| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Department of Entomology, Texas A&M University, College Station, Texas, USA; and
* Department of Radiology, Medical College of Georgia, Augusta, Georgia, USA
1Correspondence: Department of Entomology, Texas A&M University, College Station, Texas, USA. E-mail: ccoates{at}tamu.edu
SPECIFIC AIMS
Previous studies with chimeric transposases have resulted in suboptimal levels of transposition. In the present study we investigated the use of the Gal4 DNA binding domain as an N-terminal fusion with the MosI and piggyBac transposons to determine the effect on transpositional activity and ability to result in site-directed integration.
PRINCIPAL FINDINGS
1. Gal4-Mos1 and Gal4-piggyBac chimeric transposases are functional and catalyze transposition at rates comparable to unmodified transposases
The DNA binding domain of the Gal4 transactivator protein was used as an N-terminal protein fusion to generate Gal4-Mos1 and Gal4-piggyBac chimeric transposases. These constructs included a nuclear localization signal (NLS) to facilitate transportation to the nucleus and a flexible linker sequence to adequately separate the DNA binding domain from the transposase. Interplasmid transposition assays were performed in embryos of the yellow fever mosquito, Aedes aegypti. The rates of transposition were comparable to those from previous studies and were slightly higher in the case of the Gal4-Mos1 transposase (Table 1
). When a standard pGDV1 target plasmid was used, the sites of insertion were typical for assays using these elements and integrations occurred at and duplicated TA and TTAA target sites for the Mos1 and piggyBac elements, respectively. This demonstrates that the presence of the DNA binding domain as an N-terminal fusion did not interfere with the normal process of transposition for these transposons in terms of target sites and the frequency of transposition.
|
2. Interaction of the Gal4 DNA binding domain with UAS target sequences results in an increase in transpositional activity
In addition to the slight increase in transpositional activity used by the presence of the Gal4 DNA binding domain, a further large increase in transpositional activity resulted when a modified pGDV1 target plasmid was used that contained upstream activating sequences (UAS), the DNA binding site for the Gal4 DNA binding domain (Table 1)
. The Gal4-Mos1 chimeric transposase showed a 12.7-fold increase in transpositional activity in the presence of the UAS target. The Gal4-piggyBac chimeric transposase showed an 11.6-fold increase in transpositional activity in the presence of the UAS target. These results indicate that the Gal4-UAS interaction is stronger than that of the endogenous transposase DNA binding domains and thus results in a more efficient transposase target site interaction, such that the rate of transposition is increased.
3. Gal4-chimeric transposases catalyze integration in a site-directed manner
In addition to the increases seen in transposition frequencies, the transpositions catalyzed by the chimeric transposases occurred in a site-directed manner. Of the 53 transposition events recovered from assays using the Gal4-Mos1 chimeric transposase, 51 integrations occurred at a single TA target site in the pGDV1-UAS target plasmid (Fig. 1
). Insertions at this site had not been recovered from earlier transposition assays and all but one of the insertions occurred with the donor transposon in the same 5'-3' orientation. The two other insertions recovered from assays using the chimeric transposase occurred at TA target sites that had been inserted during earlier assays. Similarly, the assays performed with the Gal4-piggyBac chimeric transposase resulted in 45 of 67 recovered transposition products being inserted into a single TTAA target site, with 36 of these being in the same 5'-3' orientation. These results suggest that the Gal4-UAS interaction is limiting the choice of available target sites such that the transpositions were occurring predominantly in a site-directed manner, presumably through tethering of the transposase in close proximity to the UAS target.
|
CONCLUSIONS AND SIGNIFICANCE
Transgenic technologies enable the expression of an exogenous genetic construct in a living organism. Random integration of the transgene may result in insertional mutagenesis, transgene mis-expression, overexpression, or silencing. With available gene integration systems for insects, there are certain limitations that revolve around the degree of control over integration site specificity. Recent research of DNA-mediated transposition has opened new opportunities to improve transgenic technologies. As a part of an emerging technology, we developed chimeric transposases based on the Gal4-UAS system that have proved effective for mediating site-directed integration. This is the first evidence of site-specific integration using Mos1 and piggyBac chimeric transposases.
Since the normal specific target site sequences are very short, these transposons essentially integrate randomly into host genomes. However, all target sites are not used with equal frequency. As an example, the Sleeping Beauty transposase prefers certain TA target sites based on DNA structure surrounding the target. Sleeping beauty-mediated transposition is also strongly enhanced by CpG methylation. Interaction with host cell factors might also influence target site selection for certain transposases.
The Mos1 transposase was chosen in part because of its high degree of stability in the insect genome after integration. The helper plasmid was designed to express the Mos1 transposase under the regulation of an hr5-IE1 enhancer promoter, along with a nuclear localization signal (NLS) and Gal4 DNA binding domain (DBD). When the Gal4 binding domain binds to its target site, the tagged transposase is presumably brought into proximity and thereby increases the likelihood of transgene integration nearby. We verified the activity of the Gal4-Mos1 and Gal4-piggyBac chimeric transposases in the presence and absence of a Gal4-UAS interaction.
With the use of the Gal4-Mos1 chimeric transposase, the transposition efficiency was increased by 12.7-fold compared to controls where the UAS binding site is absent in the pGDV1 target plasmid (Table. 1)
. Increases in transpositional frequency presumably are a result of increased DNA binding affinity of the Gal4 DBD compared to the endogenous Mos1 DBD, even in the absence of a specific UAS target. Compared with other systems, the addition of an N-terminal fusion has not reduced transpositional activity.
In this system, Gal4-UAS-mediated transposition occurred almost exclusively at the 1061 site of the target plasmid. This site is 954 bp away from the UAS target site. Surrounding the 1061 site, from site 873 to 1157, there are 26 TA sites available for insertion, among which 8 (nucleotide positions 873, 922, 928, 932, 993, 1100, 1104, and 1157) have been inserted into earlier assays. It is perhaps surprising then that these sites were not used, suggesting that the Gal4-Mos1 transposase may be structurally limited after Gal4-UAS binding. It is reasonable to propose that the Gal4 DBD is binding to the UAS binding site and that the transposase donor element complex is located at some distance from this region by the linker peptide, resulting in insertion at the 1061 TA site (Fig. 1)
. Furthermore, 1061 is an unused target TA site with respect to previous assays, perhaps emphasizing the involvement of the chimeric transposase for the forcible insertion of the transgene. It is not known whether 1061 is a true cold spot, limited by DNA structure, or if by chance it had not previously been used for integration. If there is such a structural limitation, it is possible that a hot spot or more favorable TA target site sequence could be placed at an optimal distance and thus result in an even higher integration frequency. It should be noted that potential insertions into the chloramphenicol resistance gene would not be recovered due to the selection regime used for these assays.
In a parallel experiment using a Gal4-piggyBac chimeric transposase, the transposition frequency was 11.6-fold higher than controls, where the Gal4-UAS interaction is absent. In the presence of the chimeric transposase, 67% of transpositions occurred at the 1103 site of the target plasmid, which is an unused site in earlier reports of piggyBac insertions. There are five insertion sites (TTAA) close to the 1103 site (945, 968, 977, 992, and 1169). Among these, three have previously been used. Insertions primarily at 1103 suggest that the Gal4-piggyBac transposase might also be structurally limited after binding to the UAS site, and thus is forced to make an insertion in the same site. It cannot be determined whether the 1103 site is a true cold spot. It is worth noting that the two insertions sites used by the chimeric transposases are in close proximity and thus are at similar distances from the UAS target.
It is thus possible that chimeric transposases using artificial zinc finger domains could allow for integration into or around any genomic site. This represents a significant advantage over recombinases in that a prior germline integration event of an engineered target is not required, potentially allowing chimeric transposases to be utilized for somatic gene therapy.
Zinc finger domains are highly conserved DNA binding motifs that are involved in specific DNA binding and protein-protein interactions. A six-zinc finger domain protein will recognize an 18 bp target, long enough to constitute a rare address in the human genome. Independently folded domains offer a large number of combinational possibilities for DNA recognition. As an example, modifying the control of gene expression with specifically selected zinc finger combinations was used to inhibit an oncogene in a mouse cancer cell line. The site specificity of the Tn3 resolvase was recently modified by exchanging its natural DNA binding domain with the Zn finger domain of the human transcription factor, Zif268. The fusion of modifying enzymes with the DNA binding domain of Zif268 thus is promising, since Zif268 can theoretically be evolved to recognize any DNA sequence.
Similarly, approaches using zinc finger domains as artificial DBD will allow chimeric transposases to be used for the site-directed integration of therapeutic genes into precise genomic locations. The results presented here establish the potential for achieving site-directed integration of Mos1-and piggyBac elements into insect genomes. Furthermore, the recent demonstration of piggyBac activity in mammalian cells will allow these systems to be utilized in a diverse array of settings.
|
FOOTNOTES
To read the full text of this article, go to http://www.fasebj.org/cgi/doi/10.1096/fj.05-5485fje
This article has been cited by other articles:
|
|
 |
|
  N. Sharma, B. Moldt, T. Dalsgaard, T. G. Jensen, and J. G. Mikkelsen Regulated gene insertion by steroid-induced {Phi}C31 integrase Nucleic Acids Res., June 1, 2008; 36(11): e67 - e67. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. E. Morales, V. H. Mann, K. J. Kines, G. N. Gobert, M. J. Fraser Jr, B. H. Kalinna, J. M. Correnti, E. J. Pearce, and P. J. Brindley piggyBac transposon mediated transgenesis of the human blood fluke, Schistosoma mansoni FASEB J, November 1, 2007; 21(13): 3479 - 3489. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. C.-Y. Wu, Y.-J. J. Meir, C. J. Coates, A. M. Handler, P. Pelczar, S. Moisyadi, and J. M. Kaminski From the Cover: piggyBac is a flexible and highly active transposon as compared to Sleeping Beauty, Tol2, and Mos1 in mammalian cells PNAS, October 10, 2006; 103(41): 15008 - 15013. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C. Feschotte The piggyBac transposon holds promise for human gene therapy PNAS, October 10, 2006; 103(41): 14981 - 14982. [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
) and solid arrow represent the predominant site hit by Mos1 donor elements when using the pGDV1-UAS target plasmid. As indicated above the solid arrow, this site was hit 51 times during these assays. Two additional insertions recovered when using the pGDV1-UAS target plasmid are indicated by up and down arrowheads. Both experiments shown here utilized the chimeric Gal4-Mos1 transposase helper plasmid. For clarity, the 68 insertions recovered from the assays using the unmodified Mos helper plasmid are not shown; however, they show a distribution similar to the open and solid circles. The numbers represent the nucleotide location on the pGDV1 plasmid. The UAS target site is shown as a shaded box, the Chloramphenicol (Cam) resistance gene as an open box. Insertions into the antibiotic resistance gene would not be recovered from this assay due to the selection regime used to remove false positives and recombination events.