Analysis of plasmid profiles of Aeromonas salmonicida isolates by pulsed field gel electrophoresis

Abstract

Plasmid constitutions of Aeromonas salmonicida isolates were characterised by flat-bed and pulsed field gel electrophoresis. Resolution of plasmids by pulsed field gel electrophoresis was greater and more consistent than that achieved by flat-bed gel electrophoresis. The number of plasmids separated by pulsed field gel electrophoresis varied between A. salmonicida isolates, with five being the most common number present in the isolates used in this study. Plasmid profiles were diverse and the reproducibility of the distances migrated facilitated the use of principal components analysis for the characterisation of the isolates. Isolates were grouped according to the number of plasmids supported. Further principal components analysis of groups of isolates supporting five and seven plasmids showed a spatial separation of plasmids based upon distance migrated. Principal components analysis of plasmid profiles and antimicrobial minimum inhibitory concentrations could not be correlated suggesting that resistance to antimicrobial agents is not associated with either one plasmid or a particular plasmid constitution.

Aeromonas salmonicida, Pulsed field gel electrophoresis, Principal components analysis, Plasmid

1 Introduction

A. salmonicida isolates support a range of plasmids varying in molecular size from a 5–6 kb group to larger molecular mass entities [1, 2]. Furthermore, specific plasmids have been designated R plasmids [2]. Plasmid profiles of A. salmonicida when separated were considered by Belland and Trust [1] sufficiently unique to act as epidemiological markers of individual isolates. In contrast, Bast et al. [2] observed that plasmid profiles were relatively homogeneous and therefore unsuitable as epidemiological markers.

Restriction digestion and mapping of individual plasmids in conjunction with plasmid profiling offers potentially greater discrimination between strains. Separation of plasmids by conventional flat-bed agarose gel electrophoresis is insufficient to enable adequate statistical analysis, and other molecular biological techniques such as random amplified polymorphic DNA have correlated poorly with plasmid profiling [3]. Some success has been achieved using restriction digestion of a number of small conserved plasmids [3].

In this study we report the characterisation of A. salmonicida plasmids using pulsed field gel electrophoresis (PFGE), a technique recently applied to separate chromosomal macrorestriction fragments from A. hydrophila[4]. The greater separation which can be achieved using this technique permits the use of enhanced statistical analysis for delineation of plasmid groups. Discrimination between plasmid profiles is undertaken using principal components analysis (PCA) [5], which generates a two or more dimensional spatial orientation of isolates derived from the plasmid constitution of each isolate. An analysis of this form can be summarised graphically and may be correlated with antimicrobial sensitivities.

2 Materials and methods

2.1 Bacterial cultures

Eighteen isolates of A. salmonicida were obtained from the bacterial collection of the Fish Diseases Laboratory, Weymouth, UK (Table 1). These isolates were selected to represent the different characteristics and morphologies associated with A. salmonicida. Isolates were stored on slopes of tryptone soya agar (Lab M, Bury, UK) at 4°C and subcultured every 6 months.

1

Origins and MICs of the isolates in this study

FDL reference	Source	MIC judged resistant
NCIMB 833	Brook trout	OTC >8 μg/ml
13:73	Lake trout, Finland	AMX >128 μg/ml
102:77	Atlantic salmon, Sweden	TS >8 μg/ml
1:78	Brown trout, kidney
22:82	Rainbow trout
24:83	Unknown
31:93	Atlantic salmon	OTC >8 μg/ml, TS >8 μg/ml, FLM >4 μg/ml
32:93	Rainbow trout	OTC >8 μg/ml, TS >8 μg/ml, FLM >4 μg/ml
15:94	Atlantic salmon	OTC >8 μg/ml, TS >8 μg/ml
17:94	Atlantic salmon, Wales
18:94	Rainbow trout	AMX >128 μg/ml
69:94	Pike, kidney	OTC >8 μg/ml
120:94	Rainbow trout, kidney	OTC >8 μg/ml, OA >8 μg/ml, SFX >2 μg/ml, FLM >4 μg/ml, FUR >4 μg/ml
121:94	Rainbow trout
127:94	Rainbow trout	OTC >8 μg/ml, OA >8 μg/ml, TS >8 μg/ml, SFX >2 μg/ml, FLM >4 μg/ml, FUR >4 μg/ml
10:95	Rainbow trout	OTC >8 μg/ml, OA >8 μg/ml, TS >8 μg/ml, SFX >2 μg/ml, FLM >4 μg/ml, FUR >4 μg/ml
41:95	Grayling	OTC >8 μg/ml, TS >8 μg/ml, FUR >4 μg/ml
44:95	Grayling	OTC >8 μg/ml

Antimicrobials: amoxycillin (AMX), flumequine (FLM), furazolidone (FUR), oxolinic acid (OA), oxytetracycline (OTC), sarafloxin (SFX), potentiated sulfonamide (TS).

2.2 Isolation of plasmid DNA

Aliquots (10 ml) from separate multiple replicate overnight cultures of A. salmonicida isolates were harvested by centrifugation (3000×g, 10 min). Pellets were resuspended in 200 μl TEG buffer (50 mM glucose, 10 mM EDTA, 25 mM Tris-HCl pH 8.0, 4 mg/ml lysozyme) and incubated at 20°C for 5 min; 200 μl alkaline SDS solution (0.2 M NaOH, 1% sodium dodecyl sulfate) was added followed by further incubation at 20°C for 5 min. 150 μl 5 M potassium acetate, pH 4.8, was added and the samples incubated at 0–4°C for 5 min. Cell debris was removed by centrifugation and the supernatants extracted using TE buffer-saturated phenol (25 mM Tris-HCl pH 8.0, 10 mM EDTA). Plasmid DNA was ethanol-precipitated and resuspended in TE containing RNase A (1 mg/ml).

2.3 Flat-bed agarose gel electrophoresis of plasmid DNA

Agarose gels of 1% agarose (molecular biology grade) in 1×TBE buffer (90 mM Tris-borate, 1 mM EDTA, pH 8.0), ethidium bromide (0.5 μg/ml) were loaded with plasmid samples and tracking dye (0.25% (w/v) bromophenol blue, 40% (w/v) sucrose in distilled water). Electrophoresis was performed at 100 V for 3 h and gels were examined under UV illumination.

2.4 Pulsed field gel electrophoresis of plasmid DNA

Replicate agarose gels (1.25% (w/v) agarose (molecular biology grade) in 0.5×TBE) were loaded with plasmid samples and wells sealed using molten agarose (0.5% (w/v) in 0.5×TBE). PFGE was optimised to separate the plasmids (6 V/cm, 120° angle, 8°C, 18 h using a ramped pulse time of 1–8 s). Gels were stained using ethidium bromide (0.5 μg/ml in distilled water) for 30 min prior to destaining for at least 90 min in distilled water and visualisation.

2.5 Minimum inhibitory concentration (MIC) determination

Mueller-Hinton agar (Oxoid) was prepared according to the manufacturer's instructions. Appropriate drug dilutions were prepared in petri dishes. Plates containing diluted antimicrobials were inoculated with A. salmonicida and MICs determined after incubation (22°C, 48 h).

2.6 Principal components analysis

PCA (Statview 11 software) was performed on all isolates and on those with five and seven plasmids separately. Principal components were extracted and used to provide groupings according to plasmid content and correlations with MIC values.

3 Results

3.1 Pulsed field and flat-bed separations of plasmid DNA

Typical plasmid separations achieved by flat-bed agarose gel electrophoresis and PFGE are shown in Fig. 1. PFGE successfully separated a range of Aeromonas plasmids (Fig. 1A). Resolution of both large and small molecular mass plasmids was achieved on the same gel, which was scanned and the distances of the plasmids from the origin measured for use in statistical analysis. In contrast, flat-bed agarose gel electrophoresis facilitated separation of smaller Aeromonas plasmids within a shorter time than PFGE but with poorer resolution (Fig. 1B).

1

Separation of plasmids from A. salmonicida isolates. A: Pulsed field gel electrophoresis. Plasmids were separated using a ramped pulse time of 1–8 s for 22 h (1.25% agarose gel). The sizes of molecular mass standards (EcoRI/HindIII digested λ DNA) are shown in kb on the right. Plasmid DNA from the following isolates was loaded as follows: (1) 18:94, (2) 17:94, (3) 24:83, (4) NCIMB833, (5) 102:77, (6) 1:78, (7) A. sobria (included as a non-A. salmonicida representative). B: Flat-bed agarose gel electrophoresis. Plasmids were separated on a 1% agarose gel at 100 V for 3 h. The sizes of molecular mass standards (EcoRI/HindIII digested λ DNA) are shown in kb on the right. Plasmid DNA from the following isolates was loaded as follows: (1) 18:94, (2) 17:94, (3) 24:83, (4) NCIMB833, (5) 102:77, (6) 1:78, (7) A. sobria (included as a non-A. salmonicida representative).

The majority of the isolates used in this study supported a small number of plasmids of 5–8 kb. Many isolates contained a number of plasmids of approximately 20 kb and some isolates supported plasmids as large as 100 kb (Fig. 1A). The number of plasmids supported by the isolates used in this study was counted and the distances these plasmids migrated upon PFGE were measured for three independent determinations. Mean values were used in the PCA calculations. Repeated plasmid extractions over a period of 6 months and comparison of profiles from late exponential phase (18 h) and early stationary phase cultures revealed little difference in plasmid composition.

3.2 PCA of plasmids

Isolates used in this study contained a varying number of plasmids (4–8). Seven isolates contained five plasmids and five isolates contained seven plasmids (including two phenotypically atypical isolates 102:77 and 17:94). Principal components were selected and analysed to give groupings of isolates based on plasmid number (Fig. 2).

2

Principal components analysis of A. salmonicida plasmids. A plot of all isolates in relation to the selected principal components 1 and 2. Clusters A, B, C, D and E correspond to 8, 7, 6, 5 and 4 plasmids respectively.

3.3 MICs

Standardised techniques for susceptibility testing of fish pathogens do not yet exist. Consequently there are no predetermined MIC values to categorise isolates as susceptible, moderately resistant or resistant as there are in clinical medicine. An isolate was considered to be resistant to a certain compound only when its MIC was greater than 2 μg/ml. Table 1 lists the isolates and the compounds to which they were judged resistant. All isolates were considered resistant to erythromycin and sensitive to chloramphenicol, florphenicol and thiamphenicol. Plasmid number and antibiotic sensitivity could be correlated in only one incidence, that of isolates containing seven plasmids and furazolidone.

4 Discussion

PFGE has been employed effectively in the molecular characterisation of Borrelia burgdorferi isolates which demonstrated unique plasmid constitutions [6]. Resolution of Aeromonas plasmid separations achieved by PFGE appear to exceed those achieved by flat-bed agarose gel electrophoresis. In common with isolates used in previous studies [1–3] the isolates used here supported varying numbers of plasmids of diverse size. In contrast to previous studies, where a single isolate containing seven plasmids was found [1], this study revealed five isolates containing seven plasmids and one isolate containing eight plasmids.

Many studies have emphasised the greater separation of DNA achieved using PFGE [7], however, few studies have used PCA to ensure comprehensive interpretation of separations. In this study PCA has provided a summary of spatial relationships of isolates and communities through analysis of plasmid constitution which could be used as a more reliable epidemiological tool than phagetyping [8] or serotyping [9].

The lack of correlation between plasmid profile and antimicrobial MICs gives insight into the relationship of plasmid constitution and antimicrobial resistance amongst A. salmonicida isolates. As variation in MICs among isolates does not correlate with plasmid constitution and size (largely between 5 and 100 kb in this study) it would appear that drug resistance does not correlate with specific resistance plasmids or plasmid profiles in these bacteria. The findings of Aoki et al. [10] who determined R plasmid molecular masses to be 7.6 and 29 MDa (approximately 7 and 26 kb) and Belland and Trust [1] who found R plasmid sizes of 26–30 kb and 95 kb in isolates from the USA provide further evidence that resistance to antimicrobial agents is not associated with either one plasmid or a particular plasmid constitution.

Antimicrobial resistance amongst A. salmonicida isolates appears to be mediated by many plasmids of potentially differing character some of which are transferable as demonstrated by Inglis et al. [11]. Furthermore, the possibility of chromosome-encoded resistance factors amongst A. salmonicida should not be discounted as Davies [12] suggests that chromosomal material may be the source of cryptic antibiotic resistance genes and chromosomal resistance factors are frequently found in other bacteria [13].