Leucobacter iarius sp. nov., in the family Microbacteriaceae

Members of those genera of the family Microbacteriaceae that contain 2,4-diaminobutyric acid in their B-type cross-linked peptidoglycan (Schleifer & Kandler, 1972) differ from each other mainly in menaquinone type, cell wall sugars, G+C content of the DNA and physiochemical features (Morais et al., 2004). The genus Leucobacter was proposed to accommodate Gram-positive rods with a distinct peptidoglycan with an unusual amino acid, γ-aminobutyric acid, a high percentage of menaquinone 11 (MK-11) and moderate 16S rRNA gene sequence similarities with other members of the family. Diagnostic sugars were absent from the cell wall of Leucobacter komagatae, the only species available at the time of the study (Takeuchi et al., 1996). In 2004, three new members, Leucobacter chromiireducens, Leucobacter aridicollis (Morais et al., 2004) and Leucobacter albus (Lin et al., 2004), and in 2006 two new members, Leucobacter alluvii and Leucobacter luti (Morais et al., 2006), were added to the genus on the basis of the aforementioned properties.

Members of the genus Leucobacter have been isolated from diverse environments, such as activated sludge from a treatment plant containing chromium-contaminated wastewater (L. chromiireducens, L. aridicollis and L. luti) (Morais et al., 2004, 2006) and from soil in Thailand (L. albus) (Lin et al., 2004), while L. komagatae was isolated as a contaminant from an ampoule containing ‘Pseudomonas riboflavina’ (=Devosia riboflavina) (Takeuchi et al., 1996) and L. alluvii from sediment of the Alviela river, Portugal. Here we report the isolation and characterization of a novel Leucobacter species from the bacterial flora associated with the entomopathogenic nematode Steinernema thermophilum (Ganguly & Singh, 2000), recovered from soil in India.

Strain 40^T was one of the isolates found in the course of identifying bacteria from infective juveniles of the entomopathogenic nematode Steinernema thermophilum by using a procedure established by Akhurst (1980) with the modifications described by Somvanshi et al. (2006). The type strains of recognized Leucobacter species were obtained from the DSMZ and CRBIP, Paris. Cultural characteristics of strain 40^T are indicated in the species description.

Genomic DNA was extracted using a DNeasy Tissue kit (Qiagen) following the manufacturer's instructions. The 16S rRNA gene was amplified as described by Rainey et al. (1996). The PCR products were purified using a QIAquick PCR purification kit (Qiagen) and sequenced directly by using a CEQ Dye Terminator Cycle sequencing kit. The products were separated on a CEQ 8000 Genetic Analysis system. The 16S rRNA gene sequences were aligned with corresponding sequences from the database of DSMZ using the ae2 editor (Maidak et al., 1997). Evolutionary distances were calculated using the method of Jukes & Cantor (1969). A phylogenetic tree based on distance analysis, reconstructed using the neighbour-joining algorithm (Saitou & Nei, 1987), and bootstrap values were determined by analysing 1000 resamplings (Felsenstein, 1993).

The 16S rRNA gene sequence of strain 40^T (1508 nt) showed 97.3–98.5 % similarity to those of type strains of recognized species of the genus Leucobacter, the closest relative being L. aridicollis (Fig. 1⇓). However, these values were lower than those determined for certain strain pairs, such as L. albus and L. aridicollis (98.9 %), L. albus and L. komagatae (98.8 %) and L. komagatae and L. aridicollis (98.7 %). As previous studies have confirmed DNA–DNA reassociation values of less than 70 % for the latter two strain pairs with high 16S rRNA gene similarities [40 % (Lin et al., 2004) and 38 % (Morais et al., 2004, 2006), respectively], we refrained from testing this parameter and consider strain 40^T to represent a distinct genospecies (see also Stackebrandt & Ebers, 2006).

Physiological and biochemical tests on strain 40^T and the type strains of other Leucobacter species were performed at 28 °C using API 50 CH strips (bioMérieux), according to the manufacturer's instructions, with bioMérieux medium E [2 g ammonium sulphate, 0.5 g yeast extract, 1 g tryptone (bovine/porcine origin), 3.22 g disodium phosphate, 0.12 g monosodium phosphate, 10 ml trace elements, 0.17 g phenol red, 1000 ml demineralized water]. All tests were inoculated with 24 h-old cultures grown on trypticase soy agar. The reactions were scored up to 5 days, as indicated by Morais et al. (2004). Strain 40^T was also characterized by using an API 20NE strip. Biolog GP plates (Oxoid) were incubated for 24 h before being read. Wells showing a photometric value above 20 or 30 % of the highest value of an individual strain were scored as weak or positive, respectively. Conventional biochemical tests were performed according to standard methods (Smibert & Krieg, 1994). Cultural characteristics such as colony size, shape and colour were determined after 24 h incubation at 28 °C on CASO agar plates (Merck). The temperature range for growth was determined by incubating inoculated slant cultures (trypticase soy agar, pH 7.0) at 4, 25, 30, 37 and 45 °C. Growth was also assessed in trypticase soy broth (Difco) at pH 5–9. pH values of the test medium of less than 7.5 were obtained by adding HCl and pH values greater than 7.5 by addition of sterilized sodium sesquicarbonate buffer (1 M). The ability to grow in the presence of NaCl was determined using trypticase soy broth containing NaCl (5, 7 or 10 %). Growth characteristics are given in the species description.

On the basis of the results of the physiological tests, it was concluded that strains of the genus Leucobacter react negatively towards most of the substrates provided on the API 50 CH panel. There was no common substrate on the API 50 CH test panels that was acidified by all members of the genus (Table 1⇓). According to Morais et al. (2004), many tests were scored ‘weak’ for L. chromiireducens and L. aridicollis in the original description whereas, according to our results, they were negative for all the respective substrates even when read after 5 days. The Biolog GP test plates were more suitable for distinguishing Leucobacter strains (Table 2⇓). These data reveal a clear preference of the strains for utilization of amino acids over carbohydrates. Strain 40^T differed from the other strains in utilization of N-acetylglucosamine and the weak acidification of 5-ketogluconate and salicin within the API 50 CH substrate panels. In addition, a change of the indicator dye was observed for fructose (Table 1⇓). The reactions of some strains towards certain substrates differed between the Biolog and API substrate panels (Tables 1 and 2⇓⇓).

Cultures for use for determination of the fatty acid composition were grown on trypticase soy broth agar for 24 h at 28 °C. Processing and analysis of fatty acids as methyl esters were done using the protocol of the MIDI Sherlock Microbial Identification System (MIDI, 1999).

Cellular fatty acid analysis was carried out for all the type strains except L. komagatae IAM 1093^T, L. luti CIP 108818^T and L. alluvii CIP 108819^T, for which data were taken from Takeuchi et al. (1996) and Morais et al. (2006) (Table 3⇓). The major fatty acids of strain 40^T were anteiso-C_15 : 0 (41.9 %), anteiso-C_17 : 0 (33.6 %) and iso-C_16 : 0 (15.5 %). These results are in accord with the fatty acid compositions of other members of the genus, confirming the affiliation of strain 40^T with the genus.

Purified cell wall preparations were obtained by using the method of Schleifer & Kandler (1972). Amino acids and peptides in cell wall hydrolysates were analysed by two-dimensional ascending thin layer chromatography on cellulose plates using previously described systems (Schleifer & Kandler, 1972). The N-terminal amino acid of the inter-peptide bridge was determined by dinitrophenylation as described by Schleifer (1985). Strain 40^T possessed glutamic acid (1.0), l-diaminobutyric acid (0.5), alanine (1.5), glycine (0.9) and threonine (0.7), a composition that confirmed the presence of a B-type cross-linked peptidoglycan. As revealed by denitrophenylation, alanine was the N-terminal amino acid. The dipeptide Gly–Glu was detected after partial hydrolysis. 4-Aminobutyric acid, as detected in some other Leucobacter strains (L. komagatae, L. albus), was not observed. The N-terminal alanine, the presence of threonine and the Gly–Glu dipeptide corresponds to the information reported for L. chromiireducens, L. luti and L. alluvii (Morais et al., 2006). Lipoquinone and polar lipid analyses were done by using previously described methods (Groth et al., 1996). The dominant menaquinone was MK-11, consistent with the results obtained with the type strains of other Leucobacter species. MK-10 and MK-9 were present in minor amounts, while MK-12 occurred in trace amounts only. The polar lipids were phosphatidylglycerol and diphosphoglycerol. Several unidentified components occurred, i.e. a phospholipid, a glycolipid and an aminophospholipid.

In conclusion, based upon the phylogenetic and chemotaxonomic evidence, strain 40^T represents a novel member of the genus Leucobacter, family Microbacteriaceae, for which the name Leucobacter iarius sp. nov. is proposed.