Introduction

Pseudomonas aeruginosa, a Gram-negative bacteria, which is found to inhabit soil, water, and marine environments (Kimata et al., 2004; Khan et al., 2007), produces fluorescent compounds called pyoverdines that these organisms use to chelate and solubilize iron. Three different classes of pyoverdine have been identified, each recognized on the outer membrane of the organism by ferripyoverdine transporters (Cornelis et al., 1989; Ernst et al., 2003).

Several genes have been employed as taxonomic markers for the positive identification of P. aeruginosa. Some of these genes include 16S rDNA, gyrB, exoA, and algD; PCR-specific primers have also been designed for the amplification and sequencing of these genes (Laguerre et al., 1994; Kasai et al., 1998; Anzai et al., 2000; Qin et al., 2003).

Clinically, this organism plays a significant role in the survival rates of affected patients; however, this role becomes increasingly important when the organism is detected earlier and specifically identified.

Biochemical methods have, to a certain extent, facilitated the identification of P. aeruginosa; however, some of these methods are time consuming and may not be very accurate. For example, as a result of work carried out in this study on certain P. aeruginosa clinical, freshwater, and environmental isolates, as well as some Pseudomonas sp. isolated from the North Sea that were initially thought to be “P. aeruginosa,” while one isolate was identified as P. aeruginosa by the commercial method employed in this study, the other samples turned out to be false-positives after a series of experimental PCR analysis.

Much would be gained by an immuno-deficient patient who suffers from a burn wound, cancer, or cystic fibrosis and was infected by this organism if the pathogen could be detected in time, as prompt isolation and identification of this organism and other Gram-negative pathogens are often needed for patients to meet clinical trial entrance criteria (Qin et al., 2003; Xu et al., 2004). The present work used pyoverdine production, biochemical identification and antibiotyping, OprI and OprL gene amplification, serotyping, protease assays, and the more specific ferripyoverdine receptor (fpvA and fpvB) typing by PCR as a means to identify P. aeruginosa.

Experimental procedures

Preliminary Identification Methods. Preliminary identification of bacterial isolates was carried out based on growth at 37°C, pyoverdine production, biochemical identification, and antibiotyping.

Growth at 37°C. P. aeruginosa grows at an optimum temperature of 37°C but has also been observed to grow at both 42°C and 4°C (Iglewski, 1991); all the strains under study grew at 37°C. While growth at 42°C was tested for some isolates, there was no observable growth at this temperature, and after other assays, these isolates turned out not to be P. aeruginosa.

Pyoverdine production. Pyoverdine is a yellowish-green compound produced when bacteria, especially florescent Pseudomonas sp., are grown under conditions of iron limitation (Budzikiewicz, 1995; 2001; Neilands, 1995). A majority of the strains under study had been previously characterized (Pirnay et al., 2002) by pyoverdine production and growth stimulation assays, while others that had not been previously characterized were tested for pyoverdine production and selected for further analysis. However, pyoverdine-negative strains were also further analyzed due to the existence of pyoverdine-negative P. aeruginosa isolates (De Vos et al., 2001).

Protease assay. This assay tests whether bacteria produce protease exoenzymes (e.g., caseinase). Skim milk agar (SMA) was used for this purpose; bacteria that produce proteases hydrolyze casein into its constituent amino acids, producing a zone of clearing typical of microbes that produce proteases capable of degrading milk proteins. Overnight bacteria cultures were streaked on SMA plates and incubated at 37°C; these plates were then examined for a zone of transparency produced by the action of proteases for the detection and isolation of positive strains.

Antagonism against Pythium. P. aeruginosa produces phenazine compounds (Anjaiah et al., 1998), which have been found to possess anti-fungal activities. A zone of inhibition of fungal growth is observed when cultures of P. aeruginosa are incubated with fungal mycelia. This test was performed using GCA (glucose added to casamino acid medium), GCA plus iron, and PDA (potato dextrose agar) plates. On these plates containing plugs of Pythium oomycetes were inoculated 5 µl of bacteria cultures grown overnight at 37°C in CAA liquid medium. P. aeruginosa antagonism against Pythium was only observed on GCA plates.

Antibiotyping. Biochemical identification (Vitek GNI, bioMerieux) involved the testing of the production of certain biochemical compounds and possession of the following enzymatic activities in Gram-negative bacteria (e.g., beta-glucosidase, lipase, urease, alpha-galactosidase, phosphatase, and production of H₂S). In addition, this system was used to test the susceptibility of bacterial isolates to antibiotics.

PCR. The following genes were amplified for further bacteria identification: the oprI and oprL genes (De Vos et al., 1997; 1998) and the ferripyoverdine receptor genes (fpvA and fpvB) for the different pyoverdine types produced by P. aeruginosa (De Chial et al., 2003; Ghysels et al., 2004). In addition, primers for the oprI, oprL, and ferripyoverdine receptor genes (fpvA and fpvB) were designed and used for identification, and additional primers were designed to amplify the ferripyoverdine genes in the ‘difficult-to-type’ P. aeruginosa isolates, which tested negative when primers for ferripyoverdine typing by M-PCR (De Chial et al., 2003; Ghysels et al., 2004) were used.

DNA preparation method. DNA was extracted using a PCR template preparation kit (Roche) according to the manufacturer’s instructions. Chromosomal DNA preparation was performed by boiling a single colony of sample bacteria in autoclaved, filter-sterilized water, and 2 µl of extracted DNA was used as template for M-PCR amplification.

PCR cycling conditions. PCR cycling conditions differed for the amplified genes as follows:

M-PCR for OprI and OprL gene amplification. The amplification program was set at 25 to 30 cycles of denaturation at 94°C for 40 s, annealing at 57°C, and extension at 72°C for 50 s (De Vos et al., 1997). All strains tested were OprI-positive and others were both OprI- and OprL-positive. The PCR procedure was repeated with modifications to the annealing temperature (55°C instead of 57°C) for those that were OprI -positive and OprL-negative and, at this temperature, OprL-negative strains tested positive.

M-PCR for fpvA and fpvB gene amplification. Amplification was carried out separately (see pictures 1 and 2) using the same amplification program of an initial denaturation at 94°C for 3 min, followed by 30 cycles of denaturation at 94°C for 30 s, annealing at 55°C for 30 s, elongation at 72°C for 30 s, and a final extension at 72°C for 10 min (De Chial et al., 2003). An amplification program with the following conditions: initial denaturation at 94°C for 3 min followed by 35 cycles of denaturation at 94°C for 30 s, annealing at 50°C for 30 s, elongation at 72°C for 2 min and a final extension at 72°C for 10 min were used when carrying out simultaneous amplification of both fpvA and fpvB genes (see picture 3) in a reaction mixture of 50 µl comprising 23.75 µl sterile distilled water, 5 µl PCR buffer, 5 µl MgCl₂, 4 µl dNTP mix (10 mM), 1 µl each of 20 µM primers, and 0.25 µl AmpliTaq gold polymerase enzyme. All primers for fpvA and fpvB gene amplifications were synthesized by Sigma. Eight microliters amplified PCR product were put on a 1% (wt/vol) agarose gel for electrophoresis at 80 V for 90 min, and visualization of the product was performed after staining with ethidium bromide for 12 min on a UV transilluminator.

M-PCR was performed using P. aeruginosa PAO-1, ATCC 27853, 7NSK2, and 59.20 DNA as positive controls for OprI, OprL, fpvA, and fpvB gene amplification and a negative control without DNA.

Serotyping. Bacterial isolates were serotyped by slide agglutination according to the international serogrouping schema for P. aeruginosa using a panel of 16 type-O monovalent antisera following the manufacturer’s protocol, although not all isolates could be serotyped.

Figure 1

Table 1. List of primers used in this study

1-4: Degenerate primers were designed following a Clustal X alignment of all type-II fpvA gene sequences in five P. aeruginosa strains, namely, 7NKS2, ATCC27853, MSH, 1-60, and 2-164; these sets of degenerate primers were designed in this study for PCR amplification of the fpvA gene in the ‘difficult-to-type’ (by PCR) P. aeruginosa strains. Both primer sets were used to perform PCR, and amplification products of approximately 500 and 250 bp, respectively, of the fpvA gene were sequenced from these strains.

5-12: Multiplex PCR primers for the identification of fpvAI, fpvAII, and fpvAIII, six of which were designed for the simultaneous amplification ^{of the different fpvA genes, namely, primers 1F and 1R for the amplification of a 326-bp fragment corresponding to fpvAI; primers 2F and 2R for the amplification of an 897-bp fragment corresponding to fpvAII; and primers 3F and 3R for the amplification of a 506-bp fragment corresponding to fpvAIII (De Chial et al., 2003). Primers PA4168 SC Fw and PA4168 SC Rv were developed to amplify a 562-bp fragment of PA4168 (alternative ferripyoverdine receptor gene called fpvB in P. aeruginosa).}

13-16: Primers for the amplification of fpvA types IIa and IIb.

Figure 2

Table 2. List of test strains used in this study

Some of the strains above were previously identified as P. aeruginosa (Pirnay et al., 2002) and analyzed by pyoverdine production and growth stimulation assays. The positive control P. aeruginosa strains from which the fpvA gene sequences were aligned included strains MSH (Smith et al., 2005), ATCC 27853, 1-60, 2-164 (Spencer et al., 2003), and 7NSK2 (de Chial et al., 2003). These strains were all type-II pyoverdine-producing P. aeruginosa; other control strains used were PAO-1 (Stover et al., 2000) and 59.20 ( De Chial et al., 2003).

Results

Pyoverdine production. Bacterial strains were cultured on cetrimide agar (CA) plates and incubated at 37°C for 24 to 48 h. The majority of these isolates produced a yellowish green compound (pyoverdine) on CA plates, although others did not; however, these strains were selected for further analysis, including OprI, OprL, fpvA, and fpvB gene amplifications (De Vos et al., 1997; 1998; De Chial et al., 2003; Ghysels et al., 2004). Some pyoverdine-negative isolates tested positive for these genes and were later identified as P. aeruginosa. This identification was based on the outcome of a BLAST search against the NCBI database (http://www.ncbi.nlm.nih.gov/blast) using the sequences that resulted from the amplified genes. These isolates tested positive for OprI, OprL, fpvA, and fpvB genes, while other pyoverdine-negative isolates that tested positive for only the OprI and OprL genes were identified as P. putida. This finding is in agreement with other authors that non-Pseudomonas aeruginosa strains also tested positive with the OPR primers (Qin et al., 2003). The observations from this study showed that the majority of false-positive isolates never tested positive for the fpvA and fpvB genes and those that did were the result of non-specific amplification.

Figure 1. Multiplex PCR (M-PCR) amplification of the three fpvA genes, namely, fpvA type I, fpvA type II, and fpvA type III from different P. aeruginosa isolates. The marker is a 100-bp ladder (De Chial et al., 2003)

Figure 3

Figure 1

Figure 4

Figure 2. gene amplification (562 bp). The marker is a 200-bp ladder ( Ghysels ., 2004)

Figure 5

Figure 3. amplification in isolates.

Figure 6

Figure 4. amplification in non- isolates. Lanes 1-9: 1-8: test strains, 1: SWPAD40a, 2: SWPAD40b, 3: SWPAD10.20a, 4: SWPAD10.20b, 5: SWPAD10.20c, 6: SWPAD36a, 7: SWPAD36b, 8: SWPAD36c, and 9: negative control (200-bp ladder).

Figure 5. fpvA gene amplification (~500 bp) for the ‘difficult-to-type’ P. aeruginosa isolates using primer set 1 and 2 (see Table 1). SL: 200-bp Smart Ladder. Lanes 1-8; 1: Mi162, 2: Is 579, 3: So 122, 4: Lo 059, 5 and 6: Us376 and EVA 3067, respectively, 7: negative control, 8: ATCC 27853 (positive control).

Figure 7

Figure 5

These isolates were negative for this primer set; however, they tested positive for primer set 3 and 4 (see Figure 6).

Figure 6. fpvA gene amplification (~250 bp) for the ‘difficult-to-type’ P. aeruginosa isolates using primer set 3 and 4 (see Table 1). Lanes 1-8; 1: ATCC 27853 (positive control), 2: Us376, 3: EVA 3067 , 4:SWPAD40a, 5: SWPAD10.20a, 6: SWPAD36a, 7: Pr 332 , 8: negative control.

Figure 8

Figure 6

Protease assay. The majority of the bacterial isolates tested that were later identified as P. aeruginosa were positive, while only a few isolates tested negative.

Antagonism against Pythium. The majority of the bacterial isolates tested were positive for this test.

Serotyping. All bacterial isolates were serotyped and found to belong to one of 16 serotype groups of P. aeruginosa.

Discussion

P. aeruginosa is the causative agent for nosocomial infections, which has and is still being extensively studied. Research has recently been geared towards an understanding of how these groups of bacteria are able to thrive in environments that do not supply their nutritional needs, such as a requirement for iron. This element is essential to these organisms and has allowed an advance in the understanding of these organisms. Under conditions of iron limitation, these pathogens produce two siderophores or iron chelators; they can also use siderophores (heterologous siderophores) produced by other organisms within their environment (Cornelis and Matthijs, 2002) and express receptors for these heterologous siderophores. Approximately 35 TonB-dependent receptors are present in the genome of P. aeruginosa (which itself produces only two siderophores), suggesting that these organisms are well-equipped for obtaining essential nutrients. fpvA (Poole et al., 1993) and fpvB are two such receptors, and it would be interesting to know the reason for this redundancy as it provides the basis for the identification of this group of bacteria. During PCR analysis, for example, specific amplifications are only observed when PCR-specific primers, designed to amplify these genes, are used, often resulting in specific identification of P. aeruginosa and non-specific amplifications of other bacteria species. Even when the bands that arise as a result of these non-specific amplifications are sequenced, there is no significant identification of the ferripyoverdine receptor genes from P. aeruginosa, suggesting that these genes are unique to this organism.

Amplification of the ferripyoverdine receptor genes in the ‘difficult-to-type’ P. aeruginosa strains was performed using primers designed in this study (primer sets 1 & 2 and 3 & 4); these primers were designed from the alignment of the fpvA gene sequences of P. aeruginosa strains 7NSK2, MSH, 2-164, 1-60, and ATCC 27853). Specific amplification was only observed for P. aeruginosa strains. The PCR products for the different amplifications (in addition to the non-specific amplifications, Pseudomonas sp. strain SWPAD40a lane 4, picture 6) were purified and sequenced; these sequences were then used to search against the NCBI BLAST database, and the results are shown in the table below.

Figure 9

Table 3. BLAST search results for strains.

Following the alignment of sequences of those strains that were 99-100% identical at the nucleotide level to the ferripyoverdine receptor sequences of other type-II pyoverdine-producing reference P. aeruginosa strains, homology trees were constructed using DNA-manager software, and some of these strains were observed to cluster separately.

Figure 7. Homology tree constructed using the DNAMAN software shows relatedness of the fpvA gene sequence in the test P. aeruginosa strains Mi162, Pr332, EVA3067, LO059, Is579, and SO122 and reference strains which included MSH (Smith et al., 2005), ATCC 27853, 1-60, 2-164 (Spencer et al., 2003), and 7NSK2 (De Chial et al., 2003). PA denotes “P. aeruginosa.”

Figure 10

Figure 7

It has been hypothesized that ferripyoverdine receptors are targeted by antibiotics and bacteriophages and serve as entry points (Wayne and Neilands, 1975; Guerinot, 1994; Andrews et al., 2003; Smith et al., 2005). P. aeruginosa alone will go to great lengths to evade the adverse effects of antibiotics and bacteriophages (Budzikiewicz, 2001). One way that this may be accomplished, as presented in this work, may involve the ferripyoverdine receptors (as shown by the separate clustering of some of the test strains on the homology tree. Siderophore biosynthesis represents an attractive antibiotic target (Quadri, 2000), and fpvA has been proposed to drive diversity at the pyoverdine locus (Smith et al., 2005). If this is true, it would justify a correlation between amino acid sequence diversity of immunogenic bacterial proteins and evasion of host immune defense mechanisms (Tummler and Cornelis, 2005). For non-P. aeruginosa isolates (false-positives from antibiotyping), the OprI sequences were 99% identical at the nucleotide level to the OprI of P. Putida; these isolates were identified as “98.45% Trés bonne identification” to P. aeruginosa. Clinically, this would be disastrous in situations where only antibiotyping is used as the sole diagnostic tool, this work suggests the addition of other identification methods, such as experimental PCR analysis for the various ferripyoverdine receptor genes (this study), to antibiotyping (Kimata et al., 2003; Khan et al., 2007).

It would be interesting to study various aspects of these organisms, but this may be difficult because of the phenotypic (and/or genotypic) variations that exist within these bacterial species (Reid and Kirov, 2004; Zeng and Kim, 2004). Knowledge of the pyoverdine region, including the complete fpvA gene sequence (Poole et al., 1993; McMorran et al., 1996; Lamont and Martin, 2003) in the difficult-to-type P. aeruginosa strains, would be important to evolutionary studies to determine what portion of this region is favored by evolution. Such knowledge could provide rational starting points for the design of novel antimicrobial agents, especially in connection with their iron requirement (Miller et al., 1993; Clarke et al., 2001). Our future studies plan to address this issue.

Acknowledgement