Nitrosococcus watsonii sp. nov., a new species of marine obligate ammonia-oxidizing bacteria that is not omnipresent in the world's oceans: calls to validate the names ‘Nitrosococcus halophilus’ and ‘Nitrosomonas mobilis

Authors

  • Mark A. Campbell,

    1. Evolutionary and Genomic Microbiology Laboratory, Department of Biology, University of Louisville, Louisville, KY, USA
    Search for more papers by this author
  • Patrick S.G. Chain,

    1. Los Alamos National Laboratory, Genome Science Group, Bioscience Division, Los Alamos, NM, USA
    2. Metagenomics Program, Joint Genome Institute, Walnut Creek, CA, USA
    3. Center for Microbial Ecology, Michigan State University, East Lansing, MI, USA
    Search for more papers by this author
  • Hongyue Dang,

    1. Centre for Bioengineering and Biotechnology & State Key Laboratory of Heavy Oil Processing, China University of Petroleum (East China), Qingdao, China
    Search for more papers by this author
  • Amal F. El Sheikh,

    1. Evolutionary and Genomic Microbiology Laboratory, Department of Biology, University of Louisville, Louisville, KY, USA
    Search for more papers by this author
  • Jeanette M. Norton,

    1. Department of Plants, Soils and Climate, Utah State University, Logan, UT, USA
    Search for more papers by this author
  • Naomi L. Ward,

    1. Department of Molecular Biology, University of Wyoming, Laramie, WY, USA
    Search for more papers by this author
  • Bess B. Ward,

    1. Department of Geosciences, Princeton University, Princeton, NJ, USA
    Search for more papers by this author
  • Martin G. Klotz

    1. Evolutionary and Genomic Microbiology Laboratory, Department of Biology, University of Louisville, Louisville, KY, USA
    2. Centre for Bioengineering and Biotechnology & State Key Laboratory of Heavy Oil Processing, China University of Petroleum (East China), Qingdao, China
    Search for more papers by this author

  • Editor: Michael Wagner

  • Previous addresses: Patrick S.G. Chain, Lawrence Livermore National Laboratory, Livermore, CA 94550, USA.
    Naomi L. Ward, The Institute for Genomic Research, Rockville, MD 20850, USA.

Correspondence: Martin G. Klotz, Evolutionary and Genomic Microbiology Laboratory, Department of Biology, University of Louisville, 139 Life Sciences Building, Louisville, KY 40292, USA. Tel.: +1 502 852 7779; fax: +1 502 852 0725; e-mail: martin.klotz@louisville.edu

Abstract

Local associations between anammox bacteria and obligate aerobic bacteria in the genus Nitrosococcus appear to be significant for ammonia oxidation in oxygen minimum zones. The literature on the genus Nitrosococcus in the Chromatiaceae family of purple sulfur bacteria (Gammaproteobacteria, Chromatiales) contains reports on four described species, Nitrosococcus nitrosus, Nitrosococcus oceani, ‘Nitrosococcus halophilus’ and ‘Nitrosomonas mobilis’, of which only N. nitrosus and N. oceani are validly published names and only N. oceani is omnipresent in the world's oceans. The species ‘N. halophilus’ with Nc4T as the type strain was proposed in 1990, but the species is not validly published. Phylogenetic analyses of signature genes, growth-physiological studies and an average nucleotide identity analysis between N. oceani ATCC19707T (C-107, Nc9), ‘N. halophilus’ strain Nc4T and Nitrosococcus sp. strain C-113 revealed that a proposal for a new species is warranted. Therefore, the provisional taxonomic assignment Nitrosococcus watsonii is proposed for Nitrosococcus sp. strain C-113T. Sequence analysis of Nitrosococcus haoAB signature genes detected in cultures enriched from Jiaozhou Bay sediments (China) identified only N. oceani-type sequences, suggesting that different patterns of distribution in the environment correlate with speciation in the genus Nitrosococcus.

Introduction

Members of the genus Nitrosococcus are marine aerobic ammonia-oxidizing bacteria (AOB) that belong to the class Gammaproteobacteria, order Chromatiales (the purple sulfur bacteria). Aerobic chemolithotrophic oxidation of ammonia by bacteria proceeds in two consecutive steps. First, ammonia is converted to hydroxylamine via a multisubunit membrane-bound enzyme, ammonia monooxygenase (AMO, amoCAB), in the following reaction: NH3+2e+O2+2H+→NH2OH+H2O (Hooper et al., 2005). The subsequent oxidation of hydroxylamine to nitrite is facilitated by the soluble periplasmic enzyme, hydroxylamine oxidoreductase (HAO, haoA): NH2OH+H2O→NO2+5H++4e (Hooper et al., 2005). The oxidation of hydroxylamine to nitrite yields four electrons, two of which are returned to the upstream monooxygenase reaction, and the remaining two electrons are the sole source for generating useable energy and reductant. HAO is the key enzyme in bacterial ammonia catabolism due to its dual role as an extractor of reductant and as a detoxicant of the highly mutagenic intermediate hydroxylamine. For these reasons, use of its encoding gene, haoA, as a functional marker gene has been proposed (Klotz & Stein, 2008) and successfully tested in molecular ecological studies (Schmid et al., 2008). In contrast to the omnipresent and diverse betaproteobacterial AOB within the two genera Nitrosospira and Nitrosomonas, Nitrosococcus species are restricted to marine environments and salt lakes. Extensive sampling indicated that the species Nitrosococcus oceani is distributed in oceans worldwide (Ward & O'Mullan, 2002). Further, Nitrosococcus has recently been identified to cooperate as the dominant nitrifier with anammox bacteria in hypoxic layers of several oxygen minimum zones (OMZ), in which anammox dominates the process of N removal from the system (Lam et al., 2007).

The genus Nitrosococcus was originally established with Nitrosococcus nitrosus (Buchanan, 1925) as its type species. The type strain of this species has been lost from culture (agreement among attending scientists at the First International Conference on Nitrification, July 4–9, 2009, Louisville, KY), rendering N. nitrosus a type species with no validly described type strain. The only remaining validly described species with an assigned type strain in active culture is N. oceani (Watson, 1971; Skerman et al., 1980). While Koops et al. (1990) isolated and first described another species of Nitrosococcus, ‘Nitrosococcus halophilus’, this species name has not yet been validated despite being listed in Bergey's Manual of Systematic Bacteriology with strain Nc4 as the proposed type strain (Garrity & Holt, 2001). The taxonomic confusion within the genus Nitrosococcus is further compounded by the description of ‘Nitrosococcus mobilis’ (Koops et al., 1976); this species name has never been validated. Twenty years later, analysis of 16S rRNA gene sequences from ‘N. mobilis’ strains Nc2 and Nm93 indicated that ‘N. mobilis’ and recognized Nitrosomonas species constitute a monophyletic assemblage within the Betaproteobacteria (Pommerening-Röser et al., 1996), followed by Purkhold et al. (2000), Aakra et al. (2001) and others. The affilation of N. mobilis with the Betaproteobacteria was initially demonstrated by Woese et al. (1984) and later confirmed by Head et al. (1993) and Teske et al. (1994). Designation of Nc2 as the type strain of ‘N. mobilis’ has been suggested (Koops & Pommerening-Röser, 2005; Garrity et al., 2007); however, the binomial has never been validated.

Nitrosococcus oceani strains and ‘N. halophilus’ strain Nc4 can be distinguished on the basis of both phylogenetic analysis (16S rRNA gene sequences) and physiological criteria [i.e. N. halophilus is not ureolytic (Koops et al., 1990; Koops & Pommerening-Röser, 2001; Koper et al., 2004)]. However, distinction of an additional strain (Nitrosococcus sp. C-113, not yet formally described) from N. oceani has been marginal based on comparison of their 16S rRNA gene sequences (Ward & O'Mullan, 2002) and this might have prevented its recognition in reported environmental samples. The complete genome sequence of N. oceani ATCC19707 has been published recently (Klotz et al., 2006). Here we describe generation of additional genome sequences for Nitrosococcus sp. strain C-113 and ‘N. halophilus’ strain Nc4, and use of comparative genome analysis as a basis for proposed description of a new species of Nitrosococcus (Nitrosococcus watsonii) with its type strain C-113T. We also propose solutions to the current taxonomic disarray within the genus Nitrosococcus. Ecological support for the distinctiveness of N. oceani and ‘N. watsonii’ C-113 was obtained by investigating the molecular diversity of the gammaproteobacterial haoAB gene in an environment for which a highly diverse presence of betaproteobacterial AOB has been recently established (Dang et al., 2010). Our enrichment cultures from Jiaozhou Bay (China) contained exclusively haoAB genes with the highest sequence similarity to published sequences from N. oceani, but not ‘N. watsonii’ or ‘N. halophilus.’

Materials and methods

Bacterial strains, culture maintenance and DNA isolation

Nitrosococcus sp. strain C-113 was first described in 1965 by Stanley Watson (Fig. 1; Watson, 1965) and has been maintained since then as a nonaxenic enrichment culture with a predominantly coccus cell type. To purge the enrichment culture of heterotrophic bacterial contaminants visible by microscopy, the ammonia-containing growth medium [12.5 mM (NH4)2SO4] was amended with hydroxylamine and adjusted weekly to a final concentration of 0.2 mM. Medium acidification and nitrite production were monitored as described before for the growth of N. oceani (Alzerreca et al., 1999; Klotz et al., 2006). After 4 weeks of continuous exposure to hydroxylamine at 0.2 mM, inspection of the culture under the light microscope revealed only one cell type (Fig. 1). The presence of an axenic culture was also confirmed by the fact that only 16S rRNA gene sequences identical to the previously determined Nitrosococcus sp. strain C-113 (GenBank accession AF153343) were retrieved.

Figure 1.

 Electronmicrographs added in 1969 to the note of collection on November 13, 1967, on pages 112 and 113 in the culture book for AOB of Dr Stanley W. Watson.

In experiments following Koops et al. (1990) to determine the tolerance of Nitrosococcus sp. strain C-113 to its own substrate, 100 mL (in a 250-mL Erlenmeyer flask) of marine AOB medium with 12.5 mM (NH4)2SO4 were inoculated with Nitrosococcus sp. strain C-113 and incubated at 30 °C in the dark; the pH was adjusted daily with K2CO3. After 2 weeks, the culture was partitioned in 10-mL aliquots to new 100-mL Erlenmeyer flasks following adjustment of the (NH4)2SO4 concentration in duplicate flasks to 12.5, 100, 200, 400, 600 and 800 mM with a pH of 8. Like ‘N. halophilus’ (Koops et al., 1990), Nitrosococcus sp. strain C-113 does not grow at ammonium sulfate concentrations of 1 M or higher. Nitrite was measured immediately after inoculation and once daily as reported by Koops et al. (1990) for N. oceani and ‘N. halophilus’. Ammonium shift experiments were performed to identify the optimum concentration: 12.5 and 400 mM ammonium-grown cells were inoculated into fresh medium containing 100 mM ammonium sulfate and nitrite concentration was measured daily.

To prepare cells for genome sequencing, Nitrosococcus sp. strain C-113 was grown as 200–400 mL batch cultures, in 2-L Erlenmeyer flasks, for 3 weeks, at 30 °C, in the dark. Cultures were grown without shaking, in artificial seawater as described previously for N. oceani ATCC19707 (Alzerreca et al., 1999), with the modification of an increased sodium concentration (600 mM) and the presence of hydroxylamine (0.2 mM). The cultures were routinely checked for contamination.

For genomic DNA preparations, cells were harvested by centrifugation at late stationary growth phase as determined by spectrophotometric estimation of cell density and observation of decreased acid production. Genomic DNA was isolated using the Wizard genomic DNA purification kit (Promega, Madison, WI) according to the manufacturer's recommendations, and stored at −20 °C until further use. Genomic DNA was also obtained from cultures of ‘N. halophilus’ Nc4 using protocols described above for Nitrosococcus sp. strain C-113. The experimental details of sequencing the entire genomes of Nitrososoccus sp. strain C-113 and ‘N. halophilus’ Nc4 will be described elsewhere, together with the results of their analysis and characterization.

Environmental sample collection, enrichments and DNA isolation

Sediment samples were collected in October 2008 at specified locations (i.e. A3, B2, J1 or P1) in Jiaozhou Bay near Qingdao (China) and prepared as described recently (Dang et al., 2010). Enrichment cultures were started by mixing 0.5 g of frozen sediment into 200 mL of hydroxylamine-amended artificial seawater and maintained as standing batch cultures. Enrichments were propagated by transferring 5 mL of culture into 200 mL of fresh culture media after the phenyl-red pH indicator showed a change in color at least five times.

For DNA isolation, 200 mL of J1, B2, P3, D5, A3, A5 and P1 enrichment cultures were filtered (0.20 μm pore size, Fisher Scientific), the filters were cut in half and placed separately into two 1.5-mL tubes. One tube was stored at −80 °C and the other was used for subsequent DNA isolation and analysis. DNA was extracted using the UltraClean Microbial DNA Isolation Kit and a Fastprep® centrifuge as reported earlier (Poret-Peterson et al., 2008). The presence of sufficient DNA was verified for each sediment sample using the PCR and universal primers targeting the 16S rRNA gene.

PCR-cloning, sequencing and analysis of gammaproteobacterial haoAB gene fragments

PCR was performed using the degenerate primer pair HaoAF (5′-YTGYCAYAAYGGRGYNGAYCAYAAYGAGT) and HaoBRev1 (5′-ANNYGMYGRGMASYRTCCCACCA), designed for the detection of gammaproteobacterial haoAB genes, and the isolated enrichment community DNA as a template. The primers have been successfully tested using genomic DNA isolated from pure cultures of Gammaproteobacteria including N. oceani strain AFC27, ‘N. halophilus’, Nitrosococcus sp. strain C-113, Methylococcus capsulatus or Methylomicrobium album BG8 (data not shown). Genomic DNA from N. oceani ATCC19707 was used as a positive control. Amplicons, obtained only from the J1 enrichment culture and N. oceani ATCC19707, were cloned into pCR2.1-TOPO plasmids (Invitrogen, Carlsbad, CA). After transformation of the recombinant plasmids into chemically competent Escherichia coli TOP10 cells, insert-positive transformants were selected on Luria–Bertani medium agar supplemented with 50 μg mL−1 each of ampicillin and kanamycin and 40 mg mL−1 X-gal and transferred to a master plate. Ten colonies were randomly selected from the master plate for plasmid isolation (Wizard Plasmid Prep, Promega) and insert sequencing (University of Louisville DNA core facility). Insert nucleotide sequences were analyzed using sequencher (v4.6, Genecodes, Madison, WI) and seven (out of 10) nonidentical insert sequences and available Nitrosococcus haoAB gene sequences (Klotz et al., 2008; Schmid et al., 2008) were used to produce multiple sequence alignments with clustalx v.1.83 (Thompson et al., 1997) (IUB DNA weight matrix with gap opening and gap extension penalties of 35/15 and 0.75/0.35, respectively, for the pairwise/multiple sequence alignments). The DNA alignments were subjected to Bayesian inference of phylogeny using the beast package [beauti v1.5.3, beast v1.5.3, treeannotator v1.5.3, figtree v.1.3 (Drummond & Rambaut, 2007)]. Tree likelihoods (ignoring ambiguities) were determined for unique sites within the alignment by creating a Monte–Carlo Markov Chain (10 000 000 generations) in three independent runs. The searches were conducted assuming an equal distribution of rates across sites, sampling every 1000th generation and using strict molecular clock and HKY substitution models (Hasegawa et al., 1985). An unrooted 50% majority rule consensus phylogram was constructed.

Analysis of gammaproteobacterial rrsA, gyrB and rpoD gene sequences

For the AOB-specific proteins AmoA and HaoA, and for highly conserved rrsA, gyrB and rpoD genes, sequence similarities were investigated initially using the NCBI blast program (Altschul et al., 1990). Protein sequences deduced from experimentally determined nucleotide sequences were analyzed with the psort (Nakai & Kanehisa, 1991), and tmhmm (http://www.cbs.dtu.dk/services/TMHMM/) programs to identify hydrophobic domains that could serve as signal peptides for export into the periplasm or constitute membrane-spanning domains. The various deduced full-length protein sequences from Nitrososoccus sp. strain C-113 and ‘N. halophilus’ Nc4, respective blast hits retrieved from the GenBank/EMBL database and homologs found in unpublished genome sequences were used to produce multiple sequence alignments using clustalx v.1.83 (Thompson et al., 1997) (Gonnet 250 protein weight matrix with gap opening and gap extension penalties of 35/15 and 0.75/0.35, respectively, for the pairwise/multiple sequence alignments). A distance neighbor-joining guide tree was constructed using the bionj function in paup* v. 4.10b (Swofford, 1999) and observed differences between the resulting tree topology and known phylogenetic relationships were used for manual refinement of the clustalx alignments.

The refined protein and gene alignments were subjected to Bayesian inference of phylogeny using the beast package [beauti v1.5.3, beast v1.5.3, treeannotator v1.5.3, figtree v.1.3 (Drummond & Rambaut, 2007)]. Tree likelihoods (ignoring ambiguities) were determined for unique sites within the alignment by creating a Monte–Carlo Markov Chain (10 000 000 generations) in three independent runs. The searches were conducted assuming an equal or a gamma distribution of rates across sites, sampling every 1000th generation and using strict molecular clock and HKY substitution models (Hasegawa et al., 1985) for analyses of the 16S rRNA, gyrB and rpoD genes and the WAG empirical amino acid substitution model (Whelan & Goldman, 2001) for analyses of the AmoA and HaoA proteins. Unrooted 50% majority rule consensus phylograms were constructed for 16S rRNA (1428 nucleotides, Fig. 2a) and concatenated gyrB and rpoD genes (4363 nucleotides, Fig. 2b), for which only posterior probability values <1.0 are shown. Sources for the sequences used in the alignments are indicated in the respective phylogenetic trees. The mean branch lengths are characterized by a scale bar indicating the evolutionary distance (percent changes).

Figure 2.

 Unrooted 50% majority rule consensus phylograms constructed after Bayesian inference of phylogeny for 16S rRNA (1428 nucleotides, Fig. 2a) and concatenated gyrB and rpoD genes (4363 nucleotides, Fig. 2b). Only posterior probability values smaller than 1.0 are shown at the respective nodes. Sources for the sequences used in the alignments are indicated in the respective phylogenetic trees. Mean branch lengths are characterized by a scale bar indicating the evolutionary distance (percent changes).

Average nucleotide identity (ANI) between purple sulfur bacterial genomes

In order to unequivocally determine the species status of strain C-113, the ANI between the whole genome sequence of Nitrosococcus sp. strain C-113 (one chromosome and two plasmids) and genome sequences of N. oceani ATCC19707 (CP000127; plasmid: CP000126) as well as those of other purple sulfur bacteria, including N. oceani strain AFC-27 (ABSG00000000) and ‘N. halophilus’ strain Nc4 (1 chromosome and 1 plasmid; under embargo), Allochromatium vinosum strain DSM 180 (CP001896; plasmids: CP001897, CP001898), Halorhodospira halophila strain SL1 (CP000544), Alkalilimnicola ehrlichei strain MLHE-1 (CP000453), Thioalkalivibrio sp. strains HL-EbGR7 (CP001339) and K90mix (CP001905; plasmid: CP00196) and Halothiobacillus neopolitanus strain c2 (CP001801) was calculated. The ANI calculations were performed using the in silico DNA–DNA hybridization method (Konstantinidis & Tiedje, 2005; Goris et al., 2007) implemented in the jspecies software (http://www.imedea.uib-csic.es/jspecies/about.html; Richter & Rossello-Mora, 2009). blast options used were as follows; drop-off value for gapped alignment, x=150; penalty for a nucleotide mismatch, q=−1, filter query sequence, F; expectation value, e=1e−15; and number of processors to use, a=2. ANI calculation settings were 30% for identity (%), 70% for alignment and 1020 bp for length.

Strain availability

The GenBank/EMBL/DDBJ accession numbers for the 16S rRNA gene sequences are as follows: AF152243 for strain C-113; AF287298, AJ298748, Nhal_R0018 and Nhal_R0025 for strain Nc4 T; and M96403, AF287297 and AJ298728 for strain Nc2 T. Prior attempts to maintain these strains in pure culture in major type culture collections have failed. The strains above are presently in pure culture in the following research laboratories: N. watsonii strain C-113T, M.G. Klotz (University of Louisville), J.M. Norton (Utah State University) and the WHOI culture collection; ‘N. halophilus’ strain Nc4T, M.G. Klotz (University of Louisville), Youn Hwan (California State University at Fresno) and A. Pommerening-Röser (Universtät Hamburg); ‘N. mobilis’ strain Nc2T, A. Pommerening-Röser (Universtät Hamburg) and M. Wagner (Universität Wien).

Results and discussion

Phylogenetic analyses of the Nitrosococcus sp. strain C-113 AmoA and HaoA proteins generated results identical to phylogenetic analyses of AmoA (Tavormina et al., 2010) and HaoA (Klotz et al., 2008) proteins published recently. These results and the phylograms constructed for the 16S rRNA (Fig. 2a and Norton, 2010) and concatenated gyrB and rpoD genes (Fig. 2b) suggest that Nitrosococcus sp. strain C-113, ‘N. halophilus’ strain Nc4 and strains in the N. oceani species are different phylotypes. Interestingly, our analysis of Nitrosococcus haoAB gene sequences in cultures enriched from Jiaozhou Bay sediments (China) identified only N. oceani-type sequences (Fig. 3). This and the fact that ‘N. halophilus’ (strains Nc7 and Nc4 were isolated from a salt lake in Saudi Arabia and a salt lagoon off Sardinia in the Mediterranean Sea, respectively) and Nitrosococcus sp. strain C-113 (isolated in the Mediterranean Sea near Port Said) were, so far, retrieved in isolated locations whereas strains of N. oceani are found worldwide (Ward & O'Mullan, 2002) suggests that different patterns of distribution in the environment correlate with speciation in the genus Nitrosococcus.

Figure 3.

 Unrooted 50% majority rule consensus phylogram constructed after Bayesian inference of phylogeny for a 1266-nucleotide-long alignment of fragments of haoAB gene sequences obtained from Jiaozhou Bay sediment enrichment cultures, pure cultures of Nitrosococcus oceani ATCC19707 and AFC27, Nitrosococcus halophilus Nc4, Nitrosococcus watsonii C-113 and Methylomicrobium album BG8 (GQ471938) as the outgroup. Posterior probability values are shown at the nodes. Mean branch lengths are characterized by a scale bar indicating the evolutionary distance (percent changes).

Unequivocal evidence for taxonomic delineation at the species level was obtained by calculating the ANI of representative genome sequences. The ANI values of the Nitrosococcus strain C-113 with N. oceani ATCC19707 were ANIb=89.29% and and ANIm=89.87% (chromosome only, Fig. 4) as well as ANIb=89.34% and ANIm=89.85% (chromosome and plasmids), which is well below the 94% threshold for species delineation that corresponds to the recommended cut-off of 70% similarity in DNA–DNA hybridization experiments (Konstantinidis & Tiedje, 2005; Goris et al., 2007). This result and preliminary identification of massive differences in the arrangement of genes and gene clusters in these two Nitrosococcus genomes suggest that Nitrosococcus sp. strain C-113 is the representative of a species distinct from N. oceani and also from ‘N. halophilus’ (Fig. 3). This is also supported by significant differences in physiological properties between N. oceani and Nitrosococcus sp. strain C-113 as summarized in Table 1. In experiments designed to determine ammonium salt tolerance (Koops et al., 1990), Nitrosococcus sp. strain C-113, like the more distantly related ‘N. halophilus’, was significantly less tolerant to higher ammonium concentrations than N. oceani (Fig. 5). In contrast to N. oceani, ‘N. halophilus’ and Nitrosococcus sp. strain C-113 do not grow at ammonium salts concentrations of 1 M or higher, but the optimum concentration for all three species is at around 100 mM ammonium salt, as indicated by a maximum of nitrite production at this concentration (Fig. 5 and Koops et al., 1990). This is also supported for Nitrosococcus sp. strain C-113 by ammonium shift experiments to the optimum concentration (inset to Fig. 5).

Figure 4.

 Plot of the calculated ANI as a function of blastn-based similarities of the 16S rRNA gene sequences for pair-wise comparison between known, genome-sequenced purple sulfur bacteria and Nitrosococcus watsonii strain C-113. ANIb, ANI using blast alignments; ANIm, ANI using mummer alignments. Triangles, pairwise calculations including N. watsonii genome data; open circles, all other purple sulfur bacterial genome comparisons. 1, Noce ATCC19707/Noce AFC27; 2, Nwat C-113/Noce ATCC19707; Nwat C-113/Noce AFC27; 3, Nhal Nc4/Nwat C-113; Nhal Nc4/Noce ATCC19707; Nwat Nc4/Noce AFC27; 4, Hhal SL1/Noce ATCC19707; Hhal SL1/Noce AFC27; 5, circled data points: Noce or Nwat or Nhal/Alvin or Tgr7 or TK90.

Table 1.   Selected ecological, physiological, cellular and genetic properties of cultured Nitrosococcus species
SpeciesNitrosococcus oceaniNitrosococcus halophilusNitrosococcus watsonii
(type strains)ATCC19707 (C-107, Nc10)Nc4 (not validated)C-113 (not validated)
  1. Data presented in this table are a compilation of results from unpublished work in the Klotz laboratory and the following references: Fiencke & Bock (2006), Koops et al. (1990), Koops & Pommerening-Röser (2001), Koper et al. (2004) and Ward & O'Mullan (2002).

Preferred habitatMarineMarine and salt lakesMarine
Points of isolationOmnipresentSalt lagoon or lakeMediterranean Sea, Port Said
Salt requirementsObligately halophilicObligately halophilicObligately halophilic
NaCl optimumAround 500 mMAround 700 mMAround 600 mM
Temperature optimum28–32°C28–32°C28–32°C
pH optimum7.6–87.6–87.6–8
Ammonium tolerance<1200 mMup to 600 mM<1600 mM
Ammonium optimum100 mM100 mM100–200 mM
Urease genes/ureolyticPositive/positiveNegative/negativePositive/negative
Cell shapeCoccus or short rodCoccus or short rodCoccus or short rod
Membrane structurePM and intracytoplasmic membrane stackPM and intra-cytoplasmic membrane stackPM and intra-cytoplasmic membrane stack
FlagellationPolar or lophotrichousLophotrichousPolar (tufts not observed)
Genome G+C content50.3%51.6%50.1%
Nucleoid3 481 691 bp4 079 427 bp3 328 570 bp
Plasmids1 (40 420 bp)1 (65 833 bp)2 (39 105 bp, 5611 bp)
Inclusion bodiesGlycogen/starch; poly-PiGlycogen/starch; poly-PiGlycogen/starch; poly-Pi
Figure 5.

 Plot of relative nitrite production by Nitrosococcus watsonii strain C-113T, Nitrosococcus oceani strain Nc1 and ‘Nitrosococcus halophilus’ strain Nc4T when grown for 8 days at different concentrations of ammonium. Superscript ‘a’ indicates that the values for N. oceani and ‘N. halophilus’ were obtained from Koops et al. (1990). The inset shows the results of ammonium shift experiments (12.5 and 400 mM ammonium sulfate-grown cells were inoculated into fresh medium containing 100 mM ammonium sulfate and the nitrite concentration was measured daily).

Based on the data presented here, we propose and call for the validation of the provisional taxonomic assignment ‘N. watsonii’ with strain C-113 as the type strain. Furthermore, we call for validation of strain Nc4 as the type strain for ‘N. halophilus’ (Koops et al., 1990), thereby justifying inclusion of both species among the validly described species of the genus Nitrosococcus.

The original isolate of N. nitrosus was not a marine strain (Winogradsky, 1892; Koops et al., 1976), and no isolate or DNA that is not a phylotype (ribotype) similar to N. oceani or ‘N. halophilus’ (Fig. 2; Ward & O'Mullan, 2002) has since been retrieved from the open ocean. Koops et al. (1976) were not able to unambiguously classify the isolated coccus-shaped strain Nc2 either as Nitrosomonas or Nitrosococcus according to the existing key (Watson, 1974); however, to avoid the creation of a new genus and because of its morphological similarities to the type species N. nitrosus, they classified strain Nc2 in the genus Nitrosococcus (Koops et al., 1976).

Although coccoid forms are not currently found within Nitrosomonas, the order Nitrosomonadales [in which Nitrosomonas is located (Skerman et al., 1980)] also includes Nitrosospira (Skerman et al., 1980; Head et al., 1993; Purkhold et al., 2000; Aakra et al., 2001), members of which are not rod-shaped. Like all described strains of Nitrosomonas europaea and Nitrosomonas eutropha, ‘N. mobilis’ strains can also tolerate high concentrations of ammonia (Juretschko et al., 1998), providing another unifying phenotypic trait. Most importantly, the strains of ‘N. mobilis’ are Betaproteobacteria that phylogenetically cluster monophyletically with N. europaea and N. eutropha (Head et al., 1993; Pommerening-Röser et al., 1996; Purkhold et al., 2000; Aakra et al., 2001; Koops & Pommerening-Röser, 2005; Garrity et al., 2007). Therefore, this publication shall also serve as a call for validation of the new species ‘N. mobilis’ with strain Nc2 as the type strain.

Recent literature contains numerous reports on the varied contributions of Thaumarchaea and Proteobacteria to aerobic ammonia oxidation in both aquatic and terrestrial environments. However, only a few papers included experimental data on the expression of the amoA marker gene and/or nitrification rates in addition to the commonly measured abundance of amoA gene copies in the isolated community DNA. Of these more comprehensive papers, two reported on a significant role of gammaproteobacterial AOB highly similar in sequence to N. oceani in the N cycle of OMZ. Based on molecular ecological (amoA) and isotope fractionation (14N/15N) studies, Lam and colleagues reported that archaea highly similar in sequence to Nitrosopumilus maritimus (Könneke et al., 2005; Walker et al., 2010) as well as N. oceani were important nitrifiers in the OMZ of the Black Sea, each providing about half of the nitrite required by anaerobically ammonia-oxidizing Planctomycetes, the anammox bacteria (Lam et al., 2007). While archaeal and bacterial nitrification and anammox were indirectly coupled in the upper layer, anammox and nitrification by N. oceani appeared to be directly coupled in the suboxic layer of the Black Sea OMZ, where archaeal amoA transcripts were scarce (Lam et al., 2007). In addition, the recent discovery of a significant contribution to the N cycle by N. oceani in the lower layer of the Peruvian OMZ (Lam et al., 2009) confirmed that extant AOB in the genus Nitrosococcus are ecologically significant and that their ancestors might have played an important role in the oceans of the Earth's proterozoic period (Staley, 2007). Given the pioneering role of the anammox process in the early nitrogen cycle (Klotz & Stein, 2008), detailed studies of the inferred direct mutualistic interaction between Nitrosococcus and anammox bacteria in low oxygen environments are needed.

Nitrosococcus is a rare obligate aerobic representative of the predominantly anaerobic purple sulfur bacteria (Imhoff, 2005) and its finding in the lower layers of OMZs is thus puzzling (Lam et al., 2007; Staley, 2007). The genome sequences of N. oceani ATCC19707 (Klotz et al., 2006) and unpublished Nitrosococcus genomes contain a richer and more diverse gene inventory in the quinol-oxidizing branch of the electron transport chain than is found in the genomes of betaproteobacterial AOB (Arp et al., 2007; Stein et al., 2007; Norton et al., 2008) and amoA gene-encoding archaea (Hallam et al., 2006; Walker et al., 2010). Enriched categories include complexes that conserve energy coupled to the reduction of oxygen and NO (Klotz & Stein, 2010). Knowledge of the taxonomic diversity and spatial distribution of gammaproteobacterial AOB in the world's oceans, as well as additional molecular and physiological studies, are thus ecologically relevant and it can be expected that this research will help us to understand better the functional intricacies of the extant N cycle.

Emended description of the genus Nitrosococcus (GenBank Taxonomy ID: 1227) synonym: NitrosococcusWinogradsky 1892; ex ‘Micrococcus nitrosusMigula (1900); ex ‘Nitrosococcus nitrosus’ Buchanan (Buchanan, 1925; Editorial_Board, 1955; Commission, 1958). Name approved by Skerman et al. (1980), but the culture was eventually lost. A new isolate ‘Nitrosocystis oceanus strain C-107’ (sic) Watson (1965), Nitrosococcus oceani strain ATCC19707 Watson (1971), name approved by Skerman et al. (1980) is now in culture, listed as the type strain and its genome has been sequenced (Klotz et al., 2006). We propose below the validation of ‘N. halophilus’ as a species name and an additional species of the genus Nitrosococcus (N. watsonii). The etymology of Nitrosococcus is catalogued using digital optical identifiers as follows: Genus (DOI: 10.1601/nm.2107), Family Chromatiaceae (DOI: 10.1601/tx.2070; Bavendamm, 1924), Order Chromatiales (DOI: 10.1601/nm.2069; Imhoff, 2005; List_Editor, 2005), Class Gammaproteobacteria (DOI: 10.1601/tx.2068), Phylum Proteobacteria (DOI: 10.1601/tx.808; Garrity & Holt, 2001).

Description of Nitrosococcus halophilus Nc4 (GenBank Taxonomy ID: 472759): synonym: N. halophilus strain Nc4. Etymology, locality as well as culture history as in Koops et al. (1990); phylogenetic assignment to the genus Nitrosococcus (Skerman et al., 1980) by this publication with Nc4 as the type strain.

Description of Nitrosococcus watsonii sp. nov. (GenBank Taxonomy ID: 105559) synonym: Nitrosococcus sp. C-113.

Etymology, locality as well as culture history:Nitrosus (Latin masculine adjective): nitrous; coccus: (Latin masculine adjective): sphere; watsonii (Latin masculine genitive name): from Watson. The name refers to the collector of the organism (ammonia oxidizer, marine) motivated by the fact that, so far, no ammonia-oxidizing bacterium has been named in honor of Dr Stanley W. Watson.

Collected in 1967 at Port Said, Mediterranean Sea, by Stanley W. Watson. Maintained until 1994 as enrichment culture C-113 by Frederica Valois (Woods Hole Oceanographic Institution). Identified as a gammaproteobacterial AOB in an enrichment culture in 1995 by J.M. Norton (Utah State University) and later purified using a hydroxylamine treatment regime by A.F. El Sheikh, M.A. Campbell and M.G. Klotz (University of Louisville).

Appearing as large cocci or very short rods. Cells contain a well-developed intracellular membrane system of an arrangement that appears as one stack of membrane vesicles packed mainly in the center of the cell. Numerous inclusion bodies identified by comparison with N. oceani as glycogen/starch granules (white color) and polyphosphate granules (dark color). Polar flagella allow for motility, the genotype is chemotaxis-positive. Light sensitive; strictly aerobic; moderately alkaliphilic and mesophilic; grows in the presence of sodium salts between 200 and 800 mM; optimum growth temperature is 28–32 °C; optimum salt concentration is 500–700 mM; the G+C content of the DNA is 50.1%; the genome of N. watsonii strain C-113T is presently being annotated and analyzed.

Emended description of the genus Nitrosomonas (GenBank Taxonomy ID: 1227) synonym: NitrosomonasWinogradsky 1892 ex Nitrosomonas (Editorial_Board, 1955) ex Nitrosomonas (Commission, 1958), emended by Watson 1974. Approved Lists 1980 (Skerman et al., 1980). We propose below the validation of ‘N. mobilis’ as an additional species of the genus Nitrosomonas and Nc2 as its type strain.

Description of Nitrosomonas mobilis sp. comb. nov. synonym: ‘N. mobilis’ (Koops et al., 1976); ex ‘N. mobilis’ (Garrity et al., 2007).

Etymology, locality and culture history as well as diagnosis as in Koops et al. (1976); phylogenetic assignment to Nitrosomonas, Nitrosomonadaceae, Nitrosomonadales, Betaproteobacteria (Head et al., 1993; Pommerening-Röser et al., 1996; Purkhold et al., 2000; Aakra et al., 2001; Koops & Pommerening-Röser, 2005; Garrity et al., 2007).

Acknowledgements

We would like to thank Dr Jean P. Euzèby (École Nationale Vétérinaire, Toulouse, France), Dr George M. Garrity (Michigan State University) and anonymous reviewers of a previous version of this manuscript for invaluable taxonomic advice. Pertinent taxonomic information was accessed through the ‘names for life’ online tool (http://namesforlife.com). Technical assistance by undergraduate student David Griffith (UofL) is acknowledged. This project was supported in part by incentive funds provided by the UofL-EVPR office (M.A.C. and M.G.K.), U.S. National Science Foundation grants EF-0412129 (A.F.E.S. and M.G.K.) and EPS-0447681 (N.L.W.), and the China National Science Foundation grant 41076091 (H.D. and M.G.K.). The genome sequencing work conducted by the U.S. Department of Energy Joint Genome Institute is supported by the Office of Science of the U.S. Department of Energy under Contract Number DE-AC02-05CH11231.

Ancillary