Compensating for over-production inhibition of the Hsmar1 transposon in Escherichia coli using a series of constitutive promoters
Transposable elements (TEs) are a diverse group of self-mobilizing DNA elements. Transposition has been exploited as a powerful tool for molecular biology and genomics. However, transposition is sometimes limited because of auto-regulatory mechanisms that presumably allow them to cohabit within their hosts without causing excessive genomic damage. The papillation assay provides a powerful visual screen for hyperactive transposases. Transposition is revealed by the activation of a promoter-less lacZ gene when the transposon integrates into a non-essential gene on the host chromosome. Transposition events are detected as small blue speckles, or papillae, on the white background of the main Escherichia coli colony.
We analysed the parameters of the papillation assay including the strength of the transposase transcriptional and translational signals. To overcome certain limitations of inducible promoters, we constructed a set of vectors based on constitutive promoters of different strengths to widen the range of transposase expression. We characterized and validated our expression vectors with Hsmar1, a member of the mariner transposon family. The highest rate of transposition was observed with the weakest promoters. We then took advantage of our approach to investigate how the level of transposition responds to selected point mutations and the effect of joining the transposase monomers into a single-chain dimer.
We generated a set of vectors to provide a wide range of transposase expression which will be useful for screening libraries of transposase mutants. The use of weak promoters should allow screening for truly hyperactive transposases rather than those that are simply resistant to auto-regulatory mechanisms, such as overproduction inhibition (OPI). We also found that mutations in the Hsmar1 dimer interface provide resistance to OPI in bacteria, which could be valuable for improving bacterial transposon mutagenesis techniques.
KeywordsPapillation assay Hsmar1 Overproduction inhibition SETMAR Transposase Transposable elements
Inverted terminal repeat
Ribosome binding site
Transposable elements (TEs) are DNA sequences with the ability to move from one place to another in the genome. They are found in virtually all organisms and are particularly numerous in higher eukaryotes where they can represent a significant percentage of the genome [1, 2, 3]. Originally thought of as selfish elements that provide no advantage to the host, TEs have now been shown to be important drivers of genome evolution [4, 5]. Indeed, TEs can provide novel transcription factor binding sites, promoters, exons or poly(A) sites and can also be co-opted as microRNAs or long intergenic RNAs [6, 7, 8]. TEs are a diverse group of DNA sequences using a wide range of mechanisms to transpose within their hosts. One particular mechanism prevalent in eukaryotes, and used by the mariner family, is known as “cut-and-paste” transposition . Over the past several years, our group and others have described the mechanisms regulating the transposition rate of different mariner transposons, such as Himar1, Hsmar1 or Mos1 [10, 11, 12, 13, 14, 15]. In Hsmar1, a regulatory mechanism was first recognized because of the phenomenon of overproduction inhibition (OPI) . The mechanism of OPI was eventually explained by the realization that double occupancy of the transposon ends with transposase dimers blocks assembly of the transpososome . Thus, OPI curbs Hsmar1 transposition rate to avoid damaging the host genome by excessive transposition . This mechanism will apply to any transposon in which a transposase multimer binds one transposon end and then recruits the second end as naked DNA.
OPI represents a limitation in the development of hyperactive transposases for biotechnological applications. Several approaches such as modifying the binding kinetics of the transposase to the inverted terminal repeat (ITR) or the monomer-dimer equilibrium can be used to overcome OPI. Indeed, we and others previously showed that most mutations in the conserved WVPHEL motif, in Himar1 and Hsmar1, result in hyperactive transposases but at the cost of producing non-productive DNA double-strand breaks and therefore DNA damage [17, 18].
A limitation of the papillation assay is that it generally employs a transposase gene whose expression is under the control of an inducible promoter which cannot be finely regulated. We have constructed a set of vectors maintained at single copy or at ~ 13 copies per cell that carry various constitutive promoters in the absence or presence of a ribosome binding site (RBS). This set of vectors allows transposase expression across a wide range of expression levels facilitating the screening of hyperactive and/or OPI-resistant transposases. We used this set of vectors to compare an Hsmar1 transposase monomer to a single-chain dimer and to test for hyperactivity and OPI-resistance in several Hsmar1 transposase mutants. We found that one Hsmar1 mutant in the dimer interface, R141L, is resistant to OPI in E. coli.
Results and discussion
Characterization of the papillation assay using a strong inducible promoter
The papillation assay provides a visual assessment of the transposition rate, which is dependent on the concentration and activity of the transposase [12, 20]. We defined the level of transposition as the average number of papillae per colony after five days of incubation at 37 °C. In the previous papillation assay, the transposase was provided by the protein expression vector pMAL-c2x under the control of a Ptac promoter and was fused to the C-terminus of the maltose binding protein . We first characterized the papillation assay using the Hsmar1 transposase cloned downstream of the inducible Ptac promoter and investigated the effect of different concentrations of IPTG and lactose on the transposition rate (Fig. 1b and c). The Ptac promoter can be induced by IPTG or by allolactose, which is metabolized from lactose by lacZ . Therefore, lactose will not induce the Ptac promoter until a successful transposition event that leads to a lacZ+ cell occurred. Also, we investigated whether the presence of the MBP-tag affects the transposition rate (Fig. 1b and c). In the absence of transposase, the number of papillae per colony in all the conditions tested was between zero and three (Fig. 1b, no transposase column, and 1C). In presence of the transposase or MBP-transposase (middle and right columns, respectively), the number of papillae per colony varies with the concentration of IPTG and lactose.
Transposition frequencies of MBP-tagged or untagged Hsmar1 transposase
Ptac MBP-Hsmar1 transposase
1.99 (±0.43) × 10−6
Ptac Hsmar1 transposase
2.58 (±0.02) × 10− 5
Any further increase in the IPTG concentration results in a decrease of the transposition rate, consistent with the effects of OPI, which has been described for Hsmar1 in vitro, in E. coli, and in HeLa cells [12, 19]. Interestingly, the presence of the MBP tag affects the transposition rate of the transposase, potentially through its stabilization. We therefore decided to use untagged Hsmar1 transposase for the remaining experiments.
SETMAR transposition activity was lost during the same period as Hsmar1 transposase domestication
This result supports an absence of conservation of Hsmar1 transposase activity during SETMAR evolution, in agreement with recent studies which did not observe an in vivo nuclease activity of SETMAR in DNA repair assays [27, 28]. Two of the DNA binding mutants, E2K and R53C, are deleterious to Hsmar1 transposition activity in a papillation assay. It will be interesting to determine whether this effect is mediated through a change in ITR binding efficiency, which could have modified SETMAR’s ability to bind ITRs in the genome and therefore its emerging functions in regulating gene expression .
Papillation assay with a featureless DNA constitutive promoter
The EE- promoter-UTR sequence is not necessarily the highest level of transposition attainable as EE+ and EE++ might already be in OPI because of the higher translation efficiency. We therefore explored transcriptional activity with a series of progressively degraded PL-λ promoters that had been selected from a mutant library for their lack of stochastic cell-to-cell variation .
Characterization of the set of constitutive promoters
List of constitutive promoters
Average promoter metric
We first investigated the strongest non-FLAG-tagged expression vectors by performing western blots with an anti-Hsmar1 antibody (Fig. 4b). We also compared by western blotting these constructs with the Ptac inducible promoter previously used for papillation assay (Fig. 4b). Interestingly, two of our constructs (Ip5++ and Ip6++) produce a higher amount of Hsmar1 transposase than the Ptac promoter fully induced with 1 mM of IPTG.
We next quantified the strength of each expression vector by inserting an EGFP gene in each FLAG-tagged vector to investigate fluorescence levels by flow cytometry (Additional file 1: Figure S3). To determine the strength of the expression vectors, we normalized their geometric mean fluorescence value against the strongest vector, Ip6++ (Fig. 4c). Most of the single-copy expression vectors and the RBS- promoters produce an amount of EGFP fluorescence close to the background level. However, all of the ~ 13-copy expression vectors with a consensus RBS produce more fluorescence than their respective single-copy vectors. A comparison of the EGFP produced by the p5 and p6 promoters shows that the pIncQ expression vectors produced around 14 times more fluorescence than the pBAC expression vectors, indicating a potential copy number of 14 for pIncQ, which is in line with the literature . Also, the vectors with a consensus RBS produce an amount of fluorescence correlating with the promoter strength originally determined by Alper and colleagues . In contrast, all of the vectors without a RBS motif, except Ip6-, produce a fluorescence level close to the detection threshold (Fig. 4d). Similarly, the pEE promoter is also too weak to change the amount of fluorescence produced whether the RBS is present or absent.
Characterization of the papillation assay with the wild-type Hsmar1 transposase
Prior to the determination of the rate of transposition of each expression vector with the papillation assay, we visually determined the best conditions for this assay using the Ip3++ expression vector, which has a strength similar to Ptac induced with 0.1 mM IPTG (Fig. 4b) and will therefore have a limited number of papillae per colony, and a range of lactose concentrations (Additional file 1: Fig. S4). Similarly to the Ptac promoter, we observed a correlation between the number of papillae per colony and the lactose concentration (Additional file 1: Figure S4A and B). We decided to work at 0.1% lactose since it represents the best trade-off between the number of papillae per colony and the size of the papillae for quantitation at high transposition rate. To confirm that it is a lactose-specific effect, we performed papillation assays with the Bp2++ expression vector and a 0.1% concentration of different sugars: no sugar, glucose, maltose, lactose, and maltose plus lactose (Additional file 1: Figure S5). Importantly, the activity of the constitutive expression vectors is independent of the absence or presence of sugars. A higher number of papillae was only observed when lactose was added, indicating that lactose allows the late transposition events to become visible as only lacZ+ cells are able to metabolize it.
Similarly to the effect of the MBP-tag on transposition (Fig. 1), the presence of the 3x FLAG-tag also modifies the level of transposition (compare the RBS++ expression vectors from Figs. 5 and 6). However, it remains unclear how the presence of a tag affects the number of papillae per colony but it could be mediated by a change in transposase stability.
Covalently linking two Hsmar1 monomers in a dimer affects the transposition rate
When compared to the results obtained with the Hsmar1 monomer, the single chain dimer transposition rate peaks at a different set of expression vectors, Bp2- and Bp3- for the covalent dimer and Ip2- for the monomer (Fig. 7b). This might indicate that Bp2- and Bp3- are weaker expression vectors than Ip2-. We do not observe any difference in the number of papillae per colony with stronger expression vectors such as Ip3++ and Ip6++ (Fig. 7a and b). This indicates that a single chain Hsmar1 dimer is as sensitive to OPI as the Hsmar1 monomer.
Mutations in Hsmar1 dimer interface produce hyperactive mutants in bacteria
Transposition frequencies of two Hsmar1 transposase mutants expressed at optimal and high level
4.73 (±1.02) × 10−5
9.73 (±4.53) × 10−4
2.42 (±1.68) × 10− 4
3.22 (±1.02) × 10− 7
5.79 (±2.63) × 10−5
3.24 (±1.43) × 10−4
This study provides a set of expression vectors based on constitutive promoters to investigate the phenotypes of mutant transposase. It will be useful to distinguish between true hyperactive mutants and defective mutants that happen to be resistant to OPI. Compared to inducible promoters, our set of expression vectors provides a wide range of consistent transposase expression levels between individual cells. In addition to the characterization of the constitutive promoters, we also found one Hsmar1 mutation, R141L, which is OPI-resistant in E. coli and could therefore prove useful for improving bacterial transposon mutagenesis with mariner elements. Another approach in controlling the transposition rate is to use a single chain Hsmar1 dimer, which allows transposition to occur after a single translation event and would therefore permit the usage of a weak promoter with a weak RBS.
We believe our set of expression vectors will be useful or the study of other transposons and in the screening of libraries for finding hyperactive and/or OPI-resistant transposases. For transposons other than Hsmar1, the expression will have to be tuned to the system as different transposons will have different relationship between transposase concentration and transposition rate. A medium copy vector (pIncQ) with a medium promoter (p4) would be an ideal starting point. The expression can then be tuned by progressive degradation of the RBS.
Media and bacterial strains
Bacteria were grown in Luria-Bertani (LB) media at 37 °C. The following antibiotics were used at the indicated concentrations: ampicillin (Amp), 100 μg/ml), chloramphenicol (Cm), 25 μg/ml, and spectinomycin (Spec), 100 μg/ml. The following E. coli strains were used: RC5024 (identical to DH5α) [endA1 hsdR17 glnV44 thi-1 recA1 gyrA relA1 Δ (lacIZYA-argF)U169 deoR (φ80dlac Δ (lacZ)M15)], RC5094 [F- araD139 Δ (argF-lac)U169 rspL150 relA1 flbB5301 fruA25 deoC1 ptsF25 rpoS359::Tn10], RC5096 [F− fhuA2 Δ (lacZ)r1 glnV44 e14-(McrA-) trp-31 his-1 rpsL104 xyl-7 mtl-2 metB1 Δ (mcrC-mrr)114::IS10 argE::Hsmar1-lacZ’-kanR] and RC5097 (= RC5096 pOX38::miniTn10-CAT).
List and DNA sequences of constitutive promoters
Average promoter metric
Expression plasmids were built by cloning the EGFP or Hsmar1 gene in pBACe3.6, pGHM491, and pMAL-c2X (New England Biolabs) between NdeI and BamHI restriction endonuclease sites. A list of the plasmids used in this study can be found in Additional file 2: Table S1. The DNA sequences of the vectors based on pBACe3.6 and pMAL-c2X can be found in Additional file 3: Table S2. The DNA sequence of pGHM491 is unknown and therefore the DNA sequences of the vectors based on it are absent from Additional file 3: Table S2. Plasmids pRC880 and pRC1721 encode the wild-type transposase in pMAL-c2X in presence and absence of the MBP tag, respectively (Fig. 1). Plasmids pRC1782–1807 encode EGFP downstream of pEE to p6, with RBS-, RBS+, and RBS++, in pBACe3.6 and pGHM491 (Fig. 4). Plasmids pRC1723–1728 and pRC1730–1735 encode untagged Hsmar1 downstream of pEE to p6, with RBS+ and RBS++, in pBACe3.6 and pGHM491 (Figs. 4 and 5). Plasmids pRC1821–1846 encode FLAG-tagged Hsmar1 downstream of pEE to p6, with RBS-, RBS+, and RBS++, in pBACe3.6 and pGHM491 (Figs. 3 and 6). Plasmids pRC1877 to pRC1899 are derived from pMAL-c2X and encode the different Hsmar1 mutants with the mutations found in SETMAR (Fig. 2). Plasmids pRC1858–1861, 1863, 1865, 1866, 1868–1871, 1873, 1875, and 1876 encode the Hsmar1 monomer and Hsmar1 single chain dimer in Bp2-, Bp3-, Bp3++, Bp6++, Ip2-, Ip3++, and Ip6++ (Fig. 7). Plasmids pRC1739, 1740, 1746, 1747, 1752, and 1753 encode Hsmar1 F132A and R141L mutants cloned into Bp-EE+, Ip-EE+, and Ip6++ (Fig. 7).
RC5096 cells expressing EGFP were grown overnight at 37 °C in LB medium supplemented with chloramphenicol or spectinomycin. The cultures were diluted in a 1:1000 ratio in fresh LB medium complemented with antibiotics and grown to mid-log phase (OD600 ~ 0.5). The cells were pelleted at 6000 g for 5 min, washed in 1X PBS twice, and resuspended in 500 μl of 1X PBS. Flow cytometry analysis was performed on 100,000 cells with a Beckman Coulter Astrios EQ. The FlowJo software was used to analyse the data (gating of the EGFP positive cells and acquisition of the geometric mean and the number of GFP positive cells) and to construct the overlayed plots. The number of GFP positive cells can be found in Additional file 4: Table S3.
Cells containing a derivative of pMAL-c2x were grown in LB supplemented with 100 μg/ml of ampicillin at 37 °C until an OD600 of ~ 0.5 and were then induced with the required concentration of IPTG for 2 h at 37 °C. Cells containing pBACe3.6 or pGHM491 derivatives were grown in LB supplemented with respectively 100 μg/ml of spectinomycin or 50 μg/ml of chloramphenicol at 37 °C for the same amount of time as the induced cells. Promoters’ expression was analysed by pelleting ~ 1.5 × 109 cells. The samples were resuspended in SDS sample buffer, boiled for 5 min, and loaded on 10% SDS-PAGE gels. Proteins were transferred to PVDF membrane, probed with an anti-SETMAR antibody raised against the amino acids 658–671, which correspond to the domesticated Hsmar1 (goat polyclonal, 1:500 dilution, ab3823, Abcam) followed by a horseradish peroxidase-conjugated anti-goat secondary antibody (rabbit polyclonal, 1:5000 dilution, ab6741, Abcam). Proteins were visualized by using the ECL system (Promega) and Fuji medical X-ray film (Fujufilm).
The papillation assay and the reporter strain RC5096 have been described previously (Fig. 1a) . Briefly, transposase expression vectors were transformed into the RC5096 strain. It is a lac− E. coli strain encoding a transposon containing a promoter-less lacZ and a kanamycin resistance gene flanked with Hsmar1 ends, which has been integrated in a silent genomic locus. In the absence of transposase, the strain produces white colonies on X-gal indicator plates. When the transposase is supplied in trans, the integration of a transposon into the correct reading frame of an active gene will produce a lacZ fusion protein. The descendants of this cell will become visible as blue papillae on X-gal indicator plates. RC5096 transformants were plated on LB-agar medium supplemented with different concentrations of lactose (or other sugars), 40 μg/ml of X-gal and either 50 μg/ml of chloramphenicol or 100 μg/ml of spectinomycin. Plates were incubated 5 days at 37 °C and photographed. The transposition rate is determined by the number of papillae per colony. Papillation assays were performed in biological duplicates.
A chloramphenicol resistant derivative of the conjugative plasmid pOX38 has been introduced in the RC5096 papillation strains to create the donor strains RC5097. Briefly, RC5097 transformants and the recipient strain, RC5094, were grown overnight in LB supplemented with antibiotics at 37 °C. The next day, respectively one and three volumes of RC5097 and RC5094 were centrifuged for 5 min at 6000x g. Each pellet was resuspended in 3 ml of fresh LB, pooled together, and incubated in a shaking water bath for 3 h at 37 °C. After the mating, the transposition events were detected by plating 200 μl of each culture on LB-agar medium supplemented with tetracycline and kanamycin. The number of transconjugants was obtained by plating a 10− 5 fold dilution of each culture on LB-agar medium supplemented with tetracycline and chloramphenicol. The plates were incubated overnight at 37 °C and the transposition rate determined the next day by dividing the number of kanamycin-resistant colonies by the number of chloramphenicol resistant colonies.
We would like to thank Michael Chandler for his comments on the manuscript. We also thank David Onion from the University of Nottingham Flow Cytometry facility for help with flow cytometry analyses.
Performed the experiments: MT. Conceived and designed the experiments and analysed the data, MT, RC. Wrote the paper: MT, RC. All Authors read and approved the final version the manuscript.
This work was supported by the Biotechnology and Biological Sciences Research Council [grant number BB/J014508/1]. Funding for open access charge: BBSRC [BB/J014508/1].
Ethics approval and consent to participate
Consent for publication
The authors declare that they have no competing interests.
Open AccessThis article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated.