Antonie van Leeuwenhoek

, Volume 96, Issue 4, pp 515–526

Description of Tessaracoccus profundi sp.nov., a deep-subsurface actinobacterium isolated from a Chesapeake impact crater drill core (940 m depth)

Authors

    • Department of Biological Sciences, Section for MicrobiologyAarhus University
  • C. S. Cockell
    • PSSRI, Open University
  • M. A. Voytek
    • US Geological Survey
  • A. L. Gronstal
    • PSSRI, Open University
  • K. U. Kjeldsen
    • Department of Biological Sciences, Center for GeomicrobiologyUniversity of Aarhus
Original Paper

DOI: 10.1007/s10482-009-9367-y

Cite this article as:
Finster, K.W., Cockell, C.S., Voytek, M.A. et al. Antonie van Leeuwenhoek (2009) 96: 515. doi:10.1007/s10482-009-9367-y

Abstract

A novel actinobacterium, designated CB31T, was isolated from a 940 m depth sample of a drilling core obtained from the Chesapeake meteor impact crater. The strain was isolated aerobically on R2A medium agar plates supplemented with NaCl (20 g l−1) and MgCl2·6H2O (3 g l−1). The colonies were circular, convex, smooth and orange. Cells were slightly curved, rod-shaped in young cultures and often appeared in pairs. In older cultures cells were coccoid. Cells stained Gram-positive, were non-motile and did not form endospores. The diagnostic diamino acid of the peptidoglycan was ll-diaminopimelic acid. The polar lipids included phosphatidylglycerol, diphosphatidglycerol, four different glycolipids, two further phospholipids and one unidentified lipid. The dominant menaquinone was MK-9(H4) (70%). The major cellular fatty acid was anteiso C15:0 (83%). The DNA G + C content was 68 mol%. The strain grew anaerobically by reducing nitrate to nitrite or by fermenting glucose. It was catalase positive and oxidase negative. It grew between 10 and 45°C, with an optimum between 35 and 40°C. The pH range for growth was 5.7–9.3, with an optimum at pH 7.5. The closest phylogenetic neighbors based on 16S rRNA gene sequence identity were members of the genus Tessaracoccus (95–96% identity). On the basis of phenotypic and phylogenetic distinctiveness, strain CB31T is considered to represent a novel species of the genus Tessaracoccus, for which we propose the name Tessaracoccus profundi sp. nov.. It is the first member of this genus that has been isolated from a deep subsurface environment. The type strain is CB31T (=NCIMB 14440T = DSM 21240T).

Keywords

Tessaracoccus profundiActinobacteriaTaxonomyDeep subsurface

Introduction

The exploration of the subsurface of the Earth is a flourishing field of research at the cutting edge of microbiology and geology and which, since the mid 1980’s, has provided increasing evidence for the presence of a large variety of microorganisms and their activities (Teske 2005). Investigating these parts of the Earth is both a costly and technically demanding endeavor as it has to be assured that samples remain pristine and uncontaminated throughout collection and processing (White et al. 1998; D’Hondt et al. 2004; Gronstal et al. 2009). Over the last three decades, however, microorganisms have been detected in habitats as diverse as oceanic sediments (Parkes et al. 1994), aquifers (Chapelle et al. 2002), granites (Pedersen 1997), ocean crust (Thorsvik et al. 1998), coal deposits (Fry et al. 2009) and gold mines (Onstott et al. 2003), habitats that half a century ago were considered devoid of living organisms (Morita and ZoBell 1955). Generally, these habitats harbor complex microbial communities (Crocker et al. 2000; Batzke et al. 2007; Sahl et al. 2008). However, recently Onstott and coworkers have shown that a single strain of the hydrogenotrophic sulfate reducer, Candidatus Desulforudis audaxviator, dominates (almost 100% of the retrieved clones) fracture water collected from a depth of 2.8 km in a South-African gold mine (Chivian et al. 2008).

Total cell counts made on the samples obtained from subsurface habitats in general and from the oceanic deep subsurface in particular allowed Whitman et al. (1998) to conclude that the deep subsurface is the largest purely microbial habitat on Earth. This is perhaps surprising in view of the poor nutritional conditions that generally prevail in these environments and the hitherto unknown way of life of microorganisms ways under these conditions remains enigmatic (Jørgensen and Boetius 2007).

High G + C Gram-positive Actinobacteria are frequently detected in deep subsurface samples by both cultivation-dependent and independent methods, in particular the isolation of members of the genera Arthrobacter, Micrococcus and Terrabacter were reported previously (Blakwill et al. 1997; Batzke et al. 2007; Brown and Blakwill 2009; Fry et al. 2009; Gérard et al. 2009). These bacteria are generally facultative aerobic heterotrophs that are able to degrade a diverse range of substrates. This generalist nutritional potential may be an advantage in deep subsurface habitats that likely contain a large variety of substrates at low concentrations. In addition, Actinobacteria have a very efficient DNA repair mechanisms (Johnson et al. 2007), which may allow them to survive long periods of nutrient deprivation.

To increase our knowledge of microorganisms inhabiting the terrestrial deep subsurface a deep drilling program was initiated in the Chesapeake Impact crater, a structure that has already been extensively explored during numerous geological investigations. The impact crater, which is the largest known impact structure in the United States (Horton et al. 2005), was formed by a massive impact of an unidentified object about 35 million years ago. It has an average diameter of 85 km (Gohn et al. 2006) and the original crater is now buried below a 300–500 m thick layer of sediments (Poag 1997). During the drilling samples were retrieved aseptically under rigorous testing for contamination by allochthonous microorganisms (Gohn et al. 2008) from different depth intervals of the crater, including post impact sediments, sediment-clast brecchia and schists. After sampling, in coordination with other studies, a comprehensive search for microorganisms in the material obtained was initiated using multiple aerobic and anaerobic incubation techniques.

Here we report on the isolation and characterization of a novel facultative Actinobacterium that was recovered from a drilling core sample collected at 940 m depth of the Chesapeake meteor impact crater. This depth is within the sediment-clast brecchia that filled up the crater after the impact. The isolated strain was affiliated to the genus Tessaracoccus as a novel species with the designation Tessaracoccus profundi. This is to our knowledge the first time that a strain belonging to the genus Tessaracoccus has been retrieved form the deep subsurface.

Materials and methods

Sampling and isolation

A subsample for microbiological investigation was carefully withdrawn from Chesapeake drill core sample No. 31, obtained from 940 m depth in 2005 at Eyreville Farm (N 37°19.3′, W 75°58.54′), in Northampton County, VA, USA and aseptically transferred to sterile plastic bags (Gohn et al. 2006; Gronstal et al. 2009) in which it was stored frozen in the absence of oxygen. In the laboratory, a subsample (ca. 5 g) was aseptically removed in a clean N2-filled glove box, ground with a sterile pestle in a mortar and transferred to a sterile test tube that was sealed with a rubber stopper. One gram of the ground material was aerobically suspended in 10 ml PBS, vortexed for 1 min and 1 ml of the suspension was transferred to a glass bottle completely filled with 50 ml of anoxic R2A medium broth (DSM medium 830) supplemented with NaCl (20 g l−1) and MgCl2·6H2O (3 g l−1) and sealed with an aluminum screw-cap with a gastight rubber septum. Growth was followed by visual inspection of the culture bottle and microscopical examination of the culture on a daily basis.

Cultivation, physiology and chemotaxonomy

Cell biomass for chemotaxonomic analyses was obtained from cultures grown on R2A agar plates (Difco, Becton, Dickinson and Company, USA). Cells were scraped from the plates using sterile spatulas, suspended in PBS buffer in sterile centrifuge tubes and stored either freeze-dried or in 70% (v/v) isopropanol until analyzed further. All chemotaxonomic analyses (G + C mol% of the genomic DNA, quinones, fatty acids, polar lipids, and cell wall type) were carried out by the DSMZ identification service, Braunschweig, Germany, according to certified DMSZ standard protocols.

Peptidoglycan was isolated, purified and analyzed according to the methods of Schleifer and Kandler (1972), Schleifer (1985), MacKenzie (1987) and Groth et al. (1996). The cellular fatty acid composition was determined according to the Microbial Identification System (MIDI) protocols.

The growth response to pH and NaCl concentrations was tested in liquid R2A medium while the temperature response was determined on solid R2A plates supplemented with salts as described above. The pH of the medium was adjusted by the addition of either 1 M HCl or 1 M NaOH covering the range pH 4 to pH 10; the pH was monitored regularly during all subsequent incubations. Salt tolerance was determined by amending R2A agar with NaCl to give final concentrations of 0–10% w/v NaCl in 1% increments. The temperature range was determined by following growth of the cultures on agar plates incubated at 2, 5, 10, 15, 20, 30, 40, 45 and 50°C, respectively.

The physiological properties of the strain, such as substrate utilization, expressed enzymes and antibiotic-resistance, were examined with prefabricated test systems: API 20E, API 20NE, APIzym, BIOLOGGP and antibiotic containing discs. Culture preparation, inoculation and incubation of the tests were carried out according to the manufacturers instructions (Biomerieux, Craponne, France; BIOLOG, Hayward, CA).

Anaerobic growth potential was determined on R2A agar using the Oxoid Anaerobic System (Oxoid Ltd, Basingstoke, U.K.). Catalase activity was determined using 3% (v/v) H2O2 and oxidase activity was determined using prefabricated test strips containing 1% w/w tetramethyl p-phenylenediamine following the manufacturers instructions (Merck, Germany). Gram staining was performed according to Barrow and Feltham (1993).

16S rRNA gene sequence-based phylogenetic analyses and DNA G + C content

The 16S rRNA gene of strain CB31T was PCR amplified and sequenced as described by Hansen et al. (2007). The resultant 1,404 nt long sequence was added to the alignment of the Silva SSURef version 95 database (Pruesse et al. 2007) using the ARB program package (Ludwig et al. 2004). The phylogenetic placement of the strain among members of the family Propionibacteriaceae was inferred by bayesian analysis (Ronquist and Huelsenbeck 2003) as described previously (Finster et al. 2009). A dataset composed of 54 taxa (Supplementary Fig. S1) was analyzed with sampling of 1,326 sequence positions between E. coli position 52 and 1,406. Neighbor-joining (with Kimura 2-parameter distance correction)-based bootstrap analysis (100 replications) was performed using Phylip (Phylogeny Inference Package) version 3.67 (distributed by J. Felsenstein, Department of Genome Sciences, University of Washington, Seattle, WA).

The G + C mol% of the DNA of strain CB31T was determined by reverse-phase HPLC according to Mesbah et al. (1989).

Results and discussion

Enrichment and isolation

Within 2 weeks of incubation at room temperature a dense culture developed as evident from the increased turbidity of the growth medium. Microscopical examination revealed the presence of predominantly short rods. For isolation, 100 μl of the culture was spread on solid R2A agar supplemented with salts and incubated aerobically at room temperature. After 1 week small orange-pigmented colonies were observed. One distinct faint orange colony was selected and transferred to a new R2A agar plate for further purification. Purity was confirmed by microscopy and aerobic growth in TSB medium (DSM medium 545). Only the short rod-morphotype was observed, and one representative strain, designated CB31T (T = type strain), was chosen for further characterization. According to phase contrast and light microscopy at 1000× magnification (Olympus BH2, Ballerup, Denmark), cells from exponentially growing cultures of strain CB31T were about 1.5 μm long and about 0.5 μm wide (Fig. 1). The cells were immotile, did not produce endospores and stained Gram-positive. Cultures were maintained on R2A agar plates at 4°C on which they can be stored for at least 6 months.
https://static-content.springer.com/image/art%3A10.1007%2Fs10482-009-9367-y/MediaObjects/10482_2009_9367_Fig1_HTML.jpg
Fig. 1

Phase contrast micrograph (1,000× magnification) of a plate grown culture of strain CB31T. The scale bar represents 2.5 μm

DNA-based taxonomy

Initial phylogenetic analyses including comparisons of the 16S rRNA gene sequence against the NCBI (http://www.ncbi.nlm.nih.gov/) nucleotide sequence database using the BLAST tool (Altschul et al. 1997) showed a clear affiliation with members of the actinobacterial family Propionibacteriaceae (Stackebrandt et al. 1997). As is evident from Fig. 2 (and Supplementary Fig. S1) strain CB31T formed a distinct clade with the three validly described members of the genus Tessaracoccus, family Propionibacteriaceae (Kämpfer et al. 2009; Lee and Lee 2008; Maszensan et al. 1999). According to direct pairwise comparisons using the EzTaxon online tool (Chun et al. 2007), the 16S rRNA gene sequence of strain CB31T shared 96.0, 95.9 and 94.9 percent identity with those of Tessaracoccus flavescens DSM 18582T (AM393882), Tessaracoccus lubricantis DSM 19926T (FM178840) and Tessaracoccus bendigoensis ACM 5119T (AF038504), respectively. Since these values are below 97%, which is the recommended cut-off used to delineate distinct bacterial species (Stackebrandt and Goebel 1994) DNA–DNA hybridization experiments were not carried out.
https://static-content.springer.com/image/art%3A10.1007%2Fs10482-009-9367-y/MediaObjects/10482_2009_9367_Fig2_HTML.gif
Fig. 2

Phylogenetic position of strain CB31T (shown in bold face) among selected members of the family Propionibacteriaceae as inferred from Bayesian 16S rRNA gene sequence analysis. Several taxa were removed from the presented tree (a complete version is shown in supplementary Fig. S1). Numbers denote neighbor-joining-based bootstrap percentage values >50%. Bar 10% estimated sequence divergence

The DNA G + C mol% of CB31T was 68.4%, which was identical to the DNA G + C mol% of T. flavescens, while the DNA G + C mol% of T. bendigonensis was considerably higher with a value of 74% (Lee and Lee 2008; Maszensan et al. 1999) but well within the ±10 mol% range empirically observed to generally constitute the common range within prokaryotic genera (Rosselló-Mora and Amann 2001). The DNA G + C mol% of T. lubricantis has not been reported.

The GenBank/EMBL/DDBJ accession number for the 16S rRNA gene sequence of CB31T is FJ228690.

Chemotaxonomy

The results are summarized in Table 1. The peptidoglycan was of the A3γ′ type where l-alanine, which usually occurs in position 1 of peptidoglycan group A peptide subunits, is substituted by glycine (Schleifer and Seidl 1985). As characteristic of all A3γ types, the ll isomer of diaminopimelic acid occupied position 3 of the peptide subunit and glycine was identified in the interpeptide bridge bound by this isomer (Schleifer and Seidl 1985). ll-diaminopimelic acid was also present in the three other species of the genus Tessaracoccus (Maszensan et al. 1999; Lee and Lee 2008; Kämpfer et al. 2009). The murein type of T. bendigonensis (ACM 5119T) was the same as found with strain CB31T (A3γ′), while the murein type of T. flavescens DSM 18582T and T. lubricantis DSM 19926T have not been reported. The composition of the polar lipids of strain CB31T was diverse and included phosphatidylglycerol, diphosphatidglycerol, four different glycolipids (GL1-4), two further phospholipids and one unidentified polar lipid (L1). When stained with the reagent for sugars GL2 is visible while L1 and diphosphatidglycerol are not (Fig. 3). According to Dr. Brian Tindall (DSMZ), this pattern is unique and may serve as a chemical marker for the new taxon (pers. communication). T. bendigonensis also contained, in addition to the polar lipids identified in CB31T, phosphatidylinositol (Maszensan et al. 1999). In contrast, the composition of polar lipids of T. flavescens was originally reported to contain only phosphatidylglycerol and diphosphatidylglycerol (Lee and Lee 2008). The latter results were however not consistent with those results reported by Kämpfer et al. (2009) for T. flavescens, since they obtained a more complex pattern including unknown glycolipids GL2 and GL3 and the same unknown polar lipid L1 produced by strain CB31T. The polar lipid composition of T. lubricantis is complex and consisted of four predominant compounds: diphosphatidylglycerol, phosphatidylglyceral, an unknown glycolipid and an unknown polar lipid. In addition minor amounts of three unknown glycolipids, two unknown aminolipids, two unknown polar lipids and trace amounts of phosphatidylinositol were found (Kämpfer et al. 2009). The cellular fatty acid profile of strain CB31T was dominated by anteiso-C15:0 that accounted for 83%. This is a major fatty acid in most members of the family Propionibacteriaceae (Stackebrandt and Schaal 2006). Similarly, anteiso-C15:0 was reported to also dominate the fatty acid profile of T. lubricantis (Kämpfer et al. 2009) while being less prevalent in T. flavescens and T. bendigonensis, which also contain relative large amounts of anteiso-C18:0 (50 and 28%, respectively) (Lee and Lee 2008; Maszensan et al. 1999). Anteiso-C18:0 was not detected in CB31T or in T. lubricantis (Table 2).
Table 1

Differential characteristics between strain CB31T and its phylogenetically closest neighbors

 

CB 31T

Tessaracoccus flavescens

Tessaracoccus bendigonensis

Tessaracoccus lubricantis

Habitat

Chesapeake impact crater drill core (940 m depth)

Marine sediment

Sewage treatment plant Australia

Water (metal working fluid)

Colony color

Faint orange

Brilliant yellow

Cream to white

Yellow

O2 requirement

Facultative aerobic

Facultative aerobe

Facultative aerobic

Aerobic

Morphology

Short rods, coccoid in old cultures

Short rod

Coccoid, arranged in tetrads

Short rods

Quinones

MK-9(H4) (70%), MK-9(H2), MK9, MK-9(H6), MK-9(H6) in traces

MK-9(H4) (84%), MK-(H0), MK-7(H2)

MK-9(H4), MK-7(H4)

MK-9(H4), MK-7(H4)

Major fatty acids

Anteiso 15:0 (83%)

Anteiso (15:0) (50%), C18:0 (17.5%)

Anteiso 15:0 (28%), C18:0 (24.4%)

Anteiso 15:0 (83%), iso 15:0 (3.4%),

Murein type

A3-γ′

NR

A3-γ′

NR

Polar lipids

DPG, PG, PL1-2, GL1-4, L1

DPG, PG GL2, GL3, L1

PI, PG, DPG, PL

PG, DPG, GL1-4, PE, PL1-2, L1-3

DNA G + C content (mol%)

68.4

68.4

74

ND

Growth temperature range (°C)

10–45

20–30

20–37

15–36

Growth pH range

5.7–9.3

6.1–12.1

5.5–9.3

6.5–9.5

Aesculin degradation

+

(+)

+

NR

Nitrate reduction

+

+

+

NR

Carbohydrates utilized:

    d-Fructose

    d-Galactose

    Maltose

    Lactose

    Rhamnose

    Sucrose

    d-Xylose

    d-Mannose

    d-Mannitol

+

+

+

+

+

+

+

+

+

+

+

NR

+

+

+

+

+

+

+

+

+

+

Enzymes (API ZYM, NE, E)

    Esterase (C4)

+

+

+

NR

    Leucine arylamidase

+

+

+

NR

    α-Glucosidase

+

+

+

NR

    β-Galactosidase

+

+

+

NR

    Esterase lipase (C8)

+

+

+

NR

    β-Glucosidase

(+)

+

NR

    Alkaline phosphatase

+

NR

    Lipase (C14)

+

+

NR

    Valine arylamidase

+

+

NR

    Cystine arylamidase

+

+

NR

    Trypsin

NR

    α-Chymotrypsin

NR

    Acid phosphatase

+

+

NR

    Naphthol-AS-BI-phosphohydrolase

+

+

NR

    α-Galactosidase

+

+

NR

    β-Glucuronidase

NR

    N-acetyl-β-glucosaminidase

+

+

NR

    α-mannosidase

+

+

NR

    α-fucosidase

NR

    Tryptophan deaminase

+

+

NR

NR

Data for T. bendigoensis DSM 12906T were taken from Maszensan et al. (1999) and Lee and Lee (2008). Data for T. flavescens DSM 18582T were taken from Lee and Lee (2008). Data for T. lubricantis DSM 19926T were taken from Kämpfer et al. (2009). All strains were Gram-positive, oxidase negative, catalase positive, non-sporeforming, non-motile chemo-organoheterotrophs. All strains utilize arabinose and have ll-DAP as diamino acid of peptidoglycan

+, positive; (+), weakly positive; −, negative

NR not reported, DPG diphosphatidylglycerol, PG phosphatidylglycerol, PI phosphatidylinositol, PE phosphatidyethanolamine, PL phospolipid, GL Glycolipid, L Lipid

https://static-content.springer.com/image/art%3A10.1007%2Fs10482-009-9367-y/MediaObjects/10482_2009_9367_Fig3_HTML.jpg
Fig. 3

Polar lipid profile of CB31T after two-dimensional thin layer chromatography. DPG diphosphatidyl glycerol, PG phosphatidyl glycerol, PL1-PL2 unidentified phospholipids, GL1-GL4 glycolipids, L1 unidentified lipid, that produces an atypical color with the sugar reagent

Table 2

Fatty acid pattern of CB31T and its closest validly described relatives

Fatty acids (%)*

CB 31T

Tessaracoccus flavescens

Tessaracoccus bendigonensis

Tessaracoccus lubricantis

Saturated fatty acids

    C12:0

3.2

    C14:0

0.6

1.4

2.0

1.3

    C15:0

1.1

    C16:0

1.3

11.5

15.6

2.6

    C17:0

0.2

    C18:0

17.5

24.4

Unsaturated fatty acids

    C18:1ω9c

2.2

1.4

    C20:1ω9c

1.3

Branched fatty acids

    iso-C14:0

4.9

2.4

3.0

2.0

    iso-C14:0 3-OH

1.2

    iso-C15:0

0.4

3.2

1.3

3.4

    iso-C16:0

4.9

5.0

3.6

3.8

    anteiso-C15:0

83

49.6

27.5

83

    anteiso-C16:0

1.0

    anteiso-C17:0

1.5

3.6

3.8

    iso-C16:1

2.0

    anteiso-C17:1

2.5

Data for T. bendigoensis DSM 12906T were taken from Maszensan et al. (1999) and Lee and Lee (2008)

Data for T. flavescens DSM 18582T were taken from Lee and Lee (2008)

Data for T. lubricantis DSM 19926T were taken from Kämpfer et al. (2009)

* Values are percentages of total fatty acids

Similar to the other members of the genus Tessaracoccus (Maszensan et al. 1999; Lee and Lee 2008; Kämpfer et al. 2009) the major menaquinone of strain CB31T was MK-9(H4) (70%), which is present in all members of the family Propionibacteriaceae (Stackebrandt and Schaal 2006). In addition, MK-9, MK-9(H2), MK-9(H6) as well as traces of MK-9(H8) were detected. In contrast to T. flavescens and T. bendigonensis, menaquinone MK-7(H2) (Lee and Lee 2008; Maszensan et al. 1999) was not found in strain CB31T.

Physiology

Growth occurred between 10 and 45°C, with an optimum between 30 and 40°C (Table 1). No growth was observed at 50°C. Strain CB31T grew between pH 5.7 and 9.3 exhibiting optimal growth at pH 7.5 (Table 1). Growth was sustained from 0 up to 5% NaCl (w/v) with a growth optimum at 2% NaCl (w/v) (Table 1). Strain CB31T differs from its closest relatives within the genus Tessaracoccus by having an extended temperature range (Table 1). Strain CB31T grew in the absence of oxygen by fermenting sugars and by reducing nitrate to nitrite and was catalase positive and oxidase negative. These characteristics are shared by all species within the genus Tessaracoccus. In the absence of an external electron acceptor, strain CB31T produced acid from the following sugars (API 20E): arabinose, glucose, rhamnose, melobiose and saccharose, and reduced tetrazolium with the following substrates (BIOLOG GP2): adenosine, d-ribose, dextrin, α-d-glucose, glycogen, maltotriose, acetic acid, α-hydroxybutyric acid, l-lactic acid, d-lactic acid methyl-ester, pyruvic acid, pyruvic acid methyl ester, glycerol, tween 40 and tween 80. Strain CB31T differs from the other members of genus Tessaracoccus by its inability to utilize d-fructose, d-galactose, maltose, lactose and d-xylose. Activity of the following enzymes could be detected using the APIzym test-kit: acid phosphatase, alkaline phosphatase, cystine arylamidase, esterase, α-galactosidase, α-glucosidase, β-glucosidase, leucine arylamidase, lipase, α-mannosidase, N-acetyl-β-glucosaminidase, naphthol-AS-BI-phosphohydrolase, valine arylamidase. The following tests were positive on API 20E and API 20NE test strips: acetoin production, citrate utilization, β-galactosidase, β-glucosidase (aesculin), nitrate reduction, tryptophan deaminase.

Strain CB31T was resistant to the following antibiotics (μg of antibiotic per disc): Aztreonam (30), kanamycin (100) nalidixan (130), polymyxins (colistin, 150) and susceptible to ampicillin (33), bactricin (40), cefaclor (30), chloramphenicol (60), doxycycline (80), erythromycin (78), fucidin (100), gentamycin (40), kanamycin (500), neomycin (120), novobiocin (5), benzyl penicillin (5), streptomycin (100), tetracycline (10), vancomycin (70), rifampicin (30). In a comprehensive study on antibiotic resistance in bacteria isolated from the deep terrestrial subsurface, Brown and Blakwill (2009) report that resistance against a large number of antibiotics is widespread in particular among Actinobacteria. Also strain CB31T is resistant against several antibiotics affecting different aspects of cell proliferation and synthesis such cell wall synthesis (aztreonam), protein synthesis (kanamycin), RNA synthesis (nalidixan) and the disruption of the inner and outer membrane of Gram-negative cells (polymyxin). These results add to the growing body of evidence that antibiotic resistance is a common trait among environmental bacteria and that antibiotics possibly serve another role than growth inhibition in natural microbial communities (Davies 2006).

Ecological aspects

There are after all several examples of heterotrophic bacteria isolated from diverse deep subsurface environments and in many cases the isolates belong to the phylum Actinobacteria (Blakwill et al. 1997; Crocker et al. 2000; Brown and Blakwill 2009).

The large spectrum of substrate classes that can be used by strain CB31T, in addition to the fact that it both can grow in a fermentative and respiratory mode with nitrate, would allow it to exploit a variety of diverse compounds, which would improve its ability to survive under the nutrient deprived conditions of the subsurface.

The in situ conditions within the section of the core from which the sample was retrieved were as follows: temperature 45–47°C, NaCl concentration 40–45 g l−1 and the pH was uniform through the whole section at pH 7.2 (Cockell et al. unpublished data). Thus strain CB31T grows in vitro under conditions that are compatible with the conditions that are found in situ. The in situ temperature may, however, be critical for strain CB31T since it is at the upper limit of its temperature spectrum. This must be stressful for the cells and they would very likely have to allocate energy for maintenance, an aspect that we have not studied yet. Studies by Johnson et al. (2007) on 700 ky old permafrost samples have shown that Actinobacteria, the class to which strain CB31T belongs, have a high DNA repair capacity, a feature that may facilitate survival of strain CB31T at stressfully high temperature conditions.

We cannot exclude that strain CB31T cells originate from an upper part of the sediment column and were transported into the deeper strata by percolating groundwater. Since the hydraulic conductivity of the region (800–1,400 m) from which strain CB31T was isolated is poor due to its high clay content, transport of cells from upper regions into that zone appears to be limited (Cockell et al. (2009)). In addition, isolation efforts with samples from upper parts of the core have not resulted in the isolation of other CB31T strains nor have CB31T related sequences been recovered in clone libraries from these samples. These observations support the hypothesis that strain CB31T is restricted to this zone and is not continuously fed into it from above. Thus ancestors of strain CB31T may have been washed into the crater when it filled up with debris after the impact blast.

Finally, we also have to take into consideration the problem of cross contamination as the source of our isolate when collecting and handling the core samples. We have collected and handled the samples in accordance with the recommended protocols to avoid contamination (Gronstal et al. 2009) and are confident that this is not the source of our isolate.

Conclusion

On the basis of its phenotypic and genotypic characteristics strain CB 31T is assigned to a new species within the genus Tessaracoccus, for which the name Tessaracoccus profundi sp. nov. is proposed. The type strain of Tessaracoccus profundi is CB31T (=NCIMB 14440T = DSM 21240T).

Description of Tessaracoccus profundi sp. nov.

Tessaracoccus profundi (pro.fun’di. L. gen. n. profundi of depth)

Gram positive, non-endospore-forming, facultatively anaerobic, non-motile, slightly curved rods, about 1.5 μm long and about 0.5 μm wide. Cells are oxidase negative and catalase positive. The cell wall contains ll-DPM-Gly with the peptidoglycan type A3γ′. The major menaquinones are MK-9(H4) (70%), MK-9(H2), MK9, MK-9(H6), with traces of MK-9(H6). The major fatty acid is anteiso 15:0. Genomic DNA G + C content is 68 mol%. Colonies on R2A agar supplied with 20 g/l NaCl and 3 g/l MgCl2·6H2O are smooth and have an orange pigmentation. Growth occurs between 10 and 45°C, with an optimum between 35 and 40°C and within a pH range of 5.7–9.3 with an optimum pH of 7.5. Growth is sustained in the presence of up to 5% NaCl (w/v) with an optimum at 2% NaCl (w/v). Aerobic growth is supported by acetic acid, adenosine, arabinose, dextrin, α-d-glucose, glycerol, glycogen, α-hydroxybutyric acid, l-lactic acid, d-lactic acid methyl-ester, maltotriose, melobiose, pyruvic acid, pyruvic acid methyl ester, rhamnose, d-ribose, saccharose, tween 40, tween 80. Anaerobic growth is supported by fermentation and by reduction of nitrate to nitrite. The following enzymatic activities are present: acid phosphatase, alkaline phosphatase, cystine arylamidase, esterase, lipase, α-galactosidase, α-glucosidase, N-acetyl-β glucosaminidase, leucine arylamidase, α-mannosidase naphthol-AS-BI-phosphohydrolase, valine arylamidase (API zym) and β-galactosidase, β-glucosidase (aesculin), tryptophan deaminase (API E and API NE). It does not possess arginine dihydrolase, chymotrypsin, α-fucosidase, gelatinase, lysine decarboxylase, ornithine decarboxylase, trypsin, urease. It is Voges–Poskauer positive and produces acetoin. The type strain CB31T (=NCIMB 14440T = DSM 21240T) was isolated from Chesapeake meteor impact crater drill core sample No. 31, obtained from 940 m depth.

Acknowledgments

We thank Tove Wiegers for expert technical assistance. We thank Prof. Rodney A. Herbert for critically reviewing the manuscript. We also acknowledge the critical comments of two anonymous reviewers that helped to improve the manuscript. Polar lipid analyses were carried out by the Identification Service of the DSMZ and Dr B. J. Tindall, DSMZ, Braunschweig, Germany. This study was supported by the Carlsberg Foundation, Grant No. 2005-1-275.

Supplementary material

10482_2009_9367_MOESM1_ESM.pdf (92 kb)
Supplementary material 1 (PDF 92 kb)

Copyright information

© Springer Science+Business Media B.V. 2009