Pseudomonas cuatrocienegasensis sp. nov., isolated from an evaporating lagoon in the Cuatro Ciénegas valley in Coahuila, Mexico

The genus Pseudomonas was described by Migula (1894). Since then, there has been a steady increase in the rate of description of species belonging to the genus Pseudomonas; there are currently more than 191 species with validly published names (Anzai et al., 2000; ). Pseudomonas species are ubiquitous in nature and can be isolated from humans, clinical samples, the plant rhizosphere, soil and water (including seawater) (Spiers et al., 2000; Yamamoto et al., 2000; Moore et al., 2006).

In the present study, a polyphasic approach was undertaken to determine the taxonomic status of nine isolates recovered from an evaporative lagoon in the Churince system, a hydrological system in the Cuatro Ciénegas valley in Coahuila, Mexico (2 ° 50.830′ N 10 ° 09.335′ W).

A large collection of bacterial isolates was obtained from surface water samples collected at two different points (10 m apart) in the lagoon referred to as Laguna Grande. Water samples (15 ml) from each site were collected in sterile BD Falcon vials (BD Biosciences) and 200 μl from each sample was inoculated on glutamate/starch/phenol red (GSP) agar plates (Kielwein, 1971). The GSP medium had the following composition (g l⁻¹): sodium l-(+)-glutamate, 10.0; starch (soluble), 20.0; potassium dihydrogen phosphate, 2.0; magnesium sulfate, 0.5; phenol red, 0.36; agar–agar, 12.0). Incubation was performed at 29 °C for 48 h. Colonies grown on different GSP plates were purified by subculture on Luria–Bertani (LB) agar (1 % tryptone, 0.5 % yeast extract, 1 % NaCl and 1.5 % agar) plates and the cultures were maintained at −80 °C in LB broth (1 % tryptone, 0.5 % yeast extract and 1 % NaCl) with 20 % (w/v) glycerol prior to analysis.

Isolates were grown in 5 ml LB broth. DNA was extracted by using a DNeasy Blood and Tissue kit (Qiagen) according to the manufacturer's instructions. Repetitive-sequence-based PCR profiles of the isolates and reference strains were determined with BOX primers and conditions as described by Versalovic et al. (1991, 1994). The PCR conditions consisted of an initial denaturing step at 95 °C for 7 min, 30 cycles of 94 °C for 1 min, 53 °C for 1 min, 65 °C for 8 min and a final extension at 65 °C for 8 min. The PCR products were electrophoresed in 1.5 % agarose gels (0.5× TAE buffer; 2.42 g Tris–HCl, 0.57 ml glacial acetic acid, 1 ml 0.5 M EDTA (pH 8.0) and 1 l distilled and deionized water) for 5 h at 180 mV (5 V cm⁻¹). The gels were stained with ethidium bromide and a digital image was obtained for further analysis. Isolates that displayed the same number of bands and similar fragment sizes were considered identical and therefore belonging to the same bacterial group or genotype. On this basis, nine isolates recovered from different water samples were identified as having identical (or very similar) repetitive-sequence-based PCR patterns (data not shown). 16S rRNA gene sequences were obtained from two of these strains, namely 1N^T and 3N, in order to define their taxonomic affiliation. The almost-complete 16S rRNA gene sequences were obtained by PCR amplification using primers 27F and 1492R and the conditions described previously (Lane, 1991). 16S rRNA gene sequences (approx. 1450 bp) were obtained (High-Throughput Sequencing Enterprise of the University of Washington, USA) and were compared with all publicly available complete 16S rRNA gene sequences of the type strains of Pseudomonas species with validly published names, using the Ribosomal Database Project (rdp) II database and software (Cole et al., 2007). An analysis based on 16S rRNA gene sequence similarities indicated that the closest relatives of strain 1N^T were Pseudomonas peli R-20805^T (98.5 %), Pseudomonas argentinensis CH01^T (97.5 %), Pseudomonas flavescens B62^T (97.4 %) and Pseudomonas anguilliseptica NCIMB 1949^T (97.4 %). Recently, two novel Pseudomonas species, Pseudomonas guineae and Pseudomonas marincola, were described (Bozal et al., 2007; Romanenko et al., 2008) that showed 97.9 and 97.6 % identity, respectively, with respect to strain 1N^T. The phylogenetic tree based on 16S rRNA gene sequences (Fig. 1⇓) illustrates the position of strains IN^T and 3N relative to members of Pseudomonas species; both strains clustered within a single group that was clearly separated from the cluster formed by the closely related type strains of the species P. peli, P. anguilliseptica and P. guineae, and from the clusters formed by the type strains of Pseudomonas borbori, P. marincola, Pseudomonas segetis, P. argentinensis and P. flavescens. These data suggested that the nine isolates analysed in the present study might constitute a novel Pseudomonas species.

The preparation of whole-cell proteins from all of the isolates and SDS-PAGE assays were performed as described previously (Estrada-De Los Santos et al., 2001). The Pseudomonas isolates recovered from the lagoon shared almost-identical protein profiles, but their protein patterns were notably different from those of P. anguilliseptica LMG 21629^T and P. peli LMG 23201^T, as well as from those of representatives of other closely related Pseudomonas species (Fig. 2⇓). It has previously been noted that bacteria with identical or similar protein patterns possess high levels of genome similarity (Vandamme et al., 1996). On this basis, the SDS-PAGE results strongly suggested that all of the isolates represented a novel Pseudomonas species.

For qualitative and quantitative analysis of the cellular fatty acids, strains IN^T and 3N were cultivated on tryptic soy agar (Sigma-Aldrich) for 2 days at 28 °C. The whole-cell fatty acid compositions were determined by using an Agilent 6850 gas chromatograph (Agilent Technologies) and the fatty acid methyl esters were analysed by using the Sherlock 4.5 Microbial Identification System (Microbial ID). The fatty acid compositions of strains 1N^T and 3N, together with type-strain data for their closest phylogenetic relatives, are shown in Table 1⇓. The major fatty acid composition supports the affiliation of strains 1N^T and 3N to the genus Pseudomonas; in addition, there are differences in the relative abundance of fatty acids with respect to the type strains of P. anguilliseptica, P. flavescens, P. guineae and P. peli. Strains 1N^T and 3N exhibit clear quantitative differences for C_12 : 0, C_12 : 0 3-OH and C_16 : 0 with respect to the type strains of their closest relatives, i.e. P. anguilliseptica, P. marincola, P. guineae and P. peli. In addition, these strains exhibit summed feature a (C_16 : 1ω7 and/or C_16 : 1ω6c), which is not present in the closely related species (Table 1⇓). Although most of the fatty acids reported previously for the most closely related Pseudomonas species are present in the profiles of strains 1N^T and 3N, the absence of an unknown fatty acid (equivalent chain length 11.799) and the fatty acid components of summed feature b (C_16 : 1ω7 and/or iso-C_15 : 0 2-OH) is noteworthy. Special caution should be exercised with regard to comparison of the fatty acid profiles of 1N^T and 3N with that of P. guineae M8^T, as the latter was grown for 4 days at 15 °C before the fatty acid analysis, and this is not typical of the standard temperature ranges used for this type of analysis in members of the genus Pseudomonas.

To clarify the taxonomic relationships at species level, DNA–DNA hybridization experiments were performed as described previously (Estrada-De Los Santos et al., 2001); the analysis was based on relative levels of hybridization to ³²P-labelled DNA from strain 1N^T. DNA–DNA relatedness assays were performed with four strains recovered from the lagoon, with the type strain of the most closely related Pseudomonas species (as indicated by 16S rRNA gene sequence data) and with the type strain of Pseudomonas stutzeri, which has a fatty acid profile very similar to those of 1N^T and 3N according to the database of the Microbial Identification System (Microbial ID). The DNA–DNA reassociation values between the strains recovered from the lagoon were in the range 87–99 % (3N, 87 %; 2N, 96 %; 5N, 97 %), indicating close relationships at the species level (Vandamme et al., 1996; Stackebrandt et al., 2002). In contrast, low reassociation values (<15 %) were obtained in hybridizations of strain 1N^T with P. peli LMG 23201^T (14 %), P. anguilliseptica LMG 21629^T (14 %), P. flavescens LMG 18387^T (13 %) and P. stutzeri LMG 11199^T (12 %). These DNA–DNA reassociation data, together with the 16S rRNA gene sequence results, as well as the SDS-PAGE protein patterns and the fatty acid profiles, support the notion that the isolates recovered from the evaporating lagoon represent a novel species of the genus Pseudomonas, for which the name Pseudomonas cuatrocienegasensis sp. nov. is proposed.

To estimate the genome size of P. cuatrocienegasensis sp. nov., strains 1N^T, 2N, 3N, 6N, 8N, 9N, 10N and 11N were analysed by PFGE with I-CeuI endonuclease. Genomic DNA preparation and digestion of whole genomes were carried out as described by Liu et al. (1993) and Matushek et al. (1996), but with some modifications in the DNA-extraction procedure [3 h lysis time instead of overnight; two overnight washes of the lysed product instead of one; and eight washes with TE (10 mM Tris–Cl, pH 7.5; 1 mM EDTA) instead of two]. For P. cuatrocienegasensis sp. nov., the mean genome size was 2995±663 kb, which is similar to those of P. peli LMG 23201^T (3450 kb) and P. flavescens LMG 18387^T (2665 kb), but much smaller than that of P. aeruginosa PAO1 (6262 kb). Four RNA operons were detected in P. cuatrocienegasensis sp. nov. strain 1N^T (Supplementary Fig. S1, available in IJSEM Online).

Bacteriological and biochemical characterization of the novel isolates was carried after growth in LB medium for 48–72 h at 29 °C. The production of fluorescent pigments was tested on King's B medium (King et al., 1954). Other phenotypic features were assayed with the API 20NE and API 50CH systems according to the instructions of the manufacturer (bioMérieux). The API 20NE system was used to determine nitrate reduction, gelatin liquefaction, aesculin hydrolysis, urease activity, indole production, fermentative acid production from glucose, β-galactosidase activity and arginine dihydrolase activity. All colonies were grown on LB agar plates to determine the oxidase reaction as a complementary test in the API 20NE system. The utilization of various carbon sources was tested by using the API 50CH system after incubation for 72 h at 29 °C. The inoculation medium used for the API 50CH test strips was CHB/E [l⁻¹: ammonium sulfate (2 g), yeast extract (0.5 g), tryptone (1 g), disodium phosphate (3.22 g), monopotassium phosphate (0.12 g), trace elements (10 ml), phenol red (0.17 g); pH 7.4–7.8]; the incubation was performed at 20–25 °C. The results of these tests showed that the strains of P. cuatrocienegasensis sp. nov. differ from the most closely related species in terms of the utilization of specific carbon sources, e.g. d-xylose, d-fucose and gluconate; complete data are shown in Table 2⇓.