Schleiferia thermophila gen. nov., sp. nov., a slightly thermophilic bacterium of the phylum ‘Bacteroidetes’ and the proposal of Schleiferiaceae fam. nov.

Most species of the phylum ‘Bacteroidetes’ have optimum growth temperatures that range from about 25 to 45 °C. Several organisms are psychrophilic with very low growth temperatures; these include Psychroserpens burtonensis, and members of the genera Gelidibacter and Maribacter (Bowman et al., 1997; Nedashkovskaya et al., 2004). However, within this phylum, thermophilic or slightly thermophilic species are very rare. Some of these organisms have slightly elevated growth temperatures of around 40–45 °C, such as Pseudozobellia thermophila (Nedashkovskaya et al., 2009) and Lutaonella thermophilia (Arun et al., 2009), or are slightly thermophilic, namely Anaerophaga thermohalophila (Denger et al., 2002), with an optimum growth temperature around 50 °C but unable to grow at 60 °C. The species of two other genera of the phylum ‘Bacteroidetes’ are thermophilic; these are Thermonema lapsum (Hudson et al., 1989) and Thermonema rossianum (Tenreiro et al., 1997), with optimum growth temperatures of about 60 °C and maximum growth temperatures around 65 °C, and the two species of the genus Rhodothermus, namely Rhodothermus marinus and the recently described Rhodothermus profundi (Alfredsson et al., 1988; Nunes et al., 1995; Silva et al., 2000; Marteinsson et al., 2010). The two species of the genus Rhodothermus have optimum growth temperatures of over 65 °C and maximum growth temperatures just below 80 °C. The species of the genus Rhodothermus have the deepest branches of the phylum ‘Bacteroidetes’.

We recently isolated several slightly thermophilic organisms from the Furnas area, such as Elioraea tepidiphila (Albuquerque et al., 2008), Tepidicella xavieri (França et al., 2006), Truepera radiovictrix (Albuquerque et al., 2005), Meiothermus timidus (Pires et al., 2005) and one undescribed slightly thermophilic organism with an optimum growth temperature around 50–55 °C and a maximum growth temperature of 60 °C. Phylogenetic analysis of the 16S rRNA gene sequence showed that this organism represents a distinct lineage within the phylum ‘Bacteroidetes’. Based on phylogenetic, physiological and biochemical parameters, we are of the opinion that strain TU-20^T represents a novel genus and species for which we propose the name Schleiferia thermophila gen. nov., sp. nov. We are also of the opinion that this organism represents a new family for which we propose the name Schleiferiaceae fam. nov.

Strain TU-20^T was isolated from a hot spring in the Furnas area of the Island of São Miguel in the Azores. Water samples were maintained without temperature control for 6 days, and then 10 to 100 ml were filtered through membrane filters (Gelman type GN-6; pore size 0.45 µm; diameter 47 mm). The filters were placed on the surface of solidified Thermus medium (Williams & da Costa, 1992), the plates were wrapped in plastic to prevent evaporation and incubated at 50 °C for up to 4 days. Thermus medium contained the following components per litre of media: yeast extract (Difco), 1.0 g; tryptone (Difco), 1.0 g; agar (Difco), 20.0 g; 100 ml 10× concentrated macronutrients solution; 10 ml 100× concentrated trace elements solution; and 10 ml 0.17 mM FeCl₃ . 6H₂O. The 10× concentrated macronutrients solution contained per litre: nitrilotriacetic acid, 1.0 g; CaSO₄ . 2H₂O, 0.6 g; MgSO₄ . 7H₂O, 1.0 g; NaCl, 0.08 g; KNO₃, 1.03 g; NaNO₃, 6.89 g; and NaHPO₄, 1.11 g. The 100× concentrated trace elements solution contained per litre: MnSO₄ . H₂O, 0.22 g; ZnSO₄ . 7H₂O, 0.05 g; H₃BO₃, 0.05 g; CuSO₄ . 5H₂O, 0.0025 g; Na₂MoO₄ . 2H₂O, 0.0025 g; and CoCl₂ . 6H₂O, 0.0046 g. Cultures were purified by subculturing and the isolates stored at –70 °C in Thermus medium with 15 % (w/v) glycerol. Culture in Degryse medium 162 (Degryse et al., 1978), containing 0.25 % yeast extract and 0.25 % tryptone, was later adopted because this medium resulted in higher growth yields. Degryse medium contained the following components per litre of media: yeast extract (Difco), 2.5 g; tryptone (Difco), 2.5 g; agar (Difco), 20.0 g; 100 ml 10× concentrated macronutrients solution; 5 ml 100× concentrated trace elements solution; 0.5 ml 0.01 M C₆H₅O₇Fe . 5H₂O, 10 ml 0.2 M KH₂PO₄ and 15 ml 0.2 M Na₂HPO₄ . 12H₂O. The 10× concentrated macronutrients solution contained per litre: nitrilotriacetic acid, 1.0 g; CaSO₄ . 2H₂O, 0.4 g; and MgSO₄ . 7H₂O, 2.0 g. The 100× concentrated trace elements solution contained per litre: MnSO₄ . H₂O, 0.22 g; ZnSO₄ . 7H₂O, 0.05 g; H₃BO₃, 0.05 g; CuSO₄ . 5H₂O, 0.0025 g; Na₂MoO₄ . 2H₂O, 0.0025 g; CoCl₂ . 6H₂O, 0.0046 g.

The type strain of Fluviicola taffensis (DSM 16823^T) was used for comparative purposes.

Unless otherwise stated, all biochemical and tolerance tests were performed as described by Smibert & Krieg (1981) and Albuquerque et al. (2010) in Degryse medium 162 or Degryse agar 162 (Degryse et al., 1978) at 50 °C for up to 6 days. Cell morphology and motility were examined by phase-contrast microscopy during the exponential growth phase. Gliding motility was examined as described by Bernardet et al. (2002). The growth temperature range of the strains was examined in 5 °C increments between 25 and 65 °C by measuring the turbidity (610 nm) of cultures incubated in 300 ml metal-capped Erlenmeyer flasks containing 100 ml of medium in a rotary water-bath shaker at 150 r.p.m. The pH range for growth was examined in liquid medium by using 50 mM MES, HEPES, TAPS or CAPSO over a pH range from 6.0 to 10.0 in a rotary water-bath shaker. The salt tolerance of the organisms was determined in liquid medium, with NaCl ranging between 0.0 and 4.0 %. Catalase, oxidase and DNase activities were examined as described by Smibert & Krieg (1981) and Albuquerque et al. (2010). Additional enzymic activities were obtained using the API ZYM system (bioMérieux) at 50 °C. Anaerobic growth was assessed in cultures in the same medium, containing KNO₃ (1.0 g l⁻¹), incubated in anaerobic chambers (GENbox anaer; bioMérieux). Single-carbon-source assimilation tests were performed in a medium composed of Degryse basal salts (Degryse et al., 1978) to which filter-sterilized ammonium sulfate (0.5 g l⁻¹), yeast extract (0.2 g l⁻¹), vitamin B12 (40 µg l⁻¹) and the carbon source (2.0 g l⁻¹) were added. Growth of the strains on single carbon sources was examined by measuring the turbidity of cultures incubated at 50 °C in 20 ml screw-capped tubes containing 10 ml medium for up to 6 days. Acid production from carbohydrates was determined by using the API 50 CH system (bioMérieux) according to the manufacturer’s instructions, using API 50 CHB/E medium. Results were recorded after 48 h and 6 days of incubation at 50 °C.

Metabolic products were determined by growing the cultures in Degryse liquid medium overnight. The cultures were washed by centrifugation and incubated at 50 °C in 50 mM phosphate buffer, pH 7.5, containing glucose (11.1 mmol) or fructose (11.1 mmol) at a turbidity of 7–8 (OD 610 nm) for 24 h with periodic samplings and centrifuged. The samples were analysed using standard proton-nuclear magnetic resonance (¹H-NMR) on a Bruker AMX300 spectrometer (Antunes et al., 2002).

Chemolithoorganotrophic growth on sodium thiosulfate was assessed under aerobic conditions at the optimum growth temperature in DSMZ medium 27 () lacking ethanol and sulphate, and containing yeast extract (1.0 g l⁻¹), succinate (1.0 g l⁻¹) and acetate (0.5 g l⁻¹). Filter-sterilized sodium thiosulfate was added to the medium at concentrations that varied between 0.5 and 5.0 g l⁻¹. At appropriate intervals, the turbidity of the cultures was determined and the cells were harvested and centrifuged. The levels of thiosulfate and sulfate in the supernatants were determined using the methods described by Westley (1987) and Sörbo (1987), respectively.

The cultures for polar lipid analysis were grown in 1 litre Erlenmeyer flasks containing 250 ml medium at 50 °C in a rotary water-bath shaker until the late exponential phase of growth. Harvesting of the cultures and extraction of lipids was as described by da Costa et al. (2006). Individual polar lipids were separated by two-dimensional thin-layer chromatography and visualized as described previously (da Costa et al., 2006). Lipoquinones were extracted from freeze-dried cells and purified by thin-layer chromatography as described previously (da Costa et al., 2006). The lipoquinones were separated by high performance liquid chromatography with a Gilson HPLC by using a reverse-phase Hichrom 5 C18 column and methanol/heptane (10 : 2, v/v) as the mobile phase and were detected at 269 nm. Cultures for fatty acid analysis were grown on solidified Degryse medium 162 in sealed plastic bags submerged in a water bath at 50 °C for 48 h. Fatty acid methyl esters (FAMEs) were obtained from fresh wet biomass and separated, identified and quantified with the standard MIS Library Generation Software (Microbial ID) as described previously (da Costa et al., 2006).

DNA for the determination of G+C content was isolated as described by Nielsen et al. (1995). The G+C content of DNA was determined by high-performance liquid chromatography as described by Mesbah et al. (1989).

A 16S rRNA gene sequence comprising 1499 nt was determined for strain TU-20^T as previously described (Rainey et al., 1996). The 16S rRNA gene was amplified in a number of fragments using the primer combinations 27f-519r, 27f-1385r, 357f-1525r and 536f-1525r (Rainey et al., 1996). The taxonomic affiliation of strain TU-20^T, including the designation of its closest relatives based on the 16S rRNA gene sequence, was determined using the EzTaxon facility (Chun et al., 2007).

Strain TU-20^T formed very short, Gram-stain-negative, rod-shaped cells (0.5 to 0.7 µm in width and 0.5 to 1.5 µm in length) without flagella and without visible gliding motility. Colonies were orange-pigmented on Degryse medium 162. The organism had an optimum growth temperature of about 50 °C and did not grow at 25 or 62.5 °C. This organism was oxidase-positive and catalase-negative and hydrolysed hippurate, starch, gelatin and casein. Aesculin, arbutin and xylan were not hydrolysed. Other enzyme activities are listed in the species description. Small amounts of yeast extract (0.2 g l⁻¹) and vitamin B12 (40 µg l⁻¹) were added to the minimal medium to assess assimilation of single carbon sources. The organism produced acid from several carbohydrates, leading us to determine the acids produced. However, only very low levels of succinate were detected. Nitrate was reduced to nitrogen, but anaerobic growth with nitrate as electron acceptor was not observed (Table 1). Thiosulfate was completely oxidized to sulphate in media containing 0.5, 1.0 and 2.0 g thiosulfate l⁻¹ but there was no measurable increase in biomass.

The polar lipid pattern produced by thin-layer chromatography revealed the presence of one aminophospholipid, two aminolipids and three unidentified lipids. We also examined the polar lipids of F. taffensis, where we encountered one phospholipid, one aminophospholipid, one aminolipid and two unidentified lipids (Supplementary Fig S1, available in IJSEM Online), but only two lipids, an aminophospholipid and an aminolipid seem to be shared by both organisms. Moreover, strain TU-20^T possesses a major aminophospholipid which is not found in F. taffensis. The major respiratory lipoquinone was menaquinone 6, as in all members of the families Cryomorphaceae and Flavobacteriaceae examined. The fatty acid composition of strain TU-20^T was dominated by 16 : 1ω6c, 17 : 0 iso 3-OH and 15 : 0 iso (Table 2). However, many other organisms within the phylum ‘Bacteroidetes’ have large amounts of branched-chain fatty acids and 2- and 3-OH fatty acids.

Phylogenetic analyses using a dataset that included the 16S rRNA gene sequence of the type strain of all genera of the phylum ‘Bacteroidetes’ showed strain TU-20^T to represent a distinct family-level line of descent. The TU-20^T lineage falls outside of all of the described and proposed (Bergey’s Taxonomic Outline; ) families of the phylum ‘Bacteroidetes’ (Supplementary Fig S2). The highest sequence similarity was approximately 86 % with the species of the genera Owenweeksia, Cryomorpha, Lishizhenia, Brumimicrobium, Wandonia, Fluviicola and Crocinitomix (Bowman et al., 2003; Lau et al., 2005, 2006; Lee et al., 2010; O’Sullivan et al., 2005), which have been classified as members of the family Cryomorphaceae (Bowman et al., 2003). The next closest relatives, based on pairwise similarity values (82–83 %) and phylogenetic position, are the species of the family Flavobacteriaceae (Fig. 1 and Supplementary Fig. S2). Fig 1 and Supplementary Fig. S2 clearly demonstrate the lack of relationship of strain TU-20^T with any of the lineages within the phylum ‘Bacteroidetes’ and support its description as a novel taxon at the family level. On the basis of the results, we propose that strain TU-20^T represents a novel species of a novel genus for which we recommend the name Schleiferia thermophila gen. nov., sp. nov. We also propose that this organism represents a new family of the phylum ‘Bacteroidetes’ for which we recommend the name Schleiferiaceae fam. nov.