Quantification of IncP-1 Plasmid Prevalence in Environmental Samples

Abstract

To study the role of broad-host-range IncP-1 plasmids in bacterial adaptability to irregular environmental challenges, a quantitative real-time PCR assay was developed that specifically detects the korB gene, which is conserved in all IncP-1 plasmids, in environmental samples. IncP-1 plasmid dynamics in a biopurification system for pesticide wastes were analyzed.

TEXT

Horizontal gene transfer by broad-host-range plasmids plays a vital role in the adaptation and robustness of bacteria to irregular or novel environmental challenges or opportunities (1, 2). Broad-host-range plasmids are also significantly contributing to the spread of antibiotic resistance (3). Plasmids of the IncP-1 incompatibility group (also called IncP in the classification scheme of Enterobacteriaceae plasmids) are assumed especially to foster horizontal gene transfer because of their stable replication in a wide range of Gram-negative bacteria and their efficient conjugative transfer to an even wider range of taxa (4, 5). They consist of a conserved backbone carrying genes for plasmid persistence and conjugative transfer and typically regions with diverse accessory genes that vary between plasmids. Accessory genes often encode antibiotic or metal resistances or degradative pathways (6). IncP-1 plasmids were first discovered in bacteria from clinical specimens (7, 8) and subsequently found in many geographic regions and diverse environments, including agricultural soil, salt marshes, manure, compost, sewage, water, and sediment (6, 9–13). However, the environmental distribution of IncP-1 plasmids and the factors promoting their frequency in bacterial communities have not been well explored. A high abundance of these plasmids seemed to be related to environmental disturbances like pollution (10, 11, 14). Their detection in microbial community DNA was first based on PCR amplification in combination with Southern blot hybridization of fragments of the trfA gene (12), which codes for the replication initiation protein. These primers were developed on the basis of sequences of subgroups IncP-1α and -β. The discovery of IncP-1 plasmids with largely divergent backbone sequences (9, 15, 16) led to the development of new primer systems for the detection of the trfA gene of IncP-1 subgroups α, β, γ, δ, and ε (17). However, these primers also do not target all known IncP-1 plasmid types, such as the newly described ζ subgroup (18) or pKS208 (accession no. JQ432564), which is most similar to IncP-1γ plasmid pQKH54 (9). The problem of designing one primer system to detect all known IncP-1 plasmid backbones based on trfA is due to its relatively high evolutionary rate (Fig. 1), which is probably a result of adaptation to interacting host proteins in diverse hosts (20). Thus, trfA does not provide enough conserved sites as primer targets and, moreover, as an intermediate target for a TaqMan probe to enable specific quantification of IncP-1 plasmids in environmental DNA by a real-time 5′ nuclease quantitative PCR (qPCR) assay. We searched the IncP-1 backbones for such a conserved region. All common backbone genes of the IncP-1 subgroups were aligned, and relative evolutionary rates were determined (Fig. 1). Of the most conserved regions found within trbC, trbE, traG, traI, and korB, the latter target gene was best suited to the design of a qPCR system. The korB product is evolutionarily constrained because it interacts with itself to form dimers, with IncC, KorA, and multiple DNA binding sites of the plasmid as essential component of the partition system and the regulatory network of the plasmid (21, 22).

Fig 1.

Relative evolutionary rates of genes common to the backbones of all known IncP-1 subgroups (α, β1, β2, γ, δ, ε, and ζ). Genes of two representative plasmids of each subgroup (see Table 1) were aligned, and position-by-position evolutionary rates were determined by using MEGA5 (19). The scaled rates (the average evolutionary rate across all sites is 1) are displayed as moving averages with a window width of 100 bp.

Two forward and three reverse primers were combined to minimize both degeneracy and mismatches with targets (Table 1). A single TaqMan probe was sufficient for qPCR with all of IncP-1 subgroup plasmids tested. An exception was IncP-1γ plasmid pKS208, for which two mismatches with the first five 5′ bases of the TaqMan probe impeded 5′ nuclease activity. This problem could be solved by adding the TaqMan probe Pgz to the PCR mixture to target IncP-1γ and -ζ plasmids, also including more divergent variants (Table 1). Target DNA was amplified in 50-µl reaction mixtures containing 5 μl DNA solution; 1.25 U TrueStart Taq DNA polymerase (Fermentas, St. Leon-Rot, Germany); TrueStart buffer; 0.2 mM each deoxynucleoside triphosphate; 3.5 mM MgCl₂; 0.1 mg/ml bovine serum albumin (Fermentas); 0.4 μM primers F and R; 0.2 μM primers Fz, Rge, and Rd; and 0.3 μM TaqMan probes P and Pgz. Reactions were run for 5 min at 95°C and 40 cycles of 15 s at 95°C, 15 s at 54°C, and 60 s at 60°C in a real-time PCR system (CFX96; Bio-Rad, Munich, Germany). The annealing step at 54°C reduced the variance of cycle threshold values between cloned korB variants from six IncP-1 subgroups (data not shown), indicating an effect of sequence variation on amplification. To estimate the bias caused by this effect, the qPCR efficiencies of cloned korB fragments of plasmids representing the different subgroups and pKS208 were compared (Fig. 2). The amplification efficiencies of the korB gene variants did not differ significantly, as calculated from slopes of linear regression curves and a test for parallelism (P = 0.07; PROC GLM, SAS 9.3 statistical package). The efficiency of the PCR was, on average, 88%, and the detection limit was 10 copies. We achieved the lowest cycle thresholds with given initial concentrations with korB of pB10 and the highest with pKS208 and pKJK5. This variation between plasmids resulted in standard deviations of 0.4 to 0.5 log unit at environmentally relevant concentrations (Fig. 3). The standard deviation increased as the initial concentration decreased. Around one-third of this error could be explained by variation attributable to plasmid extraction and serial dilution of quantification standards, as shown by independent replicate preparations (Fig. 3).

Table 1.

Alignment of primers and TaqMan probes with corresponding sequences of korB genes which represent all of the variants described so far of the known IncP-1 plasmid subgroups

Plasmid (accession no.)	Sga	Sequence (5′−3′)
		Forward primerb	Reverse primer	TaqMan probec
		TCATCGACAACGACTACAACG (F)	TTCTTCTTGCCCTTCGCCAG (R)	TCAGYTCRTTGCGYTGCAGGTTCTCVAT (P)
		TCGTGGATAACGACTACAACG (Fz)	TTYTTCYTGCCCTTGGCCAG (Rge)	TSAGGTCGTTGCGTTGCAGGTTYTCAAT (Pgz)
			TTCTTGACTCCCTTCGCCAG (Rd)
pB3 (AJ639924)	β1	TCATCGACAACGACTACAACG	TTCTTCTTGCCCTTCGCCAG	TCAGCTCGTTGCGTTGCAGGTTCTCGAT
pBP136 (AB237782)	β1	TCATCGACAACGACTACAACG	TTCTTCTTGCCCTTCGCCAG	TCAGCTCATTGCGTTGCAGGTTCTCGAT
pA1 (AB231906)	β2	TCATCGACAACGACTACAACG	TTCTTCTTGCCCTTCGCCAG	TCAGCTCGTTGCGTTGCAGGTTCTCGAT
pA81 (AJ515144)	β2	TCATCGACAACGACTACAACG	TTCTTCTTGCCCTTCGCCAG	TCAGTTCGTTGCGTTGCAGGTTCTCGAT
pAKD4 (GQ983559)	δ	TCATCGACAACGACTACAACG	TTCTTGACTCCCTTCGCCAG	TCAGTTCATTGCGCTGCAGGTTCTCCAC
pIJB1 (JX847411)	δ	TCATCGACAACGACTACAACG	TTCTTGACTCCCTTCGCCAG	TCAGTTCATTGCGCTGCAGGTTCTCCAC
pBS228 (AM261760)	α	TCATCGACAACGACTACAACG	TTCTTCTTGCCCTTCGCCAG	TCAGCTCGTTGCGTTGCAGGTTCTCGAT
pTB11 (AJ744860)	α	TCATCGACAACGACTACAACG	TTCTTCTTGCCCTTCGCCAG	TCAGCTCGTTGCGTTGCAGGTTCTCGAT
pQKH54 (AM157767)	γ	TCATCGACAACGACTACAACG	TTTTTCTTGCCCTTGGCCAG	TCAGGTCGTTGCGTTGCAGGTTCTCAAT
pKS208 (JQ432564)	γ	GGATCGACAACGACTACAACG	TTCTTCTTGCCCTTGGCCAG	TGAGGTCGTTGCGTTGCAGGTTTTCAAT
pMCBF1 (AY950444)	ζ	TCGTGGATAACGACTACAACG	TTCTTCTTACCCTTCGCCAG	TCAGCTCGTTGCGTTGCAGGTTCTCAAT
pMCBF6 (EF107516)	ζ	TCGTGGATAACGACTACAACG	TTCTTCTTACCCTTCGCCAG	TCAGCTCGTTGCGTTGCAGGTTCTCAAT
pKJK5 (AM261282)	ε	TCATCGACAACGACTACAACG	TTTTTCCTGCCCTTGGCCAG	TCAGCTCGTTGCGTTGCAGGTTCTCGAT
pAKD16 (JN106167)	ε	TCATCGACAACAACTACAACG	TTCTTCTTGCCCTTGGCCAG	TCAGCTCGTTGCGTTGCAGGTTCTCGAT

^aSg, IncP-1 subgroup.

^bPrimer- or TaqMan probe-to-target mismatches are underlined.

^c5′ 6-carboxyfluorescein (FAM), 3′ 6-carboxytetramethylrhodamine (TAMRA).

Fig 2.

Cycle threshold values related to initial copy numbers of cloned korB genes from six plasmids (pB10, pEST4011, pKJK5, pMCBF1, pQKH54, and RP4) representing the different IncP-1 subgroups and pKS208. Solutions of korB fragments ligated into pGEM-T (Promega) with similar concentrations (optical density at 260 nm [OD₂₆₀], 0.5; OD₂₆₀/OD₂₈₀ ratio, >1.8) were serially diluted and applied to the developed real-time PCR 5′ nuclease assay. The gray area and dashed lines indicate 95% confidence limits and prediction limits of linear regression analysis, respectively.

Fig 3.

Standard deviations of korB gene quantification with respect to the initial concentration, calculated from regression curves, of seven korB variants representing the IncP-1 subgroups and pKS208 (solid line) and for four independent preparations of the quantification standards for plasmids pEST4011 and pKJK5, respectively.

The korB assay was used to analyze the abundance and dynamics of IncP-1 plasmids in a large pesticide-degrading biofilter operated on a farm near Kortrijk, Belgium, which was previously characterized (23). Four spatially separated plots in the biofilter were sampled three times over a season with continuing applications of pesticide wastes, before start-up (March), during processing (July), and after closedown (September). The biofilter, composed of coco chips, straw, and soil, received 37 different active compounds, including 16 halogenated aromatics (23). Total DNA was extracted from 0.5 g of 2-mm-sieved biofilter samples by using the FastPrep FP120 bead beating system for cell lysis and the FastDNA SPIN Kit for Soil (MP Biomedicals, Santa Ana, CA) and purified by using the GENECLEAN Spin Kit (MP Biomedicals). The korB copy numbers were determined as described above, and 16S rRNA genes were quantified by using the qPCR system developed by Suzuki et al. (24). In addition, trfA copies of IncP-1ε plasmids were measured as previously described (10), to test whether more plasmids are detected by the qPCR targeting korB. A dilution series of plasmid pKJK5 was used as a common quantification standard for both korB and trfA. Gene copy numbers were related to 16S rRNA gene copy numbers to account for differences in bacterial concentration or amplification efficiency in the samples. The korB assay detected two to four times as much IncP-1 plasmids as the trfA IncP-1ε assay, which was significant for all three samplings and which confirmed the intended broader assay specificity (Fig. 4). Concomitant with continued pesticide applications, the relative abundance of IncP-1 plasmids in the bacterial biofilter community increased from March until September, reaching values of up to 0.2%. This might indicate an important contribution of IncP-1 plasmids to pesticide degradation, as previously suggested on the basis of degradative pathways located on several IncP-1 plasmids (6).

Fig 4.

Abundance of korB of IncP-1 plasmids and trfA of IncP-1ε plasmids in a biofilter that degraded various pesticides during the growing season. Plasmid copy numbers were related to bacterial 16S rRNA gene copy numbers (rrn) quantified from the same sample. Different letters indicate significant differences (Tukey test; spatially separated compartments within the biofilter (20 by 1.2 m) were used as replicates: n = 3 for March, n = 4 for July and September).

Thus, our results indicate that we have successfully developed a rapid and sensitive method which allows the quantification of all known IncP-1 subgroups in environmental samples despite their diverse backbones. Its application will give further insight into the relative abundance of IncP-1 plasmids in bacterial populations and their role in the spread of resistance genes, degradation pathways of recalcitrant compounds, or other traits of the horizontal gene pool which allow a fast response of bacterial subpopulations to irregular environmental changes.