DNA Fingerprinting of Ralstonia paucula by Infrequent-Restriction-Site PCR and Randomly Amplified Polymorphic DNA Analysis

Abstract

Ralstonia paucula (formerly CDC group IV c-2) is an environmental organism that can cause serious human infections, occasionally clusters of nosocomial infections. In the present work, 26 strains of R. paucula (4 from the American Centers for Disease Control and Prevention collection, 10 from the Belgian Laboratorium voor Microbiologie [LMG] collection, and 12 French clinical isolates) were analyzed with infrequent-restriction-site PCR and randomly amplified polymorphic DNA analysis. Both techniques accurately distinguished between collection strains. Two close patterns obtained for all the French isolates suggested a clonal strain. Two LMG collection strains originating from human sources in the United States also showed patterns close to those of French isolates.

Since 1995, the taxonomy of many gram-negative environmental bacilli has changed (3, 14). Ralstonia paucula sp. nov., formerly CDC group IV c-2, was assigned to the genus Ralstonia in 1999 (7, 12). Isolated from pool water, groundwater, bottled mineral water (2, 4, 9), and many clinical specimens, R. paucula, despite its low pathogenicity, is now recognized as an opportunist pathogen that can generate serious infections, such as septicemia, peritonitis, abscesses, etc., especially in immunocompromised patients (1, 8, 10). Furthermore, though less often isolated than Ralstonia pickettii, R. paucula can cause clusters of nosocomial infections (8). Treatment of such infections was mainly based on the use of beta-lactams such as cefotaxime, ceftriaxone, piperacillin, and imipenem (1, 8, 10).

In 1996, two clusters of CDC group IV c-2 septicemia at the Trousseau Children's Hospital were reported (8). Comparison between these isolates and eight other blood isolates obtained from six distinct hospitals in Paris area by use of randomly amplified polymorphic DNA (RAPD) analysis showed a single pattern, and consequently the isolates could not be distinguished (8). Genotyping with pulsed-field gel electrophoresis was unsuccessful, owing to DNA degradation probably caused by strong DNase activity, which was not blocked by boiling or formaldehyde fixation (7). In contrast, parallel processing of four reference strains obtained from the Centers for Disease Control and Prevention showed four distinct patterns with both RAPD analysis and pulsed-field gel electrophoresis (8). Finally, two other typing methods, PCR ribotyping and PCR restriction of the ribosomal intergenic spacer region, were unsuccessful in typing French R. paucula clinical isolates (6). A recently described method referred to as the infrequent-restriction-site PCR (IRS-PCR) assay (5, 11) has not previously been applied to Ralstonia. In this study, DNA fingerprinting with IRS-PCR, as well as RAPD, was used to distinguish between R. paucula strains from a large collection of strains kindly provided by Peter Vandammme (Laboratorium voor Microbiologie [LMG], Ghent, Belgium) (12).

Four CDC group IV c-2 strains (G6817, G3900, G608 and F4862) and 10 strains of R. paucula from the LMG collection (LMG 3244, LMG 3245, LMG 3317, LMG 3318, LMG 3319, LMG 3320, LMG 3413, LMG 3517, LMG 3518, and LMG 15544) were studied (Table 1). Twelve R. paucula blood isolates from the Trousseau Children's Hospital and five other French hospitals were also studied. These institutions are all located in the Paris area but in quite distinct districts.

TABLE 1.

List of collection strains studied

Strain	Source, geographic origin, and year
CDC G6817	Blood, Argentina, 1991
CDC G3900	Blood, Colorado, 1989
CDC G608	Blood, Colorado, 1987
CDC F4862	Bronchial specimen, Maine, 1983
LMG 3244	Human respiratory tract, United States
LMG 3245	Human sputum, United States
LMG 3317	Human, United States
LMG 3318	Human, United States
LMG 3319	Wound newborn, Sweden, 1981
LMG 3320	Humidifier in nursery, Sweden, 1975
LMG 3413	Collection Institut Pasteur, Paris, France
LMG 3517	United States
LMG 3518	United States
LMG 15544	Human urine, United States, 1985

IRS-PCR.

The XbaI adaptors, AX1 (5′-CTA GTA CTG GCA GAC TCT-3′) and AX2 (5′-GCC AGT A-3′), were constructed as previously described by Mazurek et al. (5). The PstI adaptors PS1 (5′-GAC TCG ACT CGC ATG CA-3′) and PS2 (5′-TGC GAG T-3′) were constructed as previously described by Riffard et al. (11). The adaptors, AX1 and AX2, or PS1 and PS2, were mixed in equimolar amounts and were allowed to anneal as the mixture cooled from 80 to 4°C. Oligonucleotides PX-A, PX-T, PX-G, and PX-C (5′-AGA GTC TGC CAG TAC TAG AS-3′ [S = A, T, G, or C]) were used as primers in PCR. DNA was prepared with the QIAamp DNA Mini Kit (Qiagen, Hilden, Germany) and was digested with 10 U of PstI and 10 U of XbaI for 1 h at 37°C. T4 DNA ligase, ATP, and the AX and PS adaptors were added, and the mixture was incubated at 16°C for 1 h and then at 65°C for 15 min to inactivate T4 DNA ligase. The sample was digested with 5 U of PstI and 5 U of XbaI at 37°C for 30 min to cleave any restriction sites reformed by ligation. Amplification was performed in a GeneAmp PCR System 9700 (Perkin-Elmer, Branchburg, N.J.) with an amplification profile that consisted of an initial denaturation step at 94°C for 5 min and then 30 cycles with denaturation at 94°C for 30 s, primer annealing at 60°C for 30 s, extension at 72°C for 90 s, and a final elongation at 72°C for 5 min.

RAPD procedure.

The universal primer M13 (5′-TTATGTAAAACGACGGCCAGT-3′) was used (8, 13). DNA was amplified in 50 μl of a solution containing 50 ng of DNA, 3 μM primer, 2.5 U of AmpliTaq DNA polymerase, the four deoxynucleoside triphosphates (400 μM each), 4 mM MgCl₂, 10 mM Tris-HCl (pH 8.3), and 50 mM KCl. The mixtures were subjected to 35 cycles of amplification (95°C for 1 min, 55°C for 1 min, and 72°C for 1 min) in a thermocycler (Perkin-Elmer Cetus, Norwalk, Conn.). In a further cycle, the first denaturation step at 95°C lasted for 3 min and the incubation at 72°C lasted for 5 min.

With IRS-PCR, similar results were obtained regardless of what primer (PX-A, PX-T, PX-G, or PX-C) was used together with PS1. PX-A patterns (six to eight bands ranging in size from 900 to 110 bp) are shown in Fig. 1. IRS-PCR electrophoretic patterns enabled one to distinguish between the four Centers for Disease Control and Prevention strains, while all the French clinical isolates showed the same pattern, except for one strain that showed a one-band difference (Fig. 1). Only nine different patterns were observed over the 10 strains from the LMG collection; namely, two strains (LMG 3245 and LMG 3318) showed the same pattern, this one being surprisingly close to the major pattern of French isolates, named the Trousseau IRS-PCR pattern (Fig. 1).

FIG. 1.

IRS-PCR electrophoretic patterns of R. paucula strains. Lanes 1 through 4, Centers for Disease Control and Prevention strains; lanes 5 through 12, French clinical isolates; lanes 13 through 22, LMG collection strains; and lane 23, molecular-weight marker. The Centers for Disease Control and Prevention strains showed different patterns. The French isolates showed identical (lanes 5 to 8 and 11 and 12) to similar (lanes 9 and 10) patterns. The LMG strains showed different patterns, except for two strains (LMG 3245 and LMG 3318) that showed similar patterns (lanes 14 and 16), these ones being also close to the French isolate patterns. The primers were PX-A and PS1.

RAPD was as efficient as IRS-PCR in distinguishing between R. paucula strains (three to seven bands ranging in size from 1,114 to 190 bp). It also distinguished between the four Centers for Disease Control and Prevention strains and showed a major pattern (named Trousseau RAPD pattern) for six of the eight French isolates. The two remaining others showed one-band differences (Fig. 2). Nine different patterns were observed for the 10 strains from the LMG collection; as a matter of fact, the two strains (LMG 3245 and LMG 3318) already reported to be slightly different with IRS-PCR showed here a one-band difference. These two similar patterns were also close to the Trousseau RAPD pattern (Fig. 2). One CDC strain's (G6817) pattern appeared similar to those of French isolates (Fig. 2). This showed the value of using several primers in the RAPD technique. We have established a precedent by using six primers and by showing distinct patterns for CDC and French isolates, especially with the following primer: 5′-TCACGATGCA-3′ (7).

FIG. 2.

RAPD patterns of R. paucula strains. Lanes 1 through 4, Centers for Disease Control and Prevention strains; lanes 5 through 12, French clinical isolates; lanes 13 through 22, LMG collection strains; and lane 23, molecular- weight marker. The Centers for Disease Control and Prevention strains showed different patterns, but one pattern (lane 1) appeared similar to a French isolate pattern (lanes 5 to 8 and 11 and 12). The French isolates showed identical (lanes 5 to 8 and 11 and 12) to similar (lanes 9 and 10) patterns. The LMG strains showed different patterns except for two strains (LMG 3245 and LMG 3318) that showed similar patterns (lanes 14 and 16), these ones being also close to the French isolate patterns.

A relationship was therefore suggested between three strains that we did not expect to be related, namely, the two LMG strains originating from different human sources in the United States and the French strain representative of various isolates obtained between 1993 and 1995 from different hospitals in the Paris area.

IRS-PCR was more time consuming and labor intensive than RAPD. Though a higher number of bands was observed on the IRS-PCR patterns than on the RAPD patterns, both techniques used in this study yielded well-resolved and easily compared fragment patterns in the fingerprinting of R. paucula strains.