. A188 negative control, 8.Ladder

Please Note: Notes submitted to the Maize Genetics Cooperation Newsletter may be cited only with consent of the authors.

Return to the MNL 77 On-Line Index
Return to the Maize Newsletter Index
Return to the Maize Genome Database Page