MiniUIB, a Novel Minitransposon-Based System for Stable Insertion of Foreign DNA into the Genomes of Gram-Negative and Gram-Positive Bacteria

Abstract

Transposition of the insertion sequence (IS) ISPpu12 is actively induced after conjugative interaction. The transposase of this IS can act in trans on structures flanked by inverted repeats similar to those of the transposon. Based on that fact, an ISPpu12-based minitransposon, miniUIB, has been constructed in order to biotechnologically exploit the self-regulation of ISPpu12 and its increased activity after conjugative interaction. Mobilization of the miniUIB structure into the genome of Pseudomonas stutzeri AN10 after conjugative interaction was demonstrated. A single gene, i.e., the kanamycin resistance determinant, or large genetic structures of >12 kb, i.e., alkBFGHJKL and alkST operons of Pseudomonas putida TF4-1L (GPo1), have been easily integrated in P. stutzeri AN10 by an RP4-based delivery system. Therefore, the integration of the alk determinants by use of the miniUIB system has extended the biodegradation capabilities of this strain. Plasmid pJOC100, containing the transposase and regulator genes of ISPpu12 adjacent to the miniUIB structure, was constructed in order to extend the host range of this biotechnologically useful genetic tool to other model and real-world bacteria. The effectiveness of the system for random mutagenesis in a phylogenetic wide range of bacteria and for the insertion of novel functions has been demonstrated, even in successive steps.

INTRODUCTION

Minitransposon-based insertion systems have proven useful for inserting DNA fragments into a host chromosome. These biotechnological tools consist in nucleic acid sequences flanked by the inverted repeats (IR) of a transposon. These IRs are recognized by the transposase, an enzyme encoded outside of the mobile structure, mobilizing the IR-IR structure to another DNA region or replicon. Since the transposase gene is not mobilized with the minitransposon, the outcome is a stable insertion.

As revealed by literature, the developed minitransposon systems mini-Tn5 and mini-Tn10 (1, 2) are by far the most popularly used mainly for random transposon-mutagenesis (up to 975 cites according to the ISI Web of Knowledge at the end of September 2012). Their biotechnological use lies on the simplicity of their genetic structure together with the R6K-based suicidal plasmid (3) in RP4 conjugation delivery-based Escherichia coli λ_pir strains (2, 4). Since their description, several interesting innovations have been applied to these systems, such as the addition of novel resistance determinants (5), the insertion of an R6K replicase for rapid identification of the flanking region of insertion (6), or the introduction of the sacB marker to force the loss of antibiotic-resistance determinants by homologous recombination (7). However, the instability of these minitransposon structures in the absence of an insert between the IRs requires the use of intermediate plasmids in order to perform the constructions (2). Also, in some recipient strains, transposition frequencies of these systems are as low as the frequency of cointegration of the whole vector by homologous recombination (2), not allowing to easily discern true transposition events.

An alternative to the random insertion of mini-Tn5 are the mini-Tn7 minitransposons, which specifically integrates at the attTn7 site, located downstream the highly conserved glmS gene (8). It has been proven that the Tn7-based system is a useful tool to integrate DNA sequences into the chromosomes of different Gram-negative bacteria (9). In this sense it might be a better choice than random strategies (i.e., miniTn5) because the insertion site is known and, consequently, it is not necessary to evaluate the fitness of the recombinant (9). However, the absence of the transposition target, or even the presence of a badly conserved transposition target, could complicate obtaining recombinants. In addition, it is well known that Tn7 displays transposition immunity (i.e., Tn7 does not integrate into DNA molecules already harboring Tn7) (10). This phenomenon may complicate the use of miniTn7-derivatives in bacteria harboring Tn7 and may prevent a second integration event with the same Tn7-based system, if necessary.

ISPpu12 is an ISL3-like IS that was first described in the TOL plasmid pWW0 of Pseudomonas putida mt-2 (11). In a recent study, we showed that ISPpu12 was also present in P. stutzeri AN10 (12), in addition to another ISL3-like IS named ISPst9 with similar flanking IRs. Both ISL3-like ISs transpose use a cut-and-paste transposition mechanism by means of circular intermediates (12, 13). In addition, we showed that ISPpu12 was responsible of the transposition activation of these two ISL3-like ISs that occur in P. stutzeri AN10 cells that undertake conjugative interaction (interaction with a conjugative strain such as E. coli S17-1, probably mediated by the conjugative pili, with or without horizontal gene transfer) (13). Thus, we demonstrated that activation of transposition was due to TnpR, the transcriptional regulator encoded between the IRs of ISPpu12, and that the in trans activity of ISPpu12 transposase on ISPst9 was responsible for ISPst9 transposition in response to that stimulus (12). In fact, ISPst9 was not able to activate its own transposition after conjugative interaction because it lacks the regulatory tnpR gene (12).

We describe here the construction of the miniUIB minitransposon and a set of miniUIB-based tools designed to exploit biotechnologically the self-regulation of ISPpu12 and its increased activity after conjugative interaction. This reliable and ready-to-use system has proven effective for inserting genetic sequences in recipients compatible with the RP4 conjugation process.

MATERIALS AND METHODS

Bacterial strains, media, and culture conditions.

P. stutzeri strains AN10 (Ap^s Cm^s Km^s Sm^s nah⁺ Alk⁻ ISPpu12⁺) and AN11 (Km^s nah⁺ Alk⁻ ISPpu12-free) (14), Klebsiella pneumoniae CMD1 (15), and P. putida IS⁻, an ISPpu12-free derivative of P. putida mt-2 (16) obtained in our laboratory, were used as recipients of the minitransposons used in the present study. A total of 50 phylogenetically distinct environmental isolates from our collection of 292 previously isolated strains (17) from polluted beach-sand samples after the Prestige oil spill (Galicia, Spain) were also used. E. coli S17-1λ_pir (2) was used for hosting pGP704 plasmid (4) and its derivatives (Table 1) and for RP4-based conjugations. Conjugations were carried out as described previously (13). Briefly, aliquots of late-exponential-phase cultures of donor and recipient strains were spotted together (cell ratio of 1:1) onto a membrane filter (nitrocellulose, 0.22 μm; Millipore) and incubated at 30°C for 7 h (or 30 min, when indicated in the text) on the surface on a Luria-Bertani (LB) agar plate. P. putida TF4-1L (GPo1) (21) was used as source of alkane degradation determinants (alkBFGHJKL and alkST). When not specified, strains were cultured on LB broth at 30°C. Mineral basal medium (MBM) (22) supplemented with succinate 0.5% (wt/vol) and the appropriate antibiotics (ampicillin [Ap], 100 μg/ml; kanamycin [Km], 50 μg/ml; chloramphenicol [Cm], 40 μg/ml; and/or streptomycin [Sm], 50 μg/ml) were used for transconjugant selection. MBM supplemented with n-octane and n-hexadecane, at 2% (wt/vol) each, was used for growing alk⁺ strains.

Table 1.

Plasmids, primers, and probes used in this study

Plasmid, primer, or probe	Description or sequence (5′–3′)	Source or reference
Plasmids
pBBR1MCS-1	Cmr; broad-host-range cloning vector	18
pBC SK(−)	Cmr; ori pBBR322	Stratagene
pCSI1	Apr Cmr; source of the Cmr gene used in this study	19
pCSI2	Apr Kmr; source of the Kmr gene used in this study	19
pGP704	Apr; ori R6K, mob RP4	4
pGP704Km	Apr Kmr; pGP704-derivative containing the Kmr gene at the EcoRI site of its MCS	This study
pKNG101	Smr; ori R6K, mob RK2	20
pJOC21	Apr; pGP704-derivative containing miniUIB (IRL-MCS-IRR of ISPst9)	This study
pJOC22Km	Apr Kmr; pJOC21-derivative containing miniUIB-Km (1.6 kb, IRL-Kmr-IRR)	This study
pJOC22Cm	Apr Cmr; pJOC21-derivative containing miniUIB-Cm (2.1 kb, IRL-Cmr-IRR)	This study
pJOC22BC	Apr Cmr; pJOC21-derivative containing miniUIB-BC (3.6 kb, IRL-pBC SK-IRR)	This study
pJOC22KmBC	Apr Cmr Kmr; pJOC21-derivative containing miniUIB-KmBC (5.0 kb, IRL-Kmr-pBC SK-IRR)	This study
pJOC22KNG	Apr Smr; pJOC21-derivative containing miniUIB-KNG (7.0 kb, IRL-pKNG101-IRR)	This study
pJOC22KmKNG	Apr Kmr Smr; pJOC21-derivative containing miniUIB-KmKNG (8.4 kb, IRL-Kmr-pKNG101-IRR)	This study
pJOC22BCKmKNG	Apr Kmr Smr; pJOC21-derivative containing miniUIB-BCKmKNG (11.8 kb, IRL-pBC SK-Kmr-pKNG101-IRR)	This study
pALKB	Apr; pJOC21-derivative containing miniUIB-ALKB (8.5 kb, IRL-alkBFGHJKL-IRR)	This study
pALK	Apr; pJOC21-derivative containing miniUIB-ALK (12.8 kb, IRL-alkST-alkBFGHJKL-IRR)	This study
pJOC100	Apr; pJOC21-derivative containing miniUIB100 (IRL-MCS-IRR-ISPpu12 without IRs)	This study
pJOC100Km	Apr Kmr; pJOC100-derivative containing miniUIB100-Km (IRL-Kmr-IRR-ISPpu12 without IRs)	This study
pJOC100ALK	Apr; pJOC100-derivative containing miniUIB100-ALK (IRL-alkST-alkBFGHJKL-IRR-ISPpu12 without IRs)	This study
Primers
KM-F	AAA CGT CTT GCT CGA GGC C	This study
KM-R	GGA GAA AAC TCA CCG AGG C	This study
PCR3-F	GAG ATC TTC GGG TAT GCG GAT TTA ATG	This study
PCR3-R	GAT CTA GAC CCG GGC TAT TGT CAA GAC AG	This study
PCR1-F	CAG GTA CCG CAT GAC CGA AAT GCC CGA	This study
PCR1-R	GGA ATT CGT GGG TAT GCG GAT TTA ATG G	This study
ALK1-F	AAT CTA GAT TTC CAG CAG ACG ACG GAG C	This study
ALK1-R	CGC GCC GAG CTC CAG CGT TGT CC	This study
ALK2-F	GGC GAG TAC CAG GAC GGC GTA GGG	This study
ALK2-R	TTG AGC TCT TAG AAA ACA TAT GAC GCA CC	This study
ALKRsma-F	AAC CCG GGG CAC GTA CGG AGT GCG GG	This study
ALKRsma-R	AAC CCG GGG GCG AAG GCC GAA GTC GGC	This study
ISPPU25Eco-F	TTG AAT TCT TAC CTT TGC ATG AGA GTG AG	This study
ISPPU3348Eco-R	TTG AAT TCG GGG CAC CTT CAC CCC ATC	This study
Probes
TNPA	tnpA of ISPst9 and ISPpu12	13
TNPR	tnpR of ISPpu12	12
KM	Kmr gene of pCSI2, PCR using KM-F and KM-R primers	This study
ALK	alkBFGH genes of P. putida TF4-1L (GPo1), PCR using ALK1-F and ALK1-R primers	This study
PGP	Entire EcoRI-digested pGP704 plasmid	This study

^aCm^r, chloramphenicol resistance; Ap^r, ampicillin resistance; Sm^r, streptomycin resistance; Km^r, kanamycin resistance.

DNA manipulations.

All DNA manipulations were done as previously described (12, 13). Restriction endonuclease digestions (Roche and GE Healthcare) and ligations with T4 DNA ligase (Invitrogen) were performed as recommended by each manufacturer. PCR amplifications were done using Taq DNA polymerase (GE Healthcare). Plasmid DNA was isolated by alkaline lysis using the QIAprep spin miniprep kit (Qiagen). Southern blot hybridization against EcoRI-digested genomic DNA was performed as described previously (23). Enhanced chemiluminescence direct labeling (ECL direct nucleic acid labeling and detection system; GE Healthcare) was used for hybridization.

Primers and probes used in the present study are listed in Table 1. Probes, their targets, and the procedure for their synthesis were as follows: TNPA, for the detection of tnpA genes of both ISPst9 and ISPpu12 ISs, obtained as previously described (13); TNPR, for the detection of tnpR gene of ISPpu12, obtained as previously described (12); KM, for the detection of Km^r-gene, obtained by PCR amplification using primers KM-F and KM-R on plasmid pCSI2 (19); ALK, for alkBFGH detection, obtained by PCR amplification using primers ALK1-F and ALK1-R using genomic DNA obtained from P. putida TF4-1L (GPo1); and PGP, for pGP704 detection, obtained by direct DNA labeling of EcoRI-digested plasmid.

Plasmids and miniUIB-derivatives constructed in this work.

Plasmids used in the present study are listed in Table 1. Plasmid pJOC21, carrying the basic miniUIB genetic structure (Fig. 1A), was obtained after inserting the two IRs (IRL and IRR) of ISPst9 into pGP704. IRR was first amplified by PCR using primers PCR1F and PCR1R from P. stutzeri AN10 genomic DNA. The resulting PCR product was further digested with SacI and EcoRI and cloned between SacI and EcoRI restriction sites in pGP704 plasmid. In a second step, IRL was obtained by PCR amplification using primers PCR3F and PCR3R and cloned at the BglII and XbaI restriction sites.

Fig 1.

Physical and genetic maps of the miniUIB structures. Genetic structures harbored on pGP704 plasmid are shown. Gray and white triangles represent the left and right IRs of ISPst9, respectively. Unique restriction sites (B, BglII; E, EcoRI; K, KpnI; M, SmaI; N, NdeI; R, EcoRV; S, SacI; X, XbaI) are indicated. (A) miniUIB in pJOC21; (B) miniUIB-Km (1.6 kb) in pJOC22Km; (C) miniUIB-Cm (2.1 kb) in pJOC22Cm; (D) miniUIB-BC (3.6 kb) in pJOC22BC; (E) miniUIB-BC/Km (5.0 kb) in pJOC22BCKm; (F) miniUIB-KNG (7.0 kb) in pJOC22KNG; (G) miniUIB-Km/KNG (8.4 kb) in pJOC22KmKNG; (H) miniUIB-BC/Km/KNG (11.8 kb) in pJOC22BCKmKNG; (I) miniUIB-ALKB (8.5 kb) in pALKB; (J) miniUIB-ALK (12.8 kb) in pALK; (K) miniUIB beside IR-deleted ISPpu12 in pJOC100; (L) miniUIB-Km beside IR-deleted ISPpu12 in pJOC100Km; (M) miniUIB-ALK beside IR-deleted ISPpu12 in pJOC100ALK.

Diverse genetic material was cloned between the IRs of pJOC21 giving the miniUIB derivatives represented in Fig. 1. Plasmids pJOC22Km, harboring miniUIB-Km (Fig. 1B), and pJOC22Cm, containing miniUIB-Cm (Fig. 1C), were obtained by cloning the Km^r and the Cm^r determinants of pCSI2 and pCSI1 plasmids (19), respectively, into the XbaI (Km^r) or EcoRV (Cm^r) restriction sites of pJOC21. Plasmid pJOC22BC, harboring miniUIB-BC (Fig. 1D), was obtained by cloning the SmaI-linearized pBC SK(−) cloning vector (Cm^r; Stratagene) into the SmaI restriction site of pJOC21. Plasmid pJOC22BCKm, harboring miniUIB-BCKm (Fig. 1E), was obtained by cloning the Km^r determinant of pCSI2 into the XbaI restriction site of pJOC22BC. Plasmid pJOC22KNG, harboring miniUIB-KNG (Fig. 1F), was obtained by cloning the SmaI-linearized pKNG101 plasmid (Sm^r [20]) between the SmaI and EcoRV restriction sites of pJOC21. Plasmid pJOC22KmKNG, harboring miniUIB-KmKNG (Fig. 1G), was obtained by cloning the Km^r determinant of pCSI2 into the XbaI restriction site of pJOC22KNG. Plasmid pJOC22BCKmKNG, harboring miniUIB-BCKmKNG (Fig. 1H), was obtained by cloning the SmaI-linearized pBC SK(−) cloning vector (Cm^r; Stratagene) into the SmaI restriction site of pJOC22KmKNG.

Plasmid pALK, containing the miniUIB-ALK minitransposon (Fig. 1J) with the alkST and alkBFGHJKL determinants of P. putida TF4-1L (GPo1), was constructed in three steps. First, the alkBFGH determinants were inserted between the XbaI and SacI restriction sites of pJOC21 after PCR amplification with primers ALK1-F and ALK1-R. The alkJKL genes were then amplified by PCR using primers ALK2-F and ALK2-R and further cloned in the SacI restriction site, resulting in the pALKB plasmid that harbors the miniUIB-ALKB minitransposon (Fig. 1I). Finally, the regulator operon alkST was amplified with primers ALKRsma-F and ALKRsma-R and inserted in the SmaI restriction site of pALKB, resulting in pALK plasmid.

Finally, pJOC100 (Fig. 1K) and pJOC100ALK (Fig. 1M) plasmids, both containing the transposition machinery of ISPpu12 without the IRs, were obtained by PCR amplification of IR-free ISPpu12 of P. stutzeri AN10 with primers ISPPU-25F-eco and ISPPU-3348R-eco, followed by cloning in the EcoRI restriction site of pJOC21 and pALK, respectively. Plasmid pJOC100Km (Fig. 1L) was obtained after cloning the Km^r from pCSI2 in the EcoRV restriction site of pJOC100.

Nucleotide sequence accession number.

The sequence of the pJOC21 plasmid has been deposited in GenBank under accession number KC295444.

RESULTS AND DISCUSSION

Activated ISPpu12 has an in trans effect on genetic structures flanked by ISPpu12-like IRs.

Previous experiments revealed that ISPpu12 activates its own transposition, and it was also responsible for the transposition of ISPst9 in P. stutzeri AN10 after conjugative interaction (12). This in trans effect on ISPst9, an element of the same ISL3 family as ISPpu12 (17), was explained by the high similarity between their IRs and therefore, the plausible recognition by the transposase of ISPpu12. An ISPpu12-derived minitransposon, designated as miniUIB, was designed to explore the in trans activity of ISPpu12 on genetic structures flanked by similar IRs to those found in the transposon (Fig. 1A). The miniUIB structure was constructed on the R6K-replicase suicidal plasmid pGP704 giving pJOC21. To follow the mobilization of miniUIB, the Km^r gene of pCSI2 plasmid was cloned between the IRs, producing miniUIB-Km (Fig. 1B). The resulting plasmid, pJOC22Km, was transferred into P. stutzeri AN10 (original strain with ISPpu12 encoded in its genome) and P. stutzeri AN11 (negative control, lacking ISPpu12 in its genome) by a 7-h RP4-based conjugation using E. coli S17-1λ_pir as donor strain. Conjugative interaction was expected to activate the transposition mechanism of ISPpu12 in P. stutzeri AN10, having an in trans effect on the entering miniUIB-Km structure located on plasmid pJOC22Km. The plasmid pGP704Km, similar to pJOC22Km but with no IRs flanking the Km^r determinant, was used as a negative control. As shown in Table 2, the frequency of Km^r acquisition was almost 500 times higher in P. stutzeri AN10 when the antibiotic resistance cassette was flanked by the IRs. Also, in all analyzed cases, the Km^r transconjugants did not acquire the Ap^r determinant of the vector. On the other hand, when conjugating with pGP704Km, the Km^r gene was acquired at a low frequency, similar to that seen for the Ap^r determinant harbored in the plasmid (Table 2), indicating that the entire plasmid which is unable to replicate in Pseudomonas, was incorporated into the genome of P. stutzeri AN10 by a genetic process not related to transposition (i.e., recombination). As expected, no differences were observed between both plasmids (pJOC22Km and pGP704Km) when P. stutzeri AN11 (strain without ISPpu12 encoded in its genome) was used as conjugative recipient strain (Table 2).

Table 2.

Frequency of antibiotic resistance acquisition after conjugation

Recipient strain	Plasmid	Avg acquisition ± SD (%)a
		Kmr (Kmr Apr)b	Apr (Apr Kmr)c
P. stutzeri AN10	pGP704Km	(6.9 ± 0.9) × 10−7 (82)	(6.5 ± 1.2) × 10−7 (79)
	pJOC22Km	(3.4 ± 0.7) × 10−4 (0)	(7.4 ± 0.4) × 10−7 (82)
P. stutzeri AN11	pGP704Km	(6.1 ± 1.5) × 10−8 (74)	(6.2 ± 2.3) × 10−8 (80)
	pJOC22Km	(5.8 ± 0.9) × 10−8 (78)	(6.5 ± 1.1) × 10−8 (77)

^aE. coli S17-1λ_pir was used as a donor strain. The values are the averages of three independent mating experiments.

^bFrequencies of recipient cells that received the Km^r determinant. Frequencies were calculated by dividing the number of Km^r transconjugants obtained by the number of recipient cells counted immediately after conjugation. Transconjugants were selected on MBM plates supplemented with succinate and kanamycin. Recipient strains were counted on MBM plates supplemented with succinate. The percentage of Km^r transconjugants that also acquired the Ap^r determinant of the vector is indicated after each value in parentheses.

^cFrequencies of recipient cells that received the Ap^r determinant. Frequencies were calculated by dividing the number of Ap^r transconjugants obtained by the number of recipient cells counted immediately after conjugation. Transconjugants were selected on MBM plates supplemented with succinate and ampicillin. Recipient strains were counted on MBM plates supplemented with succinate. The percentage of Ap^r transconjugants that also acquired the Km^r determinant of the vector is indicated after each value in parentheses.

Ten of the P. stutzeri AN10 transconjugants (P. stutzeri AN10Km) which had acquired the Km^r determinant from pJOC22Km were further analyzed by genomic DNA digestion followed by Southern blotting hybridization. Between one and three hybridization bands were obtained when a specific probe for the Km^r gene (KM probe) was used (Fig. 2A), whereas no hybridization signal was obtained when pGP704 was used as a probe (result not shown). As expected, hybridizations with TNPA, a probe for the transposase gene of both ISPst9 and ISPpu12, and TNPR, a probe specific for the regulator gene of ISPpu12, revealed transposition of both ISL3-like elements of P. stutzeri AN10 (Fig. 2B).

Fig 2.

Transposition of miniUIB-Km in P. stutzeri AN10. (A and B) Southern blot assay against EcoRI-digested genomic DNA obtained from P. stutzeri AN10 (WT) and 10 different Km^r derivatives that acquired the miniUIB-Km structure after conjugation with E. coli S17-1λ_pir (pJOC22Km). (A) Detection of the acquired Km^r determinant by hybridization with the KM probe. (B) Detection of ISL3-like ISs (ISPst9 and ISPpu12) by hybridization with the TNPA probe. Open circles highlight those bands harboring ISPpu12, detected by hybridization with the TNPR probe. Asterisks indicate those transconjugants (3, 4, and 5) used to test the stability of the miniUIB-Km insertion. (C) Frequencies of Km^r acquisition by P. stutzeri AN10 after conjugation with E. coli S17-1λ_pir (pJOC22Km) using different mating times.

In order to reduce the number of miniUIB copies acquired by the recipient strain, conjugation experiments between E. coli S17-1λ_pir (pJOC22Km) and P. stutzeri AN10 using reduced mating times were performed. Maximum frequency of Km^r acquisition was obtained just after 30 min of mating time (Fig. 2C). As previously done, 10 P. stutzeri AN10 transconjugants that acquired the Km^r determinant from pJOC22Km were further analyzed by genomic DNA digestion with EcoRI, followed by Southern blotting hybridization. A single hybridization band was obtained in 9 of them when the KM probe was used (see Fig. S1 in the supplemental material). Due to this, further mating experiments were carried out with only 30 min of conjugative interaction and not the 7 h used previously.

Genomic stability of the miniUIB structure.

In order to determine the stability of the miniUIB structure when inserted in the chromosome of P. stutzeri AN10, we analyzed the loss of the miniUIB-Km harbored in different transconjugants of P. stutzeri AN10: (i) after consecutive subcultures without antibiotic pressure and (ii) after a second event of conjugative interaction. For this, three of the transconjugants that harbored the miniUIB-Km structure were selected (P. stutzeri AN10Km-3, -4, and -5, marked with an asterisk in Fig. 2). Ten consecutive subcultures in LB broth with no antibiotic pressure were done with the three AN10 derivatives. Total cell counts on LB plates and LB plates supplemented with Km showed, in all cases, no Km^r loss (result not shown). This is consistent with the fact that the transposition of both ISL3-like ISs, ISPpu12 and ISPst9, is maintained at very low rates when not induced by conjugative interaction (12). On the other hand, if a second conjugation event was carried out, miniUIB structures could form intermediate circles and be lost, as already shown for the ISL3-like elements. To demonstrate this, the pBBRMCS-1 plasmid, able to replicate in Pseudomonas, was introduced by conjugation in the three selected P. stutzeri AN10Km strains. Transconjugants that had acquired the pBBRMCS-1 plasmids were isolated on MBM plates supplemented with succinate and Cm. A total of 100 Cm^r isolates derived from each P. stutzeri AN10Km transconjugant were selected to monitor the loss of the Km^r determinant after acquisition of pBBRMCS-1 plasmid. Loss of the Km^r determinant was observed only in transconjugants that conserved an ISPpu12 copy, and the loss of the Km^r phenotype correlated to the number of copies of the Km^r gene in the genome (P. stutzeri AN10Km-3, with two copies of the Km^r gene, 15% loss; P. stutzeri AN10Km-4, with one copy of the Km^r gene, 38% loss). No loss was observed in P. stutzeri AN10Km-5, which did not conserve ISPpu12 in its genome.

Size versus frequency of miniUIB acquisition.

The idea of using miniUIB as a system for inserting genetic material in the genome of P. stutzeri AN10 and other bacteria required determining the maximum DNA size that the miniUIB structure could mobilize. For this purpose, different DNA fragments were inserted between the IRs of the miniUIB structure to generate miniUIB derivatives of increased DNA size (Fig. 1). Frequency of acquisition of the miniUIB structure was calculated following the appearance of the appropriate antibiotic resistance marker in P. stutzeri AN10 after inserting by conjugation each one of the Pseudomonas nonreplicative pJOC21 derivatives containing the different miniUIB structures. The frequency of miniUIB acquisition was higher with the smallest miniUIB derivative [miniUIB-Km, 1.6 kb, frequency of (3.4 ± 0.7) × 10⁻⁴] and decreased with the increasing size of the insert, maintaining a logarithmic regression (R² = 0.9873) up to over 12 kb (Fig. 3). The frequency of transposition of miniUIB structures over 12 kb were close to the frequencies obtained for the control plasmid pGP704Km (Table 2), unable to transpose and replicate in Pseudomonas. These data suggest that longer inserts may undergo transposition at lower frequencies, similar to the frequency of insertion by random recombination. This was corroborated by the increased detection of Ap^r colonies (up to 50%), the resistance determinant marker of pGP704 plasmid, when large miniUIB structures were used (Fig. 3).

Fig 3.

Frequency of miniUIB acquisition by P. stutzeri AN10 versus size of miniUIB structure. Strain AN10 was conjugated with E. coli S17-1λ_pir harboring different pJOC21 derivatives. Dots and error bars indicate averages and standard deviations of three independent experiments, respectively. Percentages revealed entire plasmid acquisition, evaluated by acquired Ap^r phenotype of isolated transconjugants. Dashed line shows the frequency of entire pGP704Km acquisition by P. stutzeri AN10 after conjugation with E. coli S17-1λ_pir harboring this plasmid. Continuous line shows the logarithmical regression between both parameters. Plasmids, miniUIB derivatives, and miniUIB length are as follows: A, pJOC22Km, miniUIB-Km, 1.6 kb; B, pJOC22Cm, miniUIB-Cm, 2.1 kb; C, pJOC22BC, miniUIB-BC, 3.6 kb; D, pJOC22BCKm, miniUIB-BC/Km, 5.0 kb; E, pJOC22KNG, miniUIB-KNG, 7.0 kb; F, pJOC22KmKNG, miniUIB-Km/KNG, 8.4 kb; G, pJOC22BCKmKNG, miniUIB-BC/Km/KNG, 11.8 kb.

Biotechnological applications of miniUIB in P. stutzeri AN10.

Here, we present two different strategies in which the miniUIB element was used to transform P. stutzeri AN10 into an alkane-degrading bacterium.

The first strategy consisted in introducing the alkane determinants in a single conjugation step. For this, we constructed the pALK plasmid (Table 1), which harbors the miniUIB-ALK minitransposon (Fig. 1J). The miniUIB-ALK contains the two alkane degradation operons of P. putida TF4-1L (GPo1) (21): the 4.3-kb regulation operon alkST, and the 8.3-kb catabolic operon alkBFGHJKL. Plasmid pALK was introduced into AN10 by conjugation. The mixture of mated cells were collected and incubated in MBM supplemented with n-octane and n-hexadecane (each at 2% [wt/vol]) as the sole carbon and energy source in order to select for transconjugants carrying the miniUIB-ALK structure. Genes alkBFGHJKL encode the functions necessary for the first steps of C₅-C₁₂ n-alkanes degradation such as uptake, hydroxylation, and dehydrogenation of the compounds (24). Thus, n-octane was used as carbon and energy source. The substrate can be mineralized when the cell has an active β-oxidation pathway for fatty acids. The fatty acid degradation pathway of P. stutzeri could be constitutively expressed as in P. putida (24) or could be induced only by fatty acids longer than C₁₂-like in E. coli (25). To ensure the induction of the fatty acid degradation genes and, by this way, not to be dependent on the appearance of a spontaneous constitutive mutant, we added 2% of n-hexadecane to the media, as previously done by Eggink et al. (25). After incubation in MBM supplemented with n-octane and n-hexadecane, growth was only obtained in AN10 cells mated with E. coli S17-1λ_pir (pALK) and not in the three negative controls used: (i) P. stutzeri AN10, (ii) E. coli S17-1λ_pir (pALK), and (iii) AN10 cells mated with E. coli S17-1λ_pir harboring plasmid pALKB, which contains the catabolic operon alkBFGHJKL but lacks the regulatory operon alkST. Growth also failed in AN10 cells mated with E. coli S17-1λ_pir (pALK) but incubated in MBM without alkane supplementation.

An aliquot of the culture grown in alkanes was serially diluted and plated on MBM, and further incubated with n-octane and n-hexadecane in vapor phase as unique carbon and energy sources. Eight transconjugants (P. stutzeri AN10-ALK) were isolated and proved to be AN10 derivatives by 16S rRNA gene sequence analysis. In order to discriminate whether the miniUIB-ALK transposed into the AN10 genome or whether pALK was captured, we cultivated all eight AN10-ALK transconjugants in the presence of Ap. Only one of the eight isolates analyzed (P. stutzeri AN10-ALK1) was Ap^r, indicating that in this transconjugant the whole plasmid had been acquired. This was further confirmed by Southern blotting hybridization analysis using a specific probe for the alk determinants and for linearized pGP704 (Fig. 4A). Hybridization with the ALK probe revealed the presence of the alk determinants in all transconjugants, whereas the PGP probe (for pGP704 plasmid) gave a hybridization band only with the AN10-ALK-1 transconjugant. In addition, hybridization also revealed that only one of the transconjugants (AN10-ALK8) harbored more than one copy of the miniUIB-ALK structure. Thus, it was demonstrated that the alk determinants were acquired by true transposition of the 12.8-kb miniUIB-ALK structure in 7 of the 8 transconjugants tested. P. stutzeri strains AN10-ALK3 and AN10-ALK8 were selected to carry out further growth experiments (Fig. 4B). Both AN10-ALK transconjugants could grow with only n-octane although absorbance measurements (600 nm) of the cultures almost doubled at the stationary phase when hexadecane was added. No growth was detected when hexadecane was used as sole carbon and energy source, probably because the alk genes only allow the degradation of C₅-C₁₂ n-alkanes (24). No differences in growth were detected due to the presence of one or two copies of the miniUIB-ALK structure (Fig. 4B), probably due to the enrichment strategy used for the selection of transconjugants.

Fig 4.

Alkane degradation by P. stutzeri AN10 and it derivatives harboring the miniUIB-ALK structure. (A) Southern blot hybridization with the ALK probe against EcoRI-digested genomic DNA obtained from P. stutzeri AN10 (WT) and 8 different AN10-ALK derivatives that acquired the miniUIB-ALK structure after conjugation with E. coli S17-1λ_pir (pALK). Open circle highlight the unique bands harboring the entire pALK plasmid detected by hybridization with the PGP probe. Asterisks indicate those transconjugants (3, AN10-ALK3; 8, AN10-ALK8) used to evaluate the acquired alkane degradation phenotype. (B) Growth (maximum cell density) of P. stutzeri AN10, AN10-ALK3, and AN10-ALK8 on MBM supplemented with n-octane (black bars), n-hexadecane (gray bars), and a mixture of both n-alkanes (white bars). Values are represented as the averages and standard deviations of three independent experiments.

The second strategy consisted in introducing the alkane determinants in two separate conjugation steps. The rationale of this experiment was to demonstrate that, although a second conjugation event could result in the loss of the first miniUIB structure inserted, a sufficiently high percentage of the population will end with both miniUIB structures introduced in successive events. A multiconjugation approach to introduce different DNA fragments in sequential transposition steps could also be of great interest for future biotechnological applications. This strategy of using multiple steps of DNA integration with the same minitransposon is not possible with minitransposons with transposition immunity, like Tn7 derivatives (10). For this two-step process, we used the previously constructed plasmid pALKB (harboring miniUIB-ALKB, which contains the alkBFGHJKL operon) (Fig. 1I) and a newly constructed plasmid pALKC (harboring miniUIB-ALKC, which contains the regulatory operon alkST and a chloramphenicol resistance determinant) (see Fig. S2 in the supplemental material). P. stutzeri AN10 was first mated with E. coli S17-1λ_pir (pALKC). Mated cells were then grown in liquid MBM supplemented with succinate 0.5% (wt/vol) and Cm to select those AN10 transconjugants that had acquired miniUIB-ALKC. The grown culture was used directly in a second conjugation experiment with E. coli S17-1λ_pir (pALKB). As in the first strategy, the resultant second mating event was grown in MBM supplemented with n-octane and n-hexadecane as sole carbon and energy sources. Successful growth was only obtained in double-conjugated AN10 cells and not in the negative controls (see Fig. S3 in the supplemental material).

Extending the use of the miniUIB system to ISPpu12-free model microorganisms.

The use of the miniUIB structure in bacteria lacking ISPpu12 was tested in another member of P. stutzeri species and in two different widely used model species: Klebsiella pneumoniae, related to respiratory infections (up to 14,795 references in PubMed on 1 December 2012) and P. putida, involved in hydrocarbon biodegradation (up to 5,566 references in PubMed on the same date). The strategy used was started by introducing both tnpA and tnpR genes of ISPpu12 in the suicidal vector, outside of the miniUIB structure. For this, the complete ISPpu12 sequence (without the IRs) was inserted in the EcoRI restriction site of pJOC21, obtaining pJOC100 (Fig. 1K). The gene structure of the IS was conserved in order to maintain its activation phenotype after conjugative interaction. Unlike the instability reported for mini-Tn5 (1), plasmid pJOC100 was stable in E. coli, probably due to the transcriptional repression of tnpA mediated by TnpR in the absence of stimuli (12).

In order to test the integration of miniUIB in ISPpu12-lacking strains, the miniUIB-Km structure was reconstructed by cloning the Km^r gene of pCSI2 between the IRs, giving pJOC100Km (Fig. 1L). This construction was introduced in three different bacterial strains lacking ISPpu12 in their genome: K. pneumoniae CMD1, P. stutzeri AN11, and P. putida IS⁻. The results with miniUIB-Km were compared to the ones obtained with its homolog in the miniTn5 series: the miniTn5-Km1 minitransposon (1, 2). All mating experiments were performed in triplicate. In all cases, Km^r acquisition frequencies were higher when miniUIB-Km was used [K. pneumoniae CMD1, (4.91 ± 0.6) × 10⁻⁵ for miniUIB-Km, (1.4 ± 0.7) × 10⁻⁵ for miniTn5-Km1; P. stutzeri AN11, (1.1 ± 0.3) × 10⁻⁴ for miniUIB-Km, (3.9 ± 1.2) × 10⁻⁸ for miniTn5-Km1; P. putida IS⁻, (9.7 ± 0.8) × 10⁻⁵ for miniUIB-Km, (5.2 ± 2.8) × 10⁻⁶ for miniTn5-Km1].

In order to determine whether miniUIB-Km was randomly inserted in single copy, eight P. stutzeri AN11 Km^r derivatives were selected, and a Southern blot hybridization was performed against their EcoRI-digested genomic DNAs with the KM probe. All Km^r AN11 derivatives showed different single hybridization bands (Fig. 5), suggesting that miniUIB-Km was randomly inserted by transposition in the genome of P. stutzeri AN11 and, thus, it can be used as a genetic tool for random transposon mutagenesis in ISPpu12-lacking microorganisms.

Fig 5.

Southern blot hybridization with the KM probe against EcoRI-digested genomic DNA obtained from eight different Km^r derivatives of P. stutzeri AN11 that acquired the miniUIB-Km structure after conjugation with E. coli S17-1λ_pir (pJOC100Km).

In order to prove that the miniUIB structure could be useful for introducing biotechnologically relevant foreign-DNA in ISPpu12-free bacteria, we transformed P. stutzeri AN11 into an alkane-degrading bacterium by introducing the miniUIB-ALK structure. For this purpose we constructed pJOC100ALK (Fig. 1M) by cloning the IR-free ISPpu12 of P. stutzeri AN10 into the EcoRI site of pALK. Following the strategy previously used for P. stutzeri AN10, we obtained a P. stutzeri AN11 derivative culture able to grow using n-octane and n-hexadecane as unique carbon and energy source (Fig. 6). No growth was observed in control cultures. A negative control using pALK (which previously gave a positive growth in AN10) was also included. As expected, due to the absence of ISPpu12 in the genome of P. stutzeri AN11 genome, no transposition and, consequently, no acquisition of the alk determinants of pALK could occur and, thus, no growth was observed. Ten P. stutzeri AN11 transconjugants from pJOC100ALK with an alkane degrading phenotype (P. stutzeri AN11-ALK) were isolated. In all cases, they had randomly acquired the miniUIB100ALK structure in their genomes, as shown by Southern blot hybridization analysis with the ALK probe (Fig. 6A). In most of the AN11-ALK transconjugants, we observed two hybridization bands, probably due to the enrichment strategy that selected those with higher growth rates in the presence of the mixture of alkanes. No hybridization band was observed when PGP probe was used, suggesting that all P. stutzeri AN11-ALK acquired the miniUIB-ALK by true transposition. As done for P. stutzeri AN10-ALK, two P. stutzeri AN11-ALK transconjugants were selected to carry out further growth experiments (Fig. 6B). As shown for AN10-ALK, both AN11-ALK transconjugants could grow with only the presence of n-octane, although population size increased to almost double at the stationary phase when hexadecane was added. Similarly, both AN11-ALK transconjugants were not able to grow on n-hexadecane as unique carbon and energy source, and no differences in growth were detected due to the presence of one or two copies of the miniUIB-ALK structure.

Fig 6.

Alkane degradation by P. stutzeri AN11 and its derivatives harboring the miniUIB-ALK structure. (A) Southern blot hybridization with the ALK probe against EcoRI-digested genomic DNA obtained from P. stutzeri AN11 (WT) and 10 different AN11-ALK derivatives that acquired the miniUIB-ALK structure after conjugation with E. coli S17-1λ_pir (pJOC100ALK). Asterisks indicate the transconjugants (3, AN11-ALK3; 4, AN11-ALK4) used to evaluate the acquired alkane-degradation phenotype. (B) Growth (maximum cell density) of P. stutzeri AN11, AN11-ALK3, and AN11-ALK4 on MBM medium supplemented with n-octane (black bars), n-hexadecane (gray bars), and the mixture of both n-alkanes (white bars). Values are represented as averages and standard deviations of three independent experiments.

Introduction of the miniUIB system into environmental isolates.

The comparative study with miniTn5-Km1 was extended to 50 phylogenetically distinct environmental isolates obtained from polluted sand samples of a beach after the Prestige oil spill in Galicia, Spain (17). The isolates were selected from our collection of 292 strains attending to the criteria that they were sensitive to kanamycin, and they developed visible and identifiable colonies after 48 h of incubation at 30°C in MBM supplemented with succinate (0.5% [wt/vol]). Analysis of their partial 16S rRNA gene indicated that they affiliated to Alphaproteobacteria (6 isolates of 5 different genera: Paracoccus, Roseobacter, Ruegeria, Sinorhizobium, and Thalassospira), Betaproteobacteria (7 isolates of 4 different genera: Advenella, Alcaligenes, Azoarcus, and Wautersia), Gammaproteobacteria (20 isolates of 11 different genera: Acinetobacter, Aeromonas, Alcanivorax, Klebsiella, Marinobacter, Nitrincola, Photobacterium, Pseudomonas, Psycrobacter, Shewanella, and Vibrio), Bacteroidetes (3 isolates of Cytophaga sp.), Actinobacteria (11 isolates of 5 different genera: Arthrobacter, Cellulomonas, Nocardia, Promicromonospora, and Rhodococcus), and Firmicutes (3 isolates of 2 different genera: Bacillus and Staphylococcus). Figure 7 summarizes the phylogenetic affiliation of assayed strains and the results obtained from conjugation experiments with both minitransposons. Transposition with both minitransposons failed in 15 of the 50 isolates. None of the minitransposons was able to transpose in isolates from the genera Cellulomonas, Cytophaga, Nocardia, Staphylococcus, Photobacterium, and Promicromonospora. In addition, transposition with both minitransposons failed in one of the six isolates tested of the genus Pseudomonas and in one of the three isolates tested of the genus Marinobacter. MiniUIB-Km transposed in 35 of the 50 assayed isolates, while miniTn5-Km1 was able to transpose only in 30 of them. Most interestingly, we did not observe any isolate in which miniTn5 would transpose but not miniUIB. Only miniUIB-Km, and not miniTn5-Km1, was able to transpose in several isolates of genera Acinetobacter and Paracoccus, and in some Gram-positive isolates affiliated to Arthrobacter, Bacillus, and Rhodococcus genera. When Km^r acquisition frequencies were considered, 73% of the assayed isolates presented higher frequencies of acquisition for miniUIB-Km than for minTn5-Km1. In fact, in seven of them the frequencies with miniUIB-Km were 10-fold higher than the observed for miniTn5-Km1. Meanwhile, only two of the isolates had 10-fold higher Km^r acquisition frequencies for miniTn5-Km1 compared to miniUIB-Km. Thus, it can be concluded that miniUIB-Km can be used as a random transposon mutagenesis tool in a broad range of bacteria, including some Gram-positive isolates, and that its host range is wider than the one of miniTn5-Km. In addition, these results suggest the plausible use of the miniUIB structure for introducing foreign genetic material by transposition in a wide range of bacteria and not only in P. stutzeri AN10 and the model strains mentioned above, allowing their biotechnological manipulation.

Fig 7.

Comparison of the transposition frequencies of miniUIB-Km and miniTn5-Km1 in all phylogenetically different isolates used. The phylogenetic tree (neighbor-joining dendrogram based on Jukes-Cantor 16S rRNA gene distance matrix) shows relationships between the type strains of each represented genera used in the present study. First column (isolates) indicates the number of isolates used that affiliated to each genus (28 total genera). Second and third column indicate the number of isolates that acquired the Km^r determinant of miniUIB-Km and miniTn5-Km1, respectively. Plotted comparisons (frequency of Km^r acquisition of miniUIB-Km divided by the one obtained for miniTn5-Km1) were calculated only with those 33 isolates that were able to receive both minitransposons. Black circles indicate the 22 isolates that presented higher frequencies for miniUIB-Km than for miniTn5-Km1. White circles indicate the eight isolates with higher frequencies for miniUIB-Tn5 than for miniUIB-Km. Gray circles indicate the ratio of transposition obtained for P. stutzeri AN11 (number 1), P. putida IS⁻ (number 2), and K. pneumoniae CMD1 (number 3).

Concluding comments.

In the present study we have presented the biotechnological tool miniUIB, an ISPpu12-based minitransposon. This system is an easy and ready-to-use tool for generating random mutants or for inserting genetic structures in a wide range of bacteria. Many minitransposon structures have been described till date, but the miniTn5 series (1, 2) has been the most commonly used by microbiologists. Nevertheless, the miniTn5 system presents two main disadvantages that we believe that they can be overcome with the ISPpu12-based minitransposon: (i) instability when constructions lacking an insert between the IRs (2) and (ii) the fact that the frequencies of correct minitransposon insertion versus plausible recombinational cointegration of the whole suicidal vector into the chromosome are insufficient in some strains due to the dependence of host regulation for transposition.

Plasmid pJOC100 harbors the exclusive self-regulated transposition mechanism of ISPpu12, which is strongly activated only after conjugative interaction (12, 13). This converts the minitransposon miniUIB into a host-independent system that, as intended, is highly active only during the conjugation process. Unlike miniTn5, which is unstable when nothing is cloned between its IRs (2), miniUIB is a ready-to-use system with no need of auxiliary plasmids. In this sense, miniUIB is a stable structure ready for cloning DNA fragments up to 12 kb in length into its multicloning site. The high frequency of transposition allows selecting for real transconjugants (i.e., the alk determinants) by their phenotype, without the use of nondesirable markers (i.e., antibiotic resistance determinants).

Finally, we have also demonstrated that it is possible to introduce distinct DNA-fragments in the same recipient cell by using different miniUIB structures in successive conjugation steps. This probably makes the miniUIB more attractive for its biotechnological use than other systems, such as the miniTn7-based minitransposons which display transposition immunity (9, 10).