Relative Strengths of the Class 1 Integron Promoter Hybrid 2 and the Combinations of Strong and Hybrid 1 with an Active P2 Promoter

The relative strengths of the uncommon promoters hybrid 2, hybrid1 with an active P2 promoter (hybrid 1+P2), and strong+P2, whichdrive transcription of resistance genes in class 1 integrons,were evaluated using bla_GES-1 as a reporter gene cassette. Hybrid2 was stronger than hybrid 1. Coupling P2 with the strong promoterand with hybrid 1 caused a measurable increase in GES-1 expression.

INTRODUCTION

Top

INTRODUCTION

Expression of antibiotic resistance genes in the variable regionof class 1 integrons is driven by a promoter (Pc) that is 214bases from the inner boundary of the 5' conserved segment andlocated within the coding sequence of intI1. There are fourversions of Pc, designated "weak," "strong," "hybrid 1," and"hybrid 2," exhibiting differences in the –35 and/or –10sequences, separated by 17 bases. Additionally, a second promoter,P2, complements the Pc weak promoter in various integrons. P2occurs 119 bases downstream from Pc and is formed by an insertionof three G bases optimizing the spacing (17 bases) between the–35 and –10 hexamers (P2 active) (5). The relativestrengths of the weak, strong, and hybrid 1 versions of thePc promoter, as well as that of the weak promoter complementedby P2 (weak+P2), have been determined previously (5, 9), whilethat of hybrid 2, which occurs in a relatively small numberof class 1 integrons (Table 1) (3, 4, 6, 8, 10, 14, 16), hasnot been studied. The increasing number of characterized integronsrevealed the existence of additional promoter sequences, suchas the combinations of the strong promoter and hybrid 1 withP2 (Table 1) (7, 12). In this study, we evaluated the relativestrengths of hybrid 2 and the latter promoter combinations introducedinto a GES-1 extended-spectrum-β-lactamase-encoding integron.Comparative assessments included amounts of bla_GES-1 transcripts,levels of hydrolytic activity of GES-1, and MICs of β-lactams.

View this table:
[in this window]
[in a new window]

TABLE 1. Characteristics of class 1 integrons with the uncommon promoter sequence Pc hybrid 2 and the combinations of P2 with Pc hybrid 1 and Pc strong

Cloning of a GES-1-encoding integron and construction of the Pc-P2 sequence variants.

Top

Cloning of a GES-1-encoding...

A GES-1-producing isolate of Pseudomonas aeruginosa (Pa-G1)was used as a source of In58-G1, a typical class 1 integronwith a weak promoter containing a single gene cassette, bla_GES-1(GenBank accession no. EU598463). A 1,641-bp fragment of In58-G1,including the bla_GES-1 cassette and the promoter sequence, wasamplified by PCR and cloned into the Topo TA vector (Invitrogen,Carlsbad, CA), yielding plasmid pTOPO-G1. An EcoRI fragmentof 1,690 bp from pTOPO-G1 was treated with a DNA polymeraseKlenow fragment and introduced into the single HincII site ofplasmid pACYC177, thereby disrupting the resident bla_TEM-1.A recombinant plasmid of 5,607 bp (pHP327) carrying the 1,641-bpsegment of In58-G1 in an orientation opposite that of the 246-bp5' remnant of bla_TEM-1 was selected for the construction ofthe various Pc-P2 sequence combinations. Site-specific mutagenesiswas performed using 10 mutagenic primers (Table 2) and a QuickChangesite-directed mutagenesis kit (Stratagene, La Jolla, CA). Sevenvariants of plasmid pHP327 carrying the weak, strong, hybrid1, hybrid 2, weak+P2, strong+P2, and hybrid 1+P2 promoter sequenceswere studied. Additionally, a pHP327 variant with a promoterlessintegron was constructed by reverting the 29-bp Pc sequenceand also deleting 5 of the 17 bp between the –35 and –10hexamers. Entire plasmid sequences were verified on both strandsby use of an API377 sequencer (Applied Biosystems, Foster City,CA). Escherichia coli MC4100, a ${Delta}$ ampC mutant, was used as a host.Isoelectric focusing and staining with nitrocefin showed thatGES-1, with a pI of 5.8, was the sole β-lactamase producedby the respective clones.

View this table:
[in this window]
[in a new window]

TABLE 2. Oligonucleotide primers

Characterization of bla_GES-1 transcripts by Northern hybridization and real-time reverse transcription-PCR (RT-PCR).

Top

Characterization of bla_GES...

Total RNA was prepared using RNA protect and RNeasy mini kits(Qiagen, Hilden, Germany). RNA samples (10 µg from eachstrain) were electrophoresed through formaldehyde-containingagarose, transferred to Hybond-N⁺ membranes (Amersham Biosciences,United Kingdom), and hybridized with a digoxigenin-labeled bla_GES-1-specificprobe (618 bp, nucleotides [nt] 744 to 1361; GenBank accessionno. EU598463) following standard techniques (15). These experimentsconfirmed production of similarly sized bla_GES-1 transcripts(approximately 1.2 kb) in all seven strains and also indicatedhigher transcription levels in the strains carrying the Pc strongsequence (data not shown).

RNA samples were treated with RNase-free DNase I using a TurboDNA-free kit (Ambion Inc., Foster City, CA). Real-time RT-PCRwas performed using a Quantifast Sybr green RT-PCR kit (Qiagen).Three reactions per sample were performed, using 10, 20, and50 ng of total RNA. Each reaction mixture also contained QuantifastSYBR green RT-PCR master mix (12.5 µl), Quantifast RTmix (0.25 µl), and 500 nM of each oligonucleotide primer(Ven-8P and Ven-12 [Table 2]) in a final volume of 25 µl.The pACYC177 resident gene aphA1 was used as a reference. Therespective reactions were carried out as described above withprimers Aph-F and Aph-R (Table 2). RT was carried out for 10min at 50°C and terminated by heating at 95°C for 5min. A two-step cycling PCR was then performed for 40 cycles,with a denaturation step at 95°C for 10 s and an annealing-extensionstep at 60°C for 30 s. All reactions were carried out intriplicate using a Corbett Rotor-Gene 6000 (Corbett Life Science,Sydney, Australia). Rotor-Gene 6000 software (version 1.7) wasused for data analysis. Quantities of bla_GES-1 and aphA1 inthe same sample were expressed in threshold cycle values. Thetarget-reference ratios were calculated, and the normalizedvalues were used to compare differential bla_GES-1 expressionlevels. The relative amounts of bla_GES-1 transcripts under thecontrol of the seven promoter sequences are presented in Table3. The lowest value, observed for the weak promoter, was setto 1. Differences in strength among the weak, strong, hybrid1, and weak+P2 promoters were generally in line with those reportedpreviously (9). The strong promoter appeared to be 17 timesmore efficient than the weak promoter. The respective increasewas lower for the weak+P2 sequence and marginal for hybrid 1(4.7- and 1.9-fold, respectively). Association of P2 with hybrid1 caused a 3.5-fold increase in the amount of bla_GES-1 transcript,while the respective increase in the strength of the strongsequence was minor. The efficiency of hybrid 2 was comparableto that of the weak+P2 combination and approximately 2.5-foldhigher than that of hybrid 1. bla_GES-1 transcripts from thepromoterless integron sequence were not detected (Table 3).

View this table:
[in this window]
[in a new window]

TABLE 3. Effect of integron promoter sequences on the relative amounts of bla_GES-1 transcripts, β-lactamase activities, and β-lactam resistance levels in E. coli MC4100 strains carrying a plasmid-borne bla_GES-1 gene cassette

β-Lactamase activities and resistance levels to β-lactams.

Top

β-Lactamase activities and...

β-Lactamase-containing extracts were derived by ultrasonictreatment of bacterial cell suspensions and clarified by ultracentrifugation(4). The protein content was measured with a protein assay kit(Bio-Rad Laboratories, Hercules, CA). β-Lactamase activitywas quantified by spectrophotometry using nitrocefin as a reportersubstrate ( ${Delta}$ ${varepsilon}$ _M⁴⁸² = 17,400 M^–1·cm^–1 [17]).Results were expressed as units of activity. One unit was theamount of enzyme hydrolyzing 1 nmol of substrate/min/mg of proteinat 30°C and pH 7.0. Hydrolytic activities of the extractsfrom strains carrying the Pc strong sequences were higher thanthose of the strains with the hybrid promoters. The lowest activitywas observed in the strain carrying the weak version of Pc (atleast 26- and 3-fold lower than those of the strains with strongand hybrid promoters, respectively). The β-lactamase activityof the strain with hybrid 2 was approximately twofold higherthan that of the hybrid 1-possessing strain. The presence ofa P2 promoter along with the weak and hybrid 1 sequences wasassociated with an increase in β-lactamase activity (approximately6.2- and 3.4-fold, respectively). Also, the hydrolytic activityof the strain with the strong+P2 combination was slightly higherthan that of the strain carrying the strong promoter (Table3).

Susceptibilities to ampicillin, piperacillin, cefaclor, cefuroxime,ceftazidime, aztreonam, and cefepime were determined by Etest(AB Biodisk, Solna, Sweden). MIC differences were in line withthose of the respective β-lactamase activities. As expected,differences in resistance levels were more pronounced with thenarrow-spectrum cephalosporins and ceftazidime, which are GES-1substrates preferable to piperacillin, cefepime, and aztreonam(11). The higher MICs were seen with the two strains carryingthe strong promoter and the lower MICs with the strains withthe weak and hybrid 1 promoters. MICs for E. coli strains harboringthe plasmids with the hybrid 2, weak+P2, and hybrid 1+P2 sequencesdid not differ significantly. Resistance levels of the strainscarrying strong and strong+P2 sequences were also similar. Mostβ-lactams tested were more active against the strain withhybrid 1 than against the strain possessing the hybrid 1+P2combination (Table 3).

Conclusions.

Top

Conclusions.

Determination of enzymatic activity and susceptibility testinghave been used for the characterization of the common promotersof the class 1 integrons (5, 9). We used here a similar methodologyto study hybrid 2, strong+P2, and hybrid 1+P2. Additionally,we employed a real-time RT-PCR assay for the relative quantificationof the bla_GES-1 transcripts. Results of the latter method werein agreement with the quantitative assessment of hydrolyticactivities as well as resistance levels conferred by the indicatorβ-lactamase. Additionally, classification of the weak,strong, hybrid 1, and weak+P2 promoters according to strengthwas as published by Lévesque et al. (9), although thedifferent expression systems and assays do not allow for a directcomparison. Nevertheless, the results clearly indicate thatthe hybrid 2 version of Pc is an effective promoter similarin strength to weak+P2 and stronger than hybrid 1. It is alsoshown that the active configuration of P2 is functional whencoupled with the strong and hybrid 1 promoters, resulting ina measurable increase in expression of an immediately downstreamgene cassette. The levels of this increase were comparable inall three Pc+P2 combinations studied here, suggesting that theoperation of P2 is probably not affected by the type of thepreceding Pc sequence.

Promoter strength is a key factor in determining transcriptionof resistance gene cassettes in class 1 integrons. Therefore,the evaluation of the strengths of hybrid 2, hybrid 1+P2, andstrong+P2 may be useful in explaining resistance phenotypesin strains containing these sequences. Cassette position aswell as the sequence of the respective attC site also influencesexpression in multicassette arrays (5, 13). The potential ofa resistance gene cassette to confer a wide range of resistancelevels may facilitate establishment of these determinants undervarious conditions. Also, it may have implications in the phenotypicrecognition of microorganisms carrying antibiotic resistanceintegrons.

ACKNOWLEDGMENTS

This work was supported by the Hellenic Pasteur Institute anda Kapodistrias grant from the National University of Athens.

We thank Irene Siatravani for technical assistance.

FOOTNOTES

* Corresponding author. Mailing address: Laboratory of Bacteriology, Hellenic Pasteur Institute, Vas. Sofias 127, Athens 11521, Greece. Phone: 30-210-6478810. Fax: 30-210-6423626. E-mail: miriagou{at}pasteur.gr

Published ahead of print on 10 November 2008.

REFERENCES

Top