Complex transgene locus structures implicate multiple mechanisms for plant transgene rearrangement

To more fully characterize the internal structure of transgene loci and to gain further understanding of mechanisms of transgene locus formation, we sequenced more than 160 kb of complex transgene loci in two unrelated transgenic oat (Avena sativa L.) lines transformed using microprojectile bombardment. The transgene locus sequences from both lines exhibited extreme scrambling of non-contiguous transgene and genomic fragments recombined via illegitimate recombination. A perfect direct repeat of the delivered DNA, and inverted and imperfect direct repeats were detected in the same transgene locus indicating that homologous recombination and synthesis-dependent mechanism(s), respectively, were also involved in transgene locus rearrangement. The most unexpected result was the small size of the fragments of delivered and genomic DNA incorporated into the transgene loci via illegitimate recombination; 50 of the 82 delivered DNA fragments were shorter than 200 bp. Eleven transgene and genomic fragments were shorter than the DNA lengths required for Ku-mediated non-homologous end joining. Detection of these small fragments provided evidence that illegitimate recombination was most likely mediated by a synthesis-dependent strand-annealing mechanism that resulted in transgene scrambling. Taken together, these results indicate that transgene locus formation involves the concerted action of several DNA break-repair mechanisms.     

Genomic interspersions determine the size and complexity of transgene loci in transgenic plants produced by microprojectile bombardment.

The structure of transgene loci in six transgenic allohexaploid oat (Avena sativa L.) lines produced using microprojectile bombardment was characterized using fluorescence in situ hybridization (FISH) on extended DNA fibers (fiber-FISH). The transgene loci in five lines were composed of multiple copies of delivered DNA interspersed with genomic DNA fragments ranging in size from ca. 3 kb to at least several hundred kilobases, and in greater numbers than detected using Southern blot analysis. Although Southern analysis predicted that the transgene locus in one line consisted of long tandem repeats of the delivered DNA, fiber-FISH revealed that the locus actually contained multiple genomic interspersions. These observations indicated that transgene locus size and structure were determined by the number of transgene copies and, possibly to a greater extent, the number and the length of interspersing genomic DNA sequences within the locus. Large genomic interspersions detected in several lines were most likely the products of chromosomal breakage induced either by tissue culture conditions or, more likely, by DNA delivery into the nucleus using microprojectile bombardment. We propose that copies of transgene along with other extrachromosomal DNA fragments are used as patches to repair double-strand breaks (DSBs) in the plant genome resulting in the formation of transgene loci.     

Association of transgene integration sites with chromosome rearrangements in hexaploid oat

Transgene loci in 16 transgenic oat (Avena sativa L.) lines produced by microprojectile bombardment were characterized using phenotypic and genotypic segregation, Southern blot analysis, and fluorescence in situ hybridization (FISH). Twenty-five transgene loci were detected; 8 lines exhibited single transgene loci and 8 lines had 2 or 3 loci. Double FISH of the transgene and oat C- and A/D-genome-specific dispersed and clustered repeats showed no preferences in the distribution of transgene loci among the highly heterochromatic C genome and the A/D genomes of hexaploid oat, nor among chromosomes within the genomes. Transgene integration sites were detected at different locations along individual chromosomes, although the majority of transformants had transgenes integrated into subtelomeric and telomeric regions. Transgene integration sites exhibited different levels of structural complexity, ranging from simple integration structures of two apparently contiguous transgene copies to tightly linked clusters of multiple copies of transgenes interspersed with oat DNA. The size of the genomic interspersions observed in these transgene clusters was estimated from FISH results on prometaphase chromosomes to be megabases long, indicating that some transgene loci were significantly larger than previously determined by Southern blot analysis. Overall, 6 of the 25 transgene loci were associated with rearranged chromosomes. These results suggest that particle bombardment-mediated transgene integration may result from and cause chromosomal breakage and rearrangements. Received: 29 July 1999 / Accepted: 9 November 1999     

植物转基因座位形成的分子机制

转基因座位是指染色体上插入的转基因及相邻的特定DNA序列。大多数转基因座位是以转基因片段、基因组片段和填充DNA相间而存在,仅少数含有完整的单拷贝转基因,这是由于在转基因整合过程中,转基因及基因组DNA发生缺失、重复和染色体的重排。转基因整合主要通过双链DNA断裂修复中的异常重组所产生,而同源重组也发挥了一定的作用。异常重组主要由单链复性、合成依赖链复性和依赖Ku蛋白的非同源末端连接途径调节。     

A large-scale study of rice plants transformed with different T-DNAs provides new insights into locus composition and T-DNA linkage configurations

Transgenic locus composition and T-DNA linkage configuration were assessed in a population of rice plants transformed using the dual-binary vector system pGreen (T-DNA containing the bar and gus genes)/pSoup (T-DNA containing the aphIV and gfp genes). Transgene structure, expression and inheritance were analysed in 62 independently transformed plant lines and in around 4,000 progeny plants. The plant lines exhibited a wide variety of transgenic locus number and composition. The most frequent form of integration was where both T-DNAs integrated at the same locus (56% of loci). When single-type T-DNA integration occurred (44% of loci), pGreen T-DNA was preferentially integrated. In around half of the plant lines (52%), the T-DNAs integrated at two independent loci or more. In these plants, both mixed and single-type T-DNA integration often occurred concurrently at different loci during the transformation process. Non-intact T-DNAs were present in 70–78% of the plant lines causing 14–21% of the loci to contain only the mid to right border part of a T-DNA. In 53–66% of the loci, T-DNA integrated with vector backbone sequences. Comparison of transgene presence and expression in progeny plants showed that segregation of the transgene phenotype was not a reliable indicator of either transgene inheritance or T-DNA linkage, as only 60–80% of the transgenic loci were detected by the expression study. Co-expression (28% of lines) and backbone transfer (53–66% of loci) were generally a greater limitation to the production of marker-free T₁ plants expressing the gene of interest than co-transformation (71% of lines) and unlinked integration (44% of loci).     

T-DNA locus structure in a large population of soybean plants transformed using the Agrobacterium-mediated cotyledonary-node method

Designing transformation experiments for either functional genomics or crop improvement requires knowledge of the transgene locus structure, number, transmission and expression resulting from a specific transformation method. We recently reported an improvement to the soybean [Glycine max (L.) Merrill] cotyledonary-node transformation method that resulted in the efficient production of transgenic plants. To characterize the transgene loci resulting from this method, we analysed 270 independent T0 plants and 95 randomly selected T1 progenies for T-DNA locus complexity using Southern analysis. The lines were transformed with Agrobacterium tumefaciens strains LBA4404 or EHA105 carrying the binary plasmids pGPTV, pTOK233, pCAMBIA1303 or pCAMBIA1309, and regenerated in medium supplemented with or without silver nitrate (AgNO3). Analysis in the T0 generation showed that the number of hpt-hybridizing fragments per plant ranged from 1-15, with 31.5% of the lines having a single hpt-hybridizing fragment. Each primary soybean transformant had, on average, 2.0 unlinked transgene loci and that half of the segregating loci in the T1 progenies were single, simple T-DNA insertions. Of the loci containing multiple T-DNA fragments, a low frequency had tandem and inverted repeat T-DNA structures. Integration of binary plasmid backbone sequences occurred in 37% of primary transformants. A. tumefaciens strain, binary plasmid and thiol treatment had no significant effect on transgene locus structure, numbers or expression. Interestingly, exposure of soybean explants to AgNO3 throughout shoot induction and elongation increased T-DNA locus complexity in the primary transformants and decreased silencing of gusA expression in the T1 generation.     

植物转基因整合座位的结构

通过农杆菌和直接DNA转移技术所获得的转基因植株都具有复杂的转基因座位, 转基因整合染色体和染色体区段是随机的, 但组织培养的选择作用表现为非随机性, 偏向整合于染色体的基因富集区。转基因座位除少数含有完整的单拷贝转基因之外, 大多数转基因座位中外源转基因片段、基因组片段和填充DNA相间而存在。转基因座位中转基因及基因组DNA片段产生缺失、重复和染色体的重排, 转基因的完整性对转基因表达具有重要作用。     

High-resolution structural analysis of biolistic transgene integration into the genome of wheat

Transformation of plant genomes by biolistic methods has become routine over the past decade. However, relatively little is known about how transgenes are physically integrated into the host genome. Using a high-resolution physical mapping technique, fluorescence in situ hybridization on extended DNA fibers (fiber-FISH), 13 independent transgenic wheat lines were analyzed to determine the structural arrangement of stably inherited transgenes in host-plant chromosomes. Twelve transgenic lines were transformed with a single plasmid and one line was co-transformed with two separate plasmids, which co-segregated genetically. Three basic integration patterns were observed from the fiber-FISH experiments: Type I, large tandemly repeated integration; Type II, large tandem integrations interspersed with unknown DNA; and Type III, small insertions, possibly interspersed with unknown DNA. Metaphase FISH showed that the integration of transgenes was in both hetero- and euchromatic, as well as proximal, interstitial and distal, regions of the chromosomes. In the transgenic plants, the type of promotor used, rather than the chromosomal site of transgene integration, was most critical for transgene expression. The integration of the transgenes was not associated with detectable chromosomal rearrangements. Received: 25 August 2000 / Accepted: 31 October 2000     

Zinc finger nuclease‐mediated targeting of multiple transgenes to an endogenous soybean genomic locus via non‐homologous end joining

Emerging genome editing technologies hold great promise for the improvement of agricultural crops. Several related genome editing methods currently in development utilize engineered, sequence‐specific endonucleases to generate DNA double strand breaks (DSBs) at user‐specified genomic loci. These DSBs subsequently result in small insertions/deletions (indels), base substitutions or incorporation of exogenous donor sequences at the target site, depending on the application. Targeted mutagenesis in soybean (Glycine max) via non‐homologous end joining (NHEJ)‐mediated repair of such DSBs has been previously demonstrated with multiple nucleases, as has homology‐directed repair (HDR)‐mediated integration of a single transgene into target endogenous soybean loci using CRISPR/Cas9. Here we report targeted integration of multiple transgenes into a single soybean locus using a zinc finger nuclease (ZFN). First, we demonstrate targeted integration of biolistically delivered DNA via either HDR or NHEJ to the FATTY ACID DESATURASE 2‐1a (FAD2‐1a) locus of embryogenic cells in tissue culture. We then describe ZFN‐ and NHEJ‐mediated, targeted integration of two different multigene donors to the FAD2‐1a locus of immature embryos. The largest donor delivered was 16.2 kb, carried four transgenes, and was successfully transmitted to T₁ progeny of mature targeted plants obtained via somatic embryogenesis. The insertions in most plants with a targeted, 7.1 kb, NHEJ‐integrated donor were perfect or near‐perfect, demonstrating that NHEJ is a viable alternative to HDR for gene targeting in soybean. Taken together, these results show that ZFNs can be used to generate fertile transgenic soybean plants with NHEJ‐mediated targeted insertions of multigene donors at an endogenous genomic locus.     

Transgene integration in plants: poking or patching holes in promiscuous genomes?

Transgene integration in plants transformed by either Agrobacterium or direct DNA delivery methods occurs through illegitimate recombination (IR). The precise mechanism(s) for IR-mediated transgene integration and the role of host double-strand break repair enzymes remain unknown. A recent wealth of sequenced transgene loci and investigations aimed at genetically dissecting transgene integration mechanism(s) have provided new insights into the process.     

Analysis of T-DNA-<Emphasis Type="Italic">Xa21</Emphasis> loci and bacterial blight resistance effects of the transgene <Emphasis Type="Italic">Xa21</Emphasis> in transgenic rice

The genetic loci and phenotypic effects of the transgene Xa21, a bacterial blight (BB) resistance gene cloned from rice, were investigated in transgenic rice produced through an Agrobacterium-mediated transformation system. The flanking sequences of integrated T-DNAs were isolated from Xa21 transgenic rice lines using thermal asymmetric interlaced PCR. Based on the analysis of 24 T-DNA- Xa21 flanking sequences, T-DNA loci in rice could be classified into three types: the typical T-DNA integration with the definite left and right borders, the T-DNA integration linked with the adjacent vector backbone sequences and the T-DNA integration involved in a complicated recombination in the flanking sequences. The T-DNA integration in rice was similar to that in dicotyledonous genomes but was significantly different from the integration produced through direct DNA transformation approaches. All three types of integrated transgene Xa21 could be stably inherited and expressed the BB resistance through derived generations in their respective transgenic lines. The flanking sequences of the typical T-DNA integration consisted of actual rice genomic DNA and could be used as probes to locate the transgene on the rice genetic map. A total of 15 different rice T-DNA flanking sequences were identified. They displayed restriction fragment length polymorphisms (RFLPs) between two rice varieties, ZYQ8 and JX17, and were mapped on rice chromosomes 1, 3, 4, 5, 7, 9, 10, 11 and 12, respectively, by using a double haploid population derived from a cross between ZYQ8 and JX17. The blast search and homology comparison of the rice T-DNA flanking sequences with the rice chromosome-anchored sequence database confirmed the RFLP mapping results. On the basis of genetic mapping of the T-DNA- Xa21 loci, the BB resistance effects of the transgene Xa21 at different chromosome locations were investigated using homozygous transgenic lines with only one copy of the transgene. Among the transgenic lines, no obvious position effects of the transgene Xa21 were observed. In addition, the BB resistance levels of the Xa21 transgenic plants with different transgene copy numbers and on different genetic backgrounds were also investigated. It was observed that genetic background (or genome) effects were more obvious than dosage effects and position effects on the BB resistance level of the transgenic plants.     

Targeted transgene integration in plant cells using designed zinc finger nucleases

Targeted transgene integration in plants remains a significant technical challenge for both basic and applied research. Here it is reported that designed zinc finger nucleases (ZFNs) can drive site-directed DNA integration into transgenic and native gene loci. A dimer of designed 4-finger ZFNs enabled intra-chromosomal reconstitution of a disabled gfp reporter gene and site-specific transgene integration into chromosomal reporter loci following co-transformation of tobacco cell cultures with a donor construct comprised of sequences necessary to complement a non-functional pat herbicide resistance gene. In addition, a yeast-based assay was used to identify ZFNs capable of cleaving a native endochitinase gene. Agrobacterium delivery of a Ti plasmid harboring both the ZFNs and a donor DNA construct comprising a pat herbicide resistance gene cassette flanked by short stretches of homology to the endochitinase locus yielded up to 10% targeted, homology-directed transgene integration precisely into the ZFN cleavage site. Given that ZFNs can be designed to recognize a wide range of target sequences, these data point toward a novel approach for targeted gene addition, replacement and trait stacking in plants.     

Field trial analysis of nitrate reductase co-suppression: a comparative study of 38 combinations of transgene loci

Co-syppression of host genes and 35S transgenes encoding nitrate reductase was previously reported in transgenic tobacco plants (Nicotiana tabacum cv. Paraguay or Burley) using either a full-length cDNA or fragments devoid of the 3 and/or 5 UTR. Co-suppression was previously shown to affect a limited fraction of the progeny of one transgenic tobacco line homozygous for a single transgene locus, and the phenomenon occurred at each generation. In this work, 38 combinations of transgene loci derived from 13 independent transgenic lines homozygous for a single transgene locus were field-tested under two different conditions in an attempt to determine the corresponding frequencies of co-suppression, i.e. the percentage of plants showing co-suppression.Each of the 13 homozygous lines exhibited a different frequency of co-suppression, ranging from 0% to 57%. High frequencies were found to be associated with transgene loci carrying a high number of copy of the transgene, suggesting a transgene dose effect. Combinations carrying 2 non-allelic transgene loci in a hemizygous state exhibited frequencies of co-suppression between those of each of the 2 transgene loci in a homozygous state, while combinations carrying 2 non-allelic transgene loci in a homozygous state exhibited frequencies of co-suppression higher than the sum of those of the 2 transgene loci alone in a homozygous state, clearly confirming a transgene dose effect.Co-suppression frequencies were increased when the plants were grown initially in vitro, suggesting some environmental effect. The roles of transgene copy number, number of transgene loci and environmental factors are discussed in the light of a threshold hypothesis.     

Maize DNA enrichment by representational difference analysis in a maize chromosome addition line of oat

 The recent recovery of maize (Zea mays L.) single-chromosome addition lines of oat (Avena sativa L.) from oat x maize crosses has provided novel source materials for the potential isolation of maize chromosome-specific sequences for use in genetic mapping and gene cloning. We report here the application of a technique, known as representational difference analysis (RDA), to selectively isolate maize sequences from a maize chromosome-3 addition line of oat. DNA fragments from the addition line and from the oat parent were prepared using BamHI digestion and primer ligation followed by PCR amplification. A subtractive hybridization technique using an excess of the oat parental DNA was employed to reduce the availability for amplification of DNA fragments from the addition lines that were in common with the ones from the oat parental line. After three rounds of hybridization and amplification, the resulting DNA fragments were cloned into a plasmid vector. A DNA library containing 400 clones was constructed by this method. In a test of 18 clones selected at random from this library, four (22%) detected maize-specific repetitive DNA sequences and nine (50%) showed strong hybridization to genomic DNA of maize but weak hybridization to genomic DNA of oat. Among these latter nine clones, three detected low-copy DNA sequences and two of them detected DNA sequences specific to chromosome 3 of maize, the chromosome retained in the source maize addition line of oat. The other eight out of the 13 clones that had strong hybridization to maize DNA detected repetitive DNA sequences or high-copy number sequences present on most or all maize chromosomes. We estimate that the maize DNA sequences were enriched from about 1.8% to over 72% of the total DNA by this procedure. Most of the isolated DNA fragments detected multiple or repeated DNA sequences in maize, indicating that the major part of the maize genome consists of repetitive DNA sequences that do not cross-hybridize to oat genomic sequences. Received: 18 November 1997 / Accepted: 12 March 1998     

Molecular and cytogenetic analyses of stably and unstably expressed transgene loci in tobacco.

To study the influence of genomic context on transgene expression, we have determined the T-DNA structure, flanking DNA sequences, and chromosomal location of four independent transgene loci in tobacco. Two of these loci were stably expressed in the homozygous condition over many generations, whereas the other two loci became unstable after several generations of homozygosity. The stably expressed loci comprised relatively simple T-DNA arrangements that were flanked on at least one side by plant DNA containing AT-rich regions that bind to nuclear matrices in vitro. Of the unstably expressed loci, one consisted of multiple incomplete T-DNA copies, and the second contained a single intact T-DNA; in both cases, however, binary vector sequences were directly contiguous to a right T-DNA border. Fluorescence in situ hybridization demonstrated that the two stably expressed inserts were present in the vicinity of telomeres. The two unstably expressed inserts occupied intercalary and paracentromeric locations, respectively. Results on the stability of transgene expression in F1 progeny obtained by intercrossing the four lines and the sensitivity of the four transgene loci to inactivation in the presence of an unlinked "trans-silencing" locus are also presented. The findings are discussed in the context of repetitive DNA sequences and the allotetraploid nature of the tobacco genome.     

Transgene behaviour across two generations in a large random population of transgenic rice plants produced by particle bombardment

The relationship between transgene copy number, rearrangement levels, inheritance patterns, expression levels, transgene stability and plant fertility was analysed in a random population of 95 independently transformed rice plant lines. This analysis has been conducted for both the selectable marker gene ( aphIV) and the unselected reporter gene ( gusA), in the presence or absence of flanking Matrix Attachment Regions (MARs) in order to develop a better understanding of transgene behaviour in a population of transgenic rice plants created by particle bombardment. In the first generation (T(0)), all the independently transformed plant lines contained and expressed the aphIV gene conferring resistance to hygromycin, but only 87% of the lines were co-transformed with the unselected gusA marker gene. Both transgenes seemed to be expressed independently. Most lines exhibited complex transgene rearrangements as well as an intact transgene expression unit for both aphIV and gusA transgenes. Transgene copy number was proportional to the quantity of DNA used during bombardment. In T(0) plants, high gusA copy number significantly decreased GUS expression levels but there was no correlation between expression level and transgene copy number across the entire population of lines. Four main factors impaired transgene expression in primary transgenic plants (T(0)) and their progeny (T(1)): (1) absence of transgene expression in T(0) plants (41% of lines), (2) sterility of T(0) plants (28% of lines), (3) non-transmission of intact transgenes to some or all progenies (at least 14% of lines), and (4) silencing of transgene expression in progeny plants (10% of lines). Transgene stability was significantly related to differences in transgene structure and expression levels. The presence of Rb7 MARs flanking the gusA expression unit had no effect on plant fertility or non-transmission of transgenes, but provided copy number-dependent expression of the transgene and improved expression levels and stability over two generations. Overall, only 7% of the plant lines without MARs and 17% of the lines with MARs initially generated, exhibited stable transgene expression over two generations.     

The mechanism of gene targeting in Physcomitrella patens: homologous recombination, concatenation and multiple integration

The model bryophyte Physcomitrella patens exhibits high frequencies of gene targeting when transformed with DNA constructs containing sequences homologous with genomic loci. ‘Targeted gene replacement’ (TGR) resulting from homologous recombination (HR) between each end of a targeting construct and the targeted locus occurs when either single or multiple targeting vectors are delivered. In the latter instance simultaneous, multiple, independent integration of different transgenes occurs at the targeted loci. In both single gene and ‘batch’ transformations, DNA can also be found to undergo ‘targeted insertion’ (TI), integrating at one end of the targeted locus by HR with one flanking sequence of the vector accompanied by an apparent non-homologous end-joining (NHEJ) event at the other. Untargeted integration at nonhomologous sites also occurs, but at a lower frequency. Molecular analysis of TI at a single locus shows that this occurs as a consequence of concatenation of the transforming DNA, in planta, prior to integration, followed by HR between a single site in the genomic target and two of its repeated homologues in the concatenated vector. This reinforces the view that HR is the major pathway by which transforming DNA is integrated in Physcomitrella.     

Site-specific gene integration in rice genome mediated by the FLP-FRT recombination system

Plant transformation based on random integration of foreign DNA often generates complex integration structures. Precision in the integration process is necessary to ensure the formation of full-length, single-copy integration. Site-specific recombination systems are versatile tools for precise genomic manipulations such as DNA excision, inversion or integration. The yeast FLP-FRT recombination system has been widely used for DNA excision in higher plants. Here, we report the use of FLP-FRT system for efficient targeting of foreign gene into the engineered genomic site in rice. The transgene vector containing a pair of directly oriented FRT sites was introduced by particle bombardment into the cells containing the target locus. FLP activity generated by the co-bombarded FLP gene efficiently separated the transgene construct from the vector-backbone and integrated the backbone-free construct into the target site. Strong FLP activity, derived from the enhanced FLP protein, FLPe, was important for the successful site-specific integration (SSI). The majority of the transgenic events contained a precise integration and expressed the transgene. Interestingly, each transgenic event lacked the co-bombarded FLPe gene, suggesting reversion of the integration structure in the presence of the constitutive FLPe expression. Progeny of the precise transgenic lines inherited the stable SSI locus and expressed the transgene. This work demonstrates the application of FLP-FRT system for site-specific gene integration in plants using rice as a model.     

Agrobacterium tumefaciens-mediated barley transformation

Genetically transformed barley was produced by eco-cultivating immature embryo explants with Agrobacterium tumefaciens carrying a binary vector coding for chimaeric bacterial genes, bar and gus , and selecting for bialaphos-resistant cultures from which plants were regenerated. Integration of both genes was confirmed by gel blot hybridization analysis of DNA from the transformed plants and their progenies. From 1282 embryos, plants were recovered for 54 independently transformed lines, giving a transformation efficiency of 4.2%. Transgene numbers in the different lines ranged from single copy insertion to at least ten copies. Sixteen out of 18 plants grown to maturity were fully fertile. Both marker genes, bar and gus , were expressed and co-segregated in the T₁ progeny plants. In the majority of cases, the genes showed Mendelian segregation predicted for transgene insertion at a single locus. In one family with multiple transgene insertions, molecular analysis of T₁ and T₂ plants suggested that the T-DNA had inserted at two unlinked loci.     

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.