T-DNA vector backbone sequences are frequently integrated into the genome of transgenic plants obtained by Agrobacterium-mediated transformation

Transgenic Arabidopsis and tobacco plants (125) derived from seven Agrobacterium-mediated transformation experiments were screened by polymerase chain reaction and DNA gel blot analysis for the presence of vector `backbone' sequences. The percentage of plants with vector DNA not belonging to the T-DNA varied between 20% and 50%. Neither the plant species, the explant type used for transformation, the replicon type nor the selection seem to have a major influence on the frequency of vector transfer. Only the border repeat sequence context could have an effect because T-DNA vector junctions were found in more than 50% of the plants of three different transformation series in which T-DNAs with octopine borders without inner border regions were used. Strikingly, many transgenic plants contain vector backbone sequences linked to the left T-DNA border as well as vector junctions with the right T-DNA border. DNA gel blots indicate that in most of these plants the complete vector sequence is integrated. We assume that integration into the plant genome of complete vector backbone sequences could be the result of a conjugative transfer initiated at the right border and subsequent continued copying at the left and right borders, called read-through. This model would imply that the left border is not frequently recognized as an initiation site for DNA transfer and that the right border is not efficiently recognized as a termination site for DNA transfer.     

Microhomologies between T-DNA ends and target sites often occur in inverted orientation and may be responsible for the high frequency of T-DNA-associated inversions

Sequence analysis of left and right border integration sites of independent, single-copy T-DNA inserts in Arabidopsis thaliana revealed three previously unrecognized concomitants of T-DNA integration. First, genomic pre-insertion sites shared sequence similarity not only with the T-DNA left and right border regions, as was previously reported, but also at high frequency with the inverted complement of the T-DNA right border region. Second, palindromic sequences were frequently found to overlap or lie adjacent to genomic target sites, suggesting a high recombinogenic potential for palindromic elements during T-DNA integration and a possible role during the primary contact between the T-DNA and the target DNA. Third, “filler” DNA sequences between genomic pre-insertion site DNA and T-DNA often derive from sequences in the T-DNA left and right border regions that are clustered around palindromic sequences in these T-DNA regions, suggesting that these palindromic elements are “hot spots” for filler DNA formation. The discovery of inverted sequence similarities at the right border suggests a previously unrecognized mode of T-DNA integration that involves heteroduplex formation at both T-DNA borders and with opposite strands of the target DNA. Scanning for sequence similarities in both direct and inverted orientation may increase the probability and/or effectiveness of anchoring the T-DNA to the target DNA. Variations on this scheme may also account for inversion events at the target site of T-DNA integration and inverted T-DNA repeat formation, common sequence organization patterns associated with T-DNA integration. Electronic Supplementary Material Supplementary material is available in the online version of this article at and is accessible for authorized users.     

Molecular characterization of transgenic shallots (Allium cepa L.) by adaptor ligation PCR (AL-PCR) and sequencing of genomic DNA flanking T-DNA borders

Genomic DNA blot hybridization is traditionally used to demonstrate that, via genetic transformation, foreign genes are integrated into host genomes. However, in large genome species, such as Allium cepa L., the use of genomic DNA blot hybridization is pushed towards its limits, because a considerable quantity of DNA is needed to obtain enough genome copies for a clear hybridization pattern. Furthermore, genomic DNA blot hybridization is a time-consuming method. Adaptor ligation PCR (AL-PCR) of genomic DNA flanking T-DNA borders does not have these drawbacks and seems to be an adequate alternative to genomic DNA blot hybridization. Using AL-PCR we proved that T-DNA was integrated into the A. cepa genome of three transgenic lines transformed with Agrobacterium tumefaciens EHA105 (pCAMBIA 1301). The AL-PCR patterns obtained were specific and reproducible for a given transgenic line. The results showed that T-DNA integration took place and gave insight in the number of T-DNA copies present. Comparison of AL-PCR and previously obtained genomic DNA blot hybridization results pointed towards complex T-DNA integration patterns in some of the transgenic plants. After cloning and sequencing the AL-PCR products, the junctions between plant genomic DNA and the T-DNA insert could be analysed in great detail. For example it was shown that upon T-DNA integration a 66bp genomic sequence was deleted, and no filler DNA was inserted. Primers located within the left and right flanking genomic DNA in transgenic shallot plants were used to recover the target site of T-DNA integration.     

Molecular analysis of <Emphasis Type="Italic">Agrobacterium</Emphasis> T-DNA integration in tomato reveals a role for left border sequence homology in most integration events

Studies in several plants have shown that Agrobacterium tumefaciens T-DNA can integrate into plant chromosomal DNA by different mechanisms involving single-stranded (ss) or double-stranded (ds) forms. One mechanism requires sequence homology between plant target and ssT-DNA border sequences and another double-strand-break repair in which preexisting chromosomal DSBs “capture” dsT-DNAs. To learn more about T-DNA integration in Solanum lycopersicum we characterised 98 T-DNA/plant DNA junction sequences and show that T-DNA left border (LB) and right border transfer is much more variable than previously reported in Arabidopsis thaliana and Populus tremula. The analysis of seven plant target sequences showed that regions of homology between the T-DNA LB and plant chromosomal DNA plays an important role in T-DNA integration. One T-DNA insertion generated a target sequence duplication that resulted from nucleolytic processing of a LB/plant DNA heteroduplex that generated a DSB in plant chromosomal DNA. One broken end contained a captured T-DNA that served as a template for DNA repair synthesis. We propose that most T-DNA integrations in tomato require sequence homology between the ssT-DNA LB and plant target DNA which results in the generation of DSBs in plant chromosomal DNA.     

Polymerase chain reaction analysis of transgenic plants contaminated byAgrobacterium

Agrobacterium-mediated genetic transformation is a method of choice for the development of transgenic plants. The presence of latentAgrobacterium that multiplies in the plant tissue in spite of antibiotic application confounds the results obtained by polymerase chain reaction (PCR) analysis of putative transgenic plants. The presence ofAgrobacterium can be confirmed by amplification of eitherAgrobacterium chromosomal genes or genes present out of transfer DNA (T-DNA) in the binary vector. However, the transgenic nature ofAgrobacterium-contaminated transgenic plants cannot be confirmed by PCR. Here we report a simple protocol for PCR analysis ofAgrobacterium-contaminated transgenic plants. This protocol is based on denaturation and renaturation of DNA. The contaminating plasmid vector becomes double-stranded after renaturation and is cut by a restriction enzyme having site(s) within the PCR amplicon. As a result, amplification by PCR is not possible. The genomic DNA with a few copies of the transgene remains single-stranded and unaffected by the restriction enzyme, leading to amplification by PCR. This protocol has been successfully tested with 4 different binary vectors and 3Agrobacterium tumefaciens strains: EHA105, LBA4404, and GV3101.     

A quick method to estimate the T-DNA copy number in transgenic plants at an early stage after transformation,using inverse PCR

Agrobacterium-mediated transformation of plants is known to result in transgenic plants with a variable number of integrated T-DNA copies [1, 2, 3, 7]. Our aim was to obtain transgenic tobacco plants containing one integrated T-DNA copy per genome. Therefore, a quick method was developed to estimate the T-DNA copy number of young transgenic plantlets within 10 weeks after transformation. Inverse polymerase chain reaction (IPCR) was used to amplify junction fragments, i.e. plant genomic DNA sequences flanking the known T-DNA sequences [5].     

Single-copy T-DNA insertions in Arabidopsis are the predominant form of integration in root-derived transgenics,whereas multiple insertions are found in leaf discs

Different patterns of T-DNA integration in Arabidopsis were obtained that depended on whether a root or a leaf-disc transformation method was used. An examination of 82 individual transgenic Arabidopsis plants, derived from 15 independent Agrobacterium-mediated transformations in which different cointegrate and binary constructs were used, indicated that the transformation method had a significant influence on the type and copy number of T-DNA integration events. Southern hybridizations showed that most of the transgenic plants produced by a leaf-disc method contained multiple T-DNA insertions (89%), the majority of which were organized as right-border inverted repeat structures (58%). In contrast, a root transformation method mostly resulted in single T-DNA insertions (64%), with fewer right-border inverted repeats (38%). The transformation vectors, including cointegrate and binary types, and the plant selectable markers, hygromycin phosphotransferase and dihydrofolate reductase, did not appear to influence the T-DNA integration patterns.     

Molecular Structure and Regulatory Potential of a T-DNA Integration Site in Petunia

The genomic structure surrounding a T-DNA integration site in a transgenic petunia plant, which shows deregulation of a root-specific promoter, was investigated. We have already demonstrated that T-DNA integration in this transformant (P13) had occurred close to a scaffold/matrix attachment region (S/MAR). A major question regarding the observed promoter leakiness was whether the T-DNA had integrated into the centre or at the border of the Petun-SAR and whether other regulatory elements are located within this genomic region. While small rearrangements were shown to occur during T-DNA integration in agreement with other reports, we find indications of the presence of a SINE retroposon – an apparent landmark for recombinogenic targets – at the integration site. Binding assays to both plant and animal nuclear scaffolds, supported by biomathematical analyses, reveal that the T-DNA is definitely located at the border of a strong S/MAR, which is in agreement with current models on the structure of integration sites. These results, together with a developmentally regulated leaf-specific enhancer effect of the Petun-SAR on gene expression in transgenic tobacco plants, indicate that the Petun-SAR demarcates the right border of a chromatin domain with genes predominantly active in leaves.     

Characterization of T-DNA insertions in transgenic grapevines obtained by Agrobacterium-mediated transformation

T-DNA integration patterns in 49 transgenic grapevines produced via Agrobacterium-mediated transformation were analyzed. Inverse PCR (iPCR) was performed to identify T-DNA/plant junctions. Sequence comparison revealed several deletions in the T-DNA right border (RB) and left border (LB), and filler DNA and duplications or deletions of grapevine DNA at the T-DNA insertion loci. In 20 T-DNA/grapevine genome junctions microsimilarities were found associated with the joining points and in all grapevine lines microsimilarities were present near the breaking points along the 30 bases of T-DNA adjacent to the two borders. Analysis of target site preferences of T-DNA insertions indicated a non-random distribution of the T-DNA, with a bias toward the intron regions of the grapevine genes. Compositional analysis of grapevine DNA around the T-DNA insertion sites revealed an inverse relationship between the CG and AT-skews and AT rich sequences present at 300–500 bp upstream the insertion points, near the RB of the T-DNA. PCR assays showed that vector backbone sequences were integrated in 28.6% of the transgenic plants analyzed and multiple T-DNAs frequently integrated at the same position in the plant genome, resulting in the formation of tandem and inverted repeats.     

Alternative selectable markers for potato transformation using minimal T-DNA vectors

Alternative selection systems for plant transformation are especially valuable in clonal crops, such as potato (Solanum tuberosum L.), to pyramid transgenes into the same cultivar by successive transformation events. We have modified the pGPTV series of binary vectors to construct pMOA1 to pMOA5, resulting in a series of essentially identical binary vectors except for the presence of different selectable marker genes. These selectable marker genes are tightly inserted between the left and right T-DNA borders and confer resistance to kanamycin (nptII), hygromycin (hpt), methotrexate (dhfr), phosphinothricin (bar), or phleomycin (ble). The T-DNA of all the vectors is based on the minimal features necessary for plant transformation, with no extraneous DNA segments that may be unacceptable to regulatory authorities for general release of transgenic plants. A series of unique restriction sites exists between the right border and each selectable marker gene for subsequent insertion of useful genes. We have also developed improved culture procedures for potato transformation and used the pMOA1 to pMOA5 binary vectors to define stringent selection conditions for each marker gene. Combining these advances improved the frequency of recovering transformed potato plants while maintaining a low frequency of escapes. The relative efficiency of recovering transgenic potato lines with each selectable marker gene can be summarised as: kanamycin resistance>hygromycin resistance>phosphinothricin resistance>phleomycin resistance>methotrexate resistance.     

Transgene integration and organization in Cotton (Gossypium hirsutum L.) genome

While genetically modified upland cotton (Gossypium hirsutum L.) varieties are ranked among the most successful genetically modified organisms (GMO), there is little knowledge on transgene integration in the cotton genome, partly because of the difficulty in obtaining large numbers of transgenic plants. In this study, we analyzed 139 independently derived T0 transgenic cotton plants transformed by Agrobacterium tumefaciens strain AGL1 carrying a binary plasmid pPZP-GFP. It was found by PCR that as many as 31% of the plants had integration of vector backbone sequences. Of the 110 plants with good genomic Southern blot results, 37% had integration of a single T-DNA, 24% had two T-DNA copies and 39% had three or more copies. Multiple copies of the T-DNA existed either as repeats in complex loci or unlinked loci. Our further analysis of two T1 populations showed that segregants with a single T-DNA and no vector sequence could be obtained from T0 plants having multiple T-DNA copies and vector sequence. Out of the 57 T-DNA/T-DNA junctions cloned from complex loci, 27 had canonical T-DNA tandem repeats, the rest (30) had deletions to T-DNAs or had inclusion of vector sequences. Overlapping micro-homology was present for most of the T-DNA/T-DNA junctions (38/57). Right border (RB) ends of the T-DNA were precise while most left border (LB) ends (64%) had truncations to internal border sequences. Sequencing of collinear vector integration outside LB in 33 plants gave evidence that collinear vector sequence was determined in agrobacterium culture. Among the 130 plants with characterized flanking sequences, 12% had the transgene integrated into coding sequences, 12% into repetitive sequences, 7% into rDNAs. Interestingly, 7% had the transgene integrated into chloroplast derived sequences. Nucleotide sequence comparison of target sites in cotton genome before and after T-DNA integration revealed overlapping microhomology between target sites and the T-DNA (8/8), deletions to cotton genome in most cases studied (7/8) and some also had filler sequences (3/8). This information on T-DNA integration in cotton will facilitate functional genomic studies and further crop improvement.     

Screening of transgenic plants by amplification of unknown genomic DNA flanking T-DNA.

For the screening of transfer DNA (T-DNA) integration in transgenic plant material, we developed a method based on specific amplification of genomic plant DNA flanking T-DNA borders. This approach is possible because the length of the region flanking T-DNA extremity on a restriction fragment is specific to the integration locus. We have modified an adaptor ligation PCR technique developed for amplification of unknown DNA flanking known sequence. The PCR patterns obtained were specific and reproducible for different plants from a given transgenic line. Furthermore, the number of PCR products obtained could be considered a good estimation of the T-DNA copy number. When compared to Southern blot analysis, the PCR results give valuable complementary information about the complexity of the T-DNA integration pattern and also about the integrity of the T-DNA borders. We describe the applications of this approach to populations of transgenic Arabidopsis thaliana plants.     

Chromosomal localization of foreign genes in Petunia hybrida

Summary A F1 hybrid of Petunia hybrida, heterozygous for at least one marker on each of the seven chromosomes, was transformed with a modified strain of Agrobacterium tumefaciens in which the phytohormone biosynthetic genes in the transferred DNA (T-DNA) were replaced with a NOS/NPTII/NOS chimeric gene and a wildtype nopaline synthase (NOS) gene. The chimeric gene, which confers kanamycin resistance, was used as selectable marker during the transformation process and the NOS gene was used as a scorable marker in the genetic studies. After plants had been regenerated from the transformed tissues, the transgenic plants that expressed both of these markers were backcrossed to the parental lines. The offspring were examined for the segregation of the NOS gene and the Petunia markers. Genetic mapping was thus accomplished in a single generation.By Southern hybridization analysis we confirmed the presence of the expected T-DNA fragments in the transformed plants. Four out of the six plants presented here, had just one monomeric T-DNA insertion. The sizes of the plant/T-DNA junction fragments suggest that the integration occurred in different sites of the Petunia genome. One transformant gave a more complicated hybridization pattern and possibly has two T-DNA inserts. Another transgenic plant was earlier reported (Fraley et al. 1985) to have two, possibly tandemly repeated T-DNAs.Data is presented on the genetic localization of the T-DNA inserts in six independently obtained transgenic plants. The T-DNA inserts in three plants were mapped to chromosome I. However, the distances between the NOS gene and the marker gene on this chromosome were significantly different. In another transgenic plant the NOS gene was coinherited with the marker on chromosome IV. Two other transgenic plants have the T-DNA insert on chromosome III. A three point cross enabled us to determine that both plants have the NOS gene distally located from the peroxidaseA (prxA) marker and both plants showed about 18% recombination. However, Southern hybridization analysis shows that the sizes of the plant/T-DNA junction fragments in these transgenic plants are different, thus suggesting that the integrations occurred in different sites.     

A protocol for efficient biolistic transformation of mothbean<Emphasis Type="Italic">Vigna aconitifolia</Emphasis> L. Jacq. Marechal

A protocol for efficient direct gene transfer by using particle gun bombardment was developed for mothbeanVigna aconitifolia L. Jacq. Marechal. Hypocotyl explants from 2 cultivars of mothbean were transformed with 3 plasmids: pBI121, pHS101, and pHS102. Stable transformants were regenerated on MS medium supplemented with benzyladenine, α-naphthaleneacetic acid, and kanamycin. The helium pressure, plasmid type, and cultivar that were used determined the stable transformation frequency. Complete plants were regenerated and transferred to soil. The integration of the stable transgenes and reporter genes in plant genomes was shown by means of PCR amplification of these genes from plant genomic DNA and Southern blot hybridization with gene-specific probes. This method allows high-efficiency production of transgenic plants in mothbean. Suchita Kamble and Hari S. Misra contributed equally to this work.     

Removal of the Selectable Marker Gene from Transgenic Tobacco Plants by Expression of Cre Recombinase from a Tobacco Mosaic Virus Vector through Agroinfection

Selection markers are often indispensable during the process of plant transformation, but dispensable once transgenic plants have been established. The Cre/lox site-specific recombination system has been employed to eliminate selectable marker genes from transgenic plants. Here we describe the use of a movement function-improved Tobacco Mosaic Virus (TMV) vector, m30B, to express Cre recombinase for elimination of the selectable marker gene nptII from transgenic tobacco plants. The transgenic tobacco plants were produced by Agrobacterium-mediated transformation with a specially designed binary vector pGNG which contained in its T-DNA region a sequence complex of 35S promoter-lox-the gfp coding sequence-rbcS terminator-Nos promoter-nptII-Nos terminator-lox-the gus coding region-Nos terminator. The expression of the recombinant viral vector m30B:Cre in plant cells was achieved by placing the viral vector under the control of the 35S promoter and through agroinoculation. After co-cultivating the pGNG-leaf discs with agro35S-m30B:Cre followed by shoot regeneration without any selection, plants devoid of the lox-flanked sequences including nptII were obtained with an efficiency of about 34% as revealed by histochemical GUS assay of the regenerants. Three of 11 GUS expressing regenerants, derived from two independent transgenic lines containing single copy of the pGNG T-DNA, proved to be free of the lox-flanked sequences by Southern blot analysis. Excision of the lox-flanked sequences in the three plants could be attributed to transient expression of Cre from the viral vector at the early stage of co-cultivation, since the cre sequence could not be detected in the viral RNA molecules accumulated in the plants, nor in their genomic DNA. The parental marker-free genotype was inherited in their selfed progeny, and all of the progeny were virus-free, apparently because TMV is not seed-transmissible. Therefore, expression of Cre from a TMV-based vector could be used to eliminate selectable marker genes from transgenic tobacco plants without sexual crossing and segregation, and this strategy could be extended to other TMV-infected plant species and applicable to other compatible virus–host plant systems.     

Agrobacterium tumefaciens-mediated transformation of Allium cepa L.: the production of transgenic onions and shallots

This paper describes the development of a reliable transformation protocol for onion and shallot (Allium cepa L.) which can be used year-round. It is based on Agrobacterium tumefaciens as a vector, with three-week old callus, induced from mature zygotic embryos, as target tissue. For the development of the protocol a large number of parameters were studied. The expression of the uidA gene coding for -glucuronidase was used as an indicator in the optimization of the protocol. Subspecies (onion and shallot) and cultivar were important factors for a successful transformation: shallot was better than onion and for shallot cv. Kuning the best results were obtained. Also, it was found that constantly reducing the size of the calli during subculturing and selection by chopping, thus enhancing exposure to the selective agent hygromycin, improved the selection efficiency significantly. Furthermore, callus induction medium and co-cultivation period showed a significant effect on successful stable transformation. The usage of different Agrobacterium strains, callus ages, callus sources and osmotic treatments during co-cultivation did not influence transformation efficiency. The highest transformation frequency (1.95%), was obtained with shallot cv. Kuning. A total of 11 independent transformed callus lines derived from zygotic embryos were produced: seven lines from shallot and four lines from onion. Large differences in plantlet production were observed among these lines. The best line produced over 90 plantlets. Via PCR the presence of the uidA and hpt (hygromycin phosphotransferase) genes could be demonstrated in these putative transformed plants. Southern hybridization showed that most lines originated from one transformation event. However, in one line plants were obtained indicating the occurrence and rescue of at least three independent transformation events. This suggested that T-DNA integration occurred in different cells within the callus. Most transgenic plants only had one copy of T-DNA integrated into their genomes. FISH performed on 12 plants from two different lines representing two integration events showed that original T-DNA integration had taken place on the distal end of chromosomes 1 or 5. A total of 83 transgenic plants were transferred to the greenhouse and these plants appeared to be diploid and normal in morphology.     

Deviating T-DNA transfer fromAgrobacterium tumefaciens to plants

We analyzed 29 T-DNA inserts in transgenicArabidopsis thaliana plants for the junction of the right border sequences and the flanking plant DNA. DNA sequencing showed that in most lines the right border sequences transferred had been preserved during integration, corroborating literature data. Surprisingly, in four independent transgenic lines a complete right border repeat was present followed by binary vector sequences. Cloning of two of these T-DNA inserts by plasmid rescue showed that in these lines the transferred DNA consisted of the complete binary vector sequences in addition to the T-region. On the basis of the structure of the transferred DNA we propose that in these lines T-DNA transfer started at the left-border repeat, continued through the vector part, passed the right border repeat, and ended only after reaching again this left-border repeat.     

Molecular characterization of T-DNA integration sites in transgenic birch

The integration and structure of a transgene locus can have profound effects on the level and stability of transgene expression. We screened 28 transgenic birch (Betula platyphylla Suk.) lines transformed with an insect-resistance gene (bgt) using Agrobacterium tumefaciens. Among the transgenic plants, the copy number of transgene varied from one to four. A rearrangement or partial deletion had occurred in the process of T-DNA integration. T-DNA repeat formation, detected by reverse primer PCR, was found among randomly screened transgenic lines. Sequencing of the junctions between the T-DNA inserts revealed deletions of 19–589 bp and an additional 45 bp filler DNA sequence was inserted between the T-DNA repeats at one junction. Micro-homologous sequences (1–6 bp) were observed in the junctions between the T-DNA inserts. Using SiteFinding-PCR, a relatively high percentage of AT value was found for the flanking regions. Deletion of the right border repeat was observed in 12/18 of the T-DNA/plant junctions analyzed. The number of nucleotides deleted varied from 3 to 712. Deletions of 17–89 bp were observed in all left T-DNA/plant junctions analyzed. A vector backbone DNA sequence in the transgene loci was also detected using primer pairs outside the left and right T-DNA borders. Approximately 89.3% of the lines contained some vector backbone DNA. These observations revealed that it is important to check the specificity of the integration. A mechanism of T-DNA transport and integration is proposed for this long-lived tree species.     

Agrobacterium tumefaciens-mediated transformation of the aromatic shrub Lavandula latifolia

Transgenic plants of the aromatic shrub Lavandula latifolia (Lamiaceae) were produced using Agrobacterium tumefaciens-mediated gene transfer. Leaf and hypocotyl explants from 35–40-day old lavender seedlings were inoculated with the EHA105 strain carrying the nptII gene, as selectable marker, and the reporter gusA gene with an intron. Some of the factors influencing T-DNA transfer to L. latifolia explants were assessed. Optimal transformation rates (6.0 ± 1.6% in three different experiments) were obtained when leaf explants precultured for 1 day on regeneration medium were subcultured on selection medium after a 24 h co-cultivation with Agrobacterium. Evidence for stable integration was obtained by GUS assay, PCR and Southern hybridisation. More than 250 transgenic plants were obtained from 37 independent transformation events. Twenty-four transgenic plants from 7 of those events were successfully established in soil. -glucuronidase activity and kanamycin resistance assays in greenhouse-grown plants from two independent transgenic lines confirmed the stable expression of both gusA and nptII genes two years after the initial transformation. Evidence from PCR data, GUS assays and regeneration in the presence of kanamycin demonstrated a 1:15 Mendelian segregation of both transgenes among seedlings of the T1 progeny of two plants from one transgenic L. latifolia line.     

High meiotic stability of a foreign gene introduced into tobacco by Agrobacterium-mediated transformation

Summary Two lines of transgenic Nicotiana tabacum transformed to kanamycin resistance by means of a binary Agrobacterium vector containing a nos-npt gene were investigated over three generations. Southern hybridization and crossing analyses revealed that a single copy of T-DNA had integrated in each line and that the kanamycin resistance was regularly transmitted to the progeny as a monogenic dominant trait. Homozygous transgenic plants were fully fertile, morphologically normal and did not significantly differ from wild-type plants in the quantitative characters examined (plant height, flowering time). The two lines showed very low, but significantly different levels of meiotic instability: kanamycin-sensitive plants occurred among backcross progeny from homozygous transgenic plants with frequencies of 6/45,000 and 25/45,000, respectively. The sensitive plants arose independently of each other and thus resulted from meiotic rather than mitotic events. These findings demonstrate for the first time that integrated foreign genes can be transmitted to progeny with the high degree of meiotic stability required for commercial varieties of crop plants. They emphasize the importance of non-homologous integration and of avoiding co-integration of inactive gene copies for achieving meiotically stable transformants.