Comparative analysis of the recently discovered hAT transposon TcBuster in human cells

Abstract

Background: Transposons are useful tools for creating transgenic organisms, insertional mutagenesis, and genome engineering. TcBuster, a novel hAT-family transposon system derived from the red flour beetle Tribolium castaneum, was shown to be highly active in previous studies in insect embryoes.

Methodology/principal findings: We tested TcBuster for its activity in human embryonic kidney 293 (HEK-293) cells. Excision footprints obtained from HEK-293 cells contained small insertions and deletions consistent with a hAT-type repair mechanism of hairpin formation and non-homologous end-joining. Genome-wide analysis of 23,417 piggyBac, 30,303 Sleeping Beauty, and 27,985 TcBuster integrations in HEK-293 cells revealed a uniquely different integration pattern when compared to other transposon systems with regards to genomic elements. TcBuster experimental conditions were optimized to assay TcBuster activity in HEK-293 cells by colony assay selection for a neomycin-containing transposon. Increasing transposon plasmid increased the number of colonies, whereas gene transfer activity dependent on codon-optimized transposase plasmid peaked at 100 ng with decreased colonies at the highest doses of transposase DNA. Expression of the related human proteins Buster1, Buster3, and SCAND3 in HEK-293 cells did not result in genomic integration of the TcBuster transposon. TcBuster, Tol2, and piggyBac were compared directly at different ratios of transposon to transposase and found to be approximately comparable while having their own ratio preferences.

Conclusions/significance: TcBuster was found to be highly active in mammalian HEK-293 cells and represents a promising tool for mammalian genome engineering.

Figure 1. Transposase-mediated excision of the TcBuster transposon in HEK-293 cells.

(a) An agarose gel of the excision PCR. Plasmid DNA was extracted from transfected HEK-293 cells and used as a template for nested PCR to detect the excision of the transposon DNA. Lane 1, 1 kb ladder; lane 2, PCR reaction without any DNA template added; lanes 3–7, PCR on extracts from cells transfected with either 1 µg of the transposon plasmid pTcBNeo (lane 3), 867 ng transposon pTcBNeo and 133 ng pCMVGFP negative control (lane 4), or 867 ng transposon pTcBNeo and 133 ng pXL-CMV-TcBuster_CO transposase plasmid (three separate transfections, lanes 5–7). (b) The three PCR bands shown in (a) were gel-purified and TOPO-cloned. Clones were sequenced to determine the exact excision junction. The sequence flanking the transposon in pTcBNeo is shown at the top. The TAAAG homology region is shown in blue. Mismatches are shown in lowercase pink. Dashes are used to maintain alignment and if pink, indicate a missing bp. The sequences are ranked according to (1) incidence (# of clones) followed by (2) number of bp not matching the highest incidence clone (# bp mismatches).

Figure 2. The effect of transposon and transposase plasmid dose on the number of drug-resistant colonies formed.

(a) HEK-293 cells in 6-well plates were transfected in triplicate with either 12.5 ng (light grey bars), 50 ng (dark grey bars), or 500 ng (black bars) of pTcBNeo carrying the neomycin-resistance transposon and 0 ng, 100 ng, 250 ng, or 500 ng of pCMV-TcBuster expressing the transposase. (b) HEK-293 cells in 6-well plates were transfected in triplicate with 500 ng of pTcBNeo plasmid carrying the neomycin-resistance transposon and the indicated amount of pCMV-TcBuster (0.5 ng, 1 ng, 5 ng, 10 ng, 25 ng, 50 ng, 100 ng, 250 ng, or 500 ng). In both a and b, pUC19 was used as filler DNA to increase the total amount of DNA transfected to 1 µg. Cells were diluted 1∶750 in selection media and grown for two weeks to allow drug-resistant cells to multiply and form colonies. The colonies were fixed, stained, and counted. The mean and standard error of the mean (SEM; n = 3) are shown.

Figure 3. The TcBuster transposase-related human proteins Buster1, Buster3 and SCAND3 cannot insert the TcBuster transposon into the mammalian genome.

HEK-293 cells were split onto 60 mm dishes and transfected with 1800 ng of pTcBNeo and 200 ng of Buster plasmid using FuGene6. The transfected cells (10 µl) were diluted into media containing G418 in 10 cm dishes and allowed to grow for two weeks. The resulting colonies were stained and counted (n = 3; error bars represent the standard error of the mean).

Figure 4. WebLogos generated from integration sites of piggyBac, TcBuster, and Sleeping Beauty transposons in HEK- 293 cells using the software at

weblogo.berkeley.edu .

Figure 5. High-throughput sequencing of transposon integration sites in HEK-293 cells.

(a) Integration frequency near selected genomic features. Integration site data sets for each transposon are indicated by the columns and the distance from the integration site to the genomic features by the rows. The departure from random distribution is indicated by colored tiles. Blue indicates that insertions are depleted compared to random whereas red indicates features where insertions are enriched compared to random and gray indicates that the distribution is random. Differences from random placement were scored using the ROC area method. (b) Integration frequency of Sleeping Beauty, piggyBac, and TcBuster near transcription start sites. Integration sites near transcription start sites were compiled onto a common transcription start site and the proportions mapped. The x-axis shows the distance from the transcription start sites and the y-axis shows the percentage of integration sites that were found for each range of distances.

Figure 6. Comparison of piggyBac, TcBuster, and Tol2 at varying transposase: transposon ratios.

HEK-293 cells in 6-well plates were transfected in triplicate with the indicated amount of transposase helper plasmid (+) or pUC19 (−) and plasmids carrying the neomycin-resistance transposon for each system. The amount of transposon plasmid transfected was 200 ng (a), 500 ng (b), or 900 ng (c). In c, the piggyBac (+) transposase plates had approximately 1000 colonies, too many to count (TMTC). Cells were diluted 1∶750 onto 10 cm plates in media containing geneticin for selection and incubated for two weeks to allow drug-resistant cells to form colonies that were fixed, stained and counted. Error bars represent the SEM (n = 3).

Figure 7. Transposon copy number analysis by real-time quantiative PCR (qPCR) for piggyBac, TcBuster, and Tol2.

(a) The number of copies of the transposon (Neo) per haploid genome (copy of RNaseP). There was no statistically significant difference between the transposons by the ANOVA test. Propagation of errors was used to combine errors inherent in the plates per group and replicates per plate. Error bars represent the standard error of the mean. (b) Standard curve showing the C(t) result vs number of copies of pTpB added. (c) Standard curve showing the C(t) result vs number of copies of pRNaseP added. Both sets of standards were performed in triplicate for each dilution. The equation for the line of best fit and R squared value are printed on each graph. The error bars represent the standard error of the mean.