• Cytochrome c nitrite reductase;
  • nrfA;
  • Gene probes;
  • Environmental genomics;
  • Gene detection by PCR


  1. Top of page
  2. Abstract
  3. 1Introduction
  4. 2Materials and methods
  5. 3Results
  6. 4Discussion
  7. Acknowledgements
  8. References

Degenerate primers to detect nrfA were designed by aligning six nrfA sequences including Escherichia coli K-12, Sulfurospirillum deleyianum and Wolinella succinogenes. These primers amplified a 490 bp fragment of nrfA. The ability of these primers to detect nrfA was tested with chromosomal DNA isolated from a variety of bacteria: they could distinguish between bacteria in which the gene is known to be present or absent. The positive reference organisms spanned the various classes of Proteobacteria, suggesting that these primers are probably generic. The primer pair F1 and R1 was also used successfully to analyse nrfA diversity from community DNA isolated from a sulphate reducing bioreactor, and from two established Anammox reactors (for an aerobic amm onia ox idation, in which nitrite is reduced by ammonia to dinitrogen gas). The nrfA clones isolated from these three sources grouped with the Bacteroidetes phylum. The nrfA primers also amplified 570 bp fragments from the Anammox community DNA. These fragments encoded a protein with four haem-binding motifs typical of a c-type cytochrome, but were unrelated to the NrfA nitrite reductase. A BLAST search failed to reveal similarity to any known proteins. However, similarity was found to one sequence, which was annotated as rapC (response regulator aspartate phosphatase), in the genome of the planctomycete Rhodopirellula baltica. These sequences possibly belong to a new class of c-type cytochrome that might be specific to members of the order Planctomycetales. The data are consistent with the proposal that cytochrome c nitrite reductases, present in the periplasm of Gram-negative bacteria, are widely distributed in many different environments where they provide a short circuit in the biological nitrogen cycle by reducing nitrite directly to ammonia.


  1. Top of page
  2. Abstract
  3. 1Introduction
  4. 2Materials and methods
  5. 3Results
  6. 4Discussion
  7. Acknowledgements
  8. References

The last 20 years have witnessed fundamental changes in our understanding of nitrogen transformations in the biosphere, not least with reactions that lead to the production or consumption of nitrite. For example, 20 years ago it was accepted that nitrification and denitrification required different groups of bacteria, but now it is known that some bacteria can catalyse both processes [1]. Similarly, some bacteria catalyse both denitrification and nitrogen fixation, depending on the chemical composition of their environment [2,3]. Nitrite reduction to gaseous products by denitrifying bacteria used to be considered to be a strictly anaerobic process, but this fallacy was dispelled with the discovery of aerobic denitrification [4]. Conversely, nitrification was considered to be an obligately aerobic process until the discovery that, in the Anammox process, the strictly anaerobic planctomycetes Candidatus“Brocadia anammoxidans”, Candidatus“Kuenenia stuttgartiensis” and Candidatus“Scalindua sorokinii” convert one mole each of nitrite and ammonia to dinitrogen and water [5–7]. Anammox bacteria are found associated with different types of bacteria in wastewater treatment plants and anaerobic marine sediments [7–9]. However, they cannot be cultured in the laboratory. Metabolic diversity occurs in conventional wastewater treatment plants, in the gastro-intestinal tracts of animals, in soil and in sediments. An outstanding challenge, therefore, remains the development of methods to detect the diversity of bacteria in mixed populations that contribute to, or compete with these newly discovered processes of the biological nitrogen cycle. To this end, degenerate oligonucleotide primer sets have been designed to detect the presence of genes encoding key enzymes in nitrogen transformations [10–14]. Primer sets provide the first step that, when combined with reverse transcriptase PCR (RT-PCR) to measure RNA transcripts, allow detailed analysis of gene expression even in highly complex environments.

Five main types of bacterial nitrite reductase have previously been characterised. The first committed step in denitrification is the reduction of nitrite to nitric oxide, catalysed by either the copper-containing nitrite reductase encoded by nirK, or the cytochrome cd1 nitrite reductase encoded by nirS. The other three types of nitrite reductase catalyse the reduction of nitrite to ammonia [15]. These are the assimilatory nitrite reductases that use reduced pyrimidine nucleotides or ferredoxin as the electron donor; the cytoplasmic NADH-dependent nitrite reductase (Nir) from fermentative bacteria; and the periplasmic nitrite reductase, NrfA (for n itrite r eduction by f ormate). In enteric bacteria, formate is the preferred physiological electron donor, and menaquinone mediates electron transfer to the periplasmic Nrf components. Our recent, unpublished analysis of whole genome databases suggests that the Nir pathway might be restricted to a narrow group of facultative anaerobic bacteria, for example, enteric bacteria, nitrate-reducing bacilli, and Staphylococcus carnosus. In contrast, cytochrome c nitrite reductases and nrfA genes encoding them have been found in a much wider range of bacteria, suggesting that the periplasmic nitrite reductases might be more significant environmentally than their cytoplasmic NirB counterparts. In the Anammox process, electrons released during ammonia oxidation drive nitrite reduction to dinitrogen. Whether a novel nitrite reductase or one of the well-characterised nitrite reductases is involved in this process is presently unknown. Oligonucleotide primers that detect the diversity of genes encoding nitrite reductases involved in denitrification have been described. However, the diversity of nirK and nirS genes that can be detected remains to be established [10,13]. Denitrification dominates environments rich in nitrate but relatively deficient in electron donors. Conversely, nitrate and nitrite reduction to ammonia dominate electron-rich environments where nitrate is available in low concentration, such as the gastro-intestinal tract of humans and animals, polluted estuarine sediments, and sulphide-rich environments [16–19]. The periplasmic nitrate and nitrite reductases, Nap and Nrf, play key roles in such environments that are cohabited by a diverse range of bacteria. Our aim was to design effective, highly degenerate oligonucleotide primers suitable for detecting the presence of nrfA genes in diverse bacterial communities. The immediate objective was to use the probes to determine whether nrfA-containing bacteria persist and whether the NrfA protein plays a significant role in different types of wastewater treatment plant, especially those designed to exploit the Anammox process.

2Materials and methods

  1. Top of page
  2. Abstract
  3. 1Introduction
  4. 2Materials and methods
  5. 3Results
  6. 4Discussion
  7. Acknowledgements
  8. References

2.1Strains and growth conditions

Escherichia coli strain JCB387 was from laboratory stocks [20]. E. coli strain LCB2048 was provided by G Giordano, CNRS, Marseilles, France. Bacteria were grown with aeration in 20 ml of Luria-Bertani broth (LB: [21]). Stenotrophomonas nitritireducens was provided by N-P Revsbech, Aarhus, Denmark; Alcaligenes faecalis and Alcaligenes defragrans were from K. van de Pas-Schoonen, Nijmegen, The Netherlands. These strains were grown statically at 30 °C for 7 days in test tubes containing 5 ml LB supplemented with 0.4% sodium succinate and 5 mM NaNO2. A Desulfovibrio fairfieldensis cell suspension was obtained from I Pereira, ITQB, Oeiras, Portugal.

2.2Sources of genomic DNA

Samples of genomic DNA from Desulfovibrio desulfuricans NCIMB8307, D. desulfuricans G200, D. vulgaris Hildenborough and D. salexigens NCIMB 8365 were gifts from G. Voordouw, Calgary; Canada. D. desulfuricans ATCC 27774 DNA was provided by L. Saraiva, ITQB, Oeiras, Portugal; Aeromonas caviae strain Sch DNA was from J. Shaw, Sheffield, UK; and Campylobacter jejuni strain 99/198 DNA was a gift from C. Penn, Birmingham, UK. DNA was also isolated from 1 ml samples of overnight cultures of E. coli strains, or from 5 ml of the S. nitritireducens, A. faecalis and A. defragrans cultures. Chromosomal DNA was isolated from cell pellets using the GFX Genomic Blood DNA purification kit (Amersham Pharmacia Biotech) following the method described for Gram-negative bacteria.

Community DNA from Anammox reactor samples obtained from Delft, The Netherlands, from Nijmegen, The Netherlands, from Santiago de Compostella, Spain, and from the sulphate reducing reactor from Paques, Balk, The Netherlands, was extracted using the GFX kit. Bacteria were resuspended in twice the recommended volume of Proteinase K buffer and incubated with Proteinase K for 30 min.

2.3PCR amplification

Touchdown PCR was used to amplify genomic DNA fragments. The reaction mixture contained in a final volume of 50 μl: 5 μl DNA (diluted 10–20-fold), 5 μl 10-fold concentrated NH4+ reaction buffer, 1.5 μl MgCl2 (50 mM), 1 μl dNTP (200 μM, Bioline), 5 μl of each forward and reverse primer (10 μM; Table 1) and 0.5 μl of Taq polymerase (Bioline). The programme consisted of the following steps: (1) 94 °C for 5 min; (2) 30 cycles at 94 °C for 1 min, starting with an annealing temperature of 60 °C for 1 min, which was decreased by 0.5 °C every cycle; 72 °C for 90 s; (3) 30 cycles of 94 °C for 30 s; 45 °C for 30 s, 72 °C for 1 min; (4): a final incubation at 72 °C for 10 min. Amplified products were checked by electrophoresis in 1.5% agarose gels and subsequent staining with ethidium bromide (0.5 μg ml−1).

Table 1.  Sequences of nrfA and T Easy primers used in this study
PrimerLocationSequence (5′ to 3′)
  1. R=A + G, S=C + G, W=A + T, Y=C + T, N=A + C + G + T.


2.4Cloning of PCR products into pGEM-T easy

Products obtained were excised from agarose gels and the DNA extracted using the Gel Extraction kit (Qiagen). The pGEM-T Easy System II (Promega) was used to ligate the DNA to the pGEM-T Easy vector by an overnight incubation at 4 °C. The ligated plasmid was transformed into competent JM109 cells (Promega). The transformants were selected by plating onto LB plates supplemented with X-gal, IPTG and ampicillin. The resulting clones from community DNA from Delft, Santiago de Compostella and Paques were given the prefix ANAD, ANAS and P, respectively. Plasmid DNA was extracted from overnight cultures of the clones by using a miniprep DNA kit (Qiagen). The cloned inserts were checked by agarose gel electrophoresis of Eco R1 digested miniprep DNA.

2.5Sequencing and phylogenetic analyses of the cloned fragments

The Plasmid to Profile sequencing service provided by the Functional Genomics Laboratory, School of Biosciences, University of Birmingham, was used to obtain DNA sequences of the cloned fragments. Plasmid DNA and the pGEM-T Easy primer (FWD or REV, Table 1) were mixed and the sequencing reaction set up using the Big Dye Terminator kit (Applied Biosystems) in a fully automated MWG Primus HT 96 well format PCR thermocycler. The sequencing products were purified and the sequences obtained by capillary electrophoresis on an ABI Prism 3700 DNA analyser. DNA and amino acid sequences were aligned with the program Pileup. Their identity was confirmed using the BLAST search program of the National Centre for Biotechnology Information website (

For phylogenetic analyses of nrfA genes and their deduced products, a nrfA sequence database for the ARB program package ( was created by downloading previously published nrfA gene and NrfA protein sequences from the GenBank database. After transfer to ARB, nucleic acid sequences were translated and the resulting protein sequences were aligned using ClustalW in the ARB software package. The alignment was corrected by visual inspection. Nucleic acid sequences were realigned according to the protein sequences. Gene fragments of nrfA derived in this study were translated into amino acids and automatically aligned to the NrfA protein sequences using the ARB alignment tool. Different treeing methods previously described for the ammonia monooxygenase sequences [22] were used for the phylogenetic analyses of the deduced NrfA sequences.


  1. Top of page
  2. Abstract
  3. 1Introduction
  4. 2Materials and methods
  5. 3Results
  6. 4Discussion
  7. Acknowledgements
  8. References

3.1Design of PCR primers for the detection of nrfA

At the start of this project, the E. coli NrfA amino acid sequence was used to search the GenBank database for sequences similar to the E. coli protein. Six sequences were retrieved including NrfA from Sulfurospirillum deleyianum and Wolinella succinogenes. These sequences were aligned using the programme Pileup (the first six sequences in Fig. 1). The conserved regions of the aligned sequences were used as the target for designing six degenerate primers, three forward primers, F1, F2 and F3, and three reverse primes, R1, R2 and R3 (Table 1). Their locations in the aligned sequences are shown in Fig. 1.


Figure 1. Alignment of NrfA sequences used (except the C. jejuni sequence) to design nrfA primers. The sequences were retrieved from GenBank and aligned by the programme Pileup from the Wisconsin Package Version 10.2, Genetics Computer Group, Madison, Wisconsin. The arrows represent the positions of the three forward primers (F1, F2 and F3) and the three reverse primers (R1, R2 and R3). The full names of the bacteria from which the sequences were derived were Wolinella succinogenes, Sulfospirillum deleyianum, E. coli O157 H7, Haemophilus influenzae, Pasteurella multocida and Campylobacter jejuni. Residues shaded black, dark grey or pale grey are 100%, more than 80%, or more than 60% identical, or with only functionally conservative substitutions, respectively.

Download figure to PowerPoint

3.2Performance of the degenerate primers with E. coli DNA

Optimum conditions for the PCR reactions were evaluated for various primer pairs. Initially E. coli DNA from plasmid p JG1, which encodes nrfA[20], and from two nrfA+E. coli strains, JCB387 and LCB2048, were used as templates to test the ability of the primers to direct the amplification of predicted nrfA fragments. Due to the high degeneracy of the primers, the best performance was obtained with a Touchdown PCR protocol. Data obtained with the various primer pairs tested together with their corresponding expected product sizes are summarised in Table 2. Products of the expected size were obtained with the plasmid p JG1 with all primer pairs tested, but the products with F1 + R2 and F2 + R2 were faint. With the two E. coli strains, the three forward primers yielded products of the expected sizes when paired with the reverse primers R1 and R3. The reverse primer R2 produced multiple bands with primers F1 and F2, the most intense bands being 2 and 1.5 kb in size. This primer probably recognises sequences elsewhere in the E. coli genome and was the only primer out of the six that was unsuitable for detecting nrfA.

Table 2.  Performance of the various primer pairs tested
Forward primerReverse primerExpected product (bp)Products obtained with E. coli DNA from
   Plasmid p JG1JCB 387LCB 2048
  1. + or − indicate that products of the correct size were detected or absent; −/+ indicates that a faint product was detected.


3.3Detection of nrfA genes in various bacteria

DNA samples isolated from a wide variety of bacteria were screened to determine whether the primers could discriminate between groups of bacteria in which the nrfA gene was known to be present or absent. The bacteria tested are listed in Table 3 together with the PCR products obtained. Either no products, or multiple bands that stained weakly with ethidium bromide, were obtained with template DNA from denitrifying bacteria such as A. defragrans and A. faecalis, which were expected to lack the nrfA gene. Conversely, single products of the expected sizes were obtained reproducibly with primer pairs F1 + R1 or F3 + R1 and DNA templates from diverse bacterial species whereas with F1 + R3 and F2 + R1 correct products were detectable but in some cases multiple bands were obtained. Fig. 2(a) shows the quality of products obtained with primers F1 and R1 with some of the templates tested. These fragments were cloned into the vector, pGEM-T Easy, and sequenced. The sequences obtained confirmed that the primers had identified the correct fragment.

Table 3.  List of bacteria screened for the nrfA gene
Bacteria testednrfA known to be present/absentProducts obtained with primers F1 + R1
Aeromonas caviaePresent500 bp product
Alcaligenes defragransAbsentNone
Alcaligenes faecalisAbsentNone
Campylobacter jejuniPresent500 bp
Desulfovibrio desulfuricans
 ATCC 2774Present500 bp
 NCIB 8307Unknown500 bp
Desulfovibrio fairfieldensisUnknown500 bp
Desulfovibrio salexigens NCIB 8365Unknown500 bp
Desulfovibrio vulgaris HildenboroughPresent500 bp
Paracoccus pantotrophusAbsentNone
Paracoccus versutusUnknownNone
Stenotrophomonas nitritireducensAbsentNone
Sulfurospirillum barnesiiPresent500 bp

Figure 2. (a) Detection of the nrfA genes from various bacteria. Primers F1 and R1 were used to amplify genomic DNA. Track 1: DNA standards; Tracks 2–7: PCR products obtained with DNA from plasmid p JG1, E. coli JCB387, E. coli LCB2048, Desulfovibrio desulfuricans 27774, Aeromonas caviae Sch3, Campylobacter jejuni 99/198, respectively. (b) PCR products produced by Anammox community DNA. Track 1: molecular mass markers; tracks 2–5 are the amplification products from duplicate reactions set up on different days using the same community DNA as template.

Download figure to PowerPoint

After the start of this study the sequence of a Campylobacter jejuni NrfA homologue became available. Despite far greater differences between the putative NrfA sequence of C. jejuni and the bacteria used to design the primers (Fig. 1), a single amplicon was generated using primers F1 and R1 with C. jejuni DNA as template (Fig. 2(a), track 7). This fragment, which due to six insertions between codons 168 and 364 was larger than the corresponding fragment from other bacteria, was also cloned and shown to be correct by sequencing. It was concluded that the primer pair, F1 and R1, was suitable for screening community DNA from mixed cultures for the presence of nrfA genes. As they amplified a larger fragment of nrfA than other primer pairs, this combination was selected for use in all subsequent experiments.

3.4Detection and diversity of nrfA genes in the Anammox community

As Anammox is a strictly anaerobic process and the planctomycetes involved cannot be isolated in pure culture, it was of interest to determine whether the nrfA gene could be detected in the Anammox community. Laboratory scale Anammox reactors of various sizes have been maintained in Delft, Nijmegen and Santiago de Compostella. Primers F1 and R1 were used initially to screen community DNA isolated from cells from an Anammox reactor maintained in Delft. An intense but broad double band of about 500 bp was detected by electrophoresis in an agarose gel (Fig. 2(b)). The reaction mixtures were pooled, separated on a preparative agarose gel, and the 500 bp products were recovered. The purified DNA from three independent experiments was cloned into pGEM-T Easy vector and more than 90 of the resulting clones were screened for the presence of different nrfA sequences.

Sequenced fragments were either 490 or 570 bp long, consistent with the double band apparent in Fig. 2(b). The 490 bp fragments were confirmed as nrfA by a BLAST search. An alignment of the corresponding amino acid sequences showed that they fell into three distinct groups, represented by clones ANADA06, ANADB01 and ANADB07 (Fig. 3). The four expected haem-binding motifs were present in all cases except for clone ANAB07 in which the first C of the CXXCK motif was replaced by V. Whether this is due to an amplification error by Taq polymerase or an alternative haem-binding motif could not be verified. The data from these experiments suggest that three closely related strains of bacteria that encode NrfA are present in the Anammox community.


Figure 3. Amino acid sequences encoded by the 490 base pair fragments. Alignment of the amino acid sequences corresponding to the nrfA fragments amplified with community DNA isolated from Anammox reactor maintained at Delft (ANAD prefix) and Santiago de Compostella (ANAS prefix) and from sulphate reducing reactor (P prefix). The Pileup programme was used. Clones screened from one of the three experiments for the Delft Anammox community DNA are included in the diagram. Note that the sequencing primers enabled the presence of the primer sequences F1 and R1 to be confirmed, as shown in the figure. Residues shaded black, dark grey or pale grey are 100%, more than 80%, or more than 60% identical, or with only functionally conservative substitutions, respectively.

Download figure to PowerPoint

The amino acid sequences encoded by the 570 bp fragments also include four CXXCH haem binding motifs (Fig. 4; represented by clone ANADC04 in Fig. 5). These sequences showed no overall similarity to NrfA, but were probably amplified using the primer pair F1–R1 because 12 out of the 17 bp of the F1 sequence matched the amplified fragment, and all 17 bases of the R1 sequence were identical to the 3′ end of the amplified 570 bp fragment. Furthermore, the reading frame of the 570 bp fragment includes the amino acid sequence – DCHMP, which is highly conserved in NrfA and provided the basis for the design of primer R1. BLAST searches did not reveal any similarity between the 570 bp fragment and other known sequences in the database. However, the genome of the planctomycete, Rhodopirellula baltica[23], contained an open reading frame annotated as rapC (response regulator aspartate phosphatase), which was 37% identical to ANADC04. Based on these observations, it is possible that these fragments represent a c-type cytochrome that is specific to the Planctomycetales. It is unknown whether this c-type cytochrome represents a new class of nitrite reductase relevant to the Anammox reaction.


Figure 4. Amino acid sequence encoded by the 570 bp fragment amplified from the Anammox community DNA isolated from a reactor maintained in Delft. Amino acids in bold type show the four C–X–X–C–H haem-binding motifs.

Download figure to PowerPoint


Figure 5. “Fitsch–Margoliash” tree of NrfA protein sequences derived from databases and amplified nrfA fragments. Trees were constructed either including (not shown) or excluding primer sequences (as shown in the figure). No change in tree topology could be observed. Sequences used to design the nrfA primers are underlined. NrfA sequences of reference organisms from various bacterial groups that were used to test the specificity of the nrfA primers are shown in boxes. The sequences amplified from community DNA samples of the Anammox reactors of Delft and Santiago de Compostella as well as the sulfate reducing sludge from Balk are depicted in bold. The 570 bp fragment of the novel c-type cytochrome, including the RapC of Rhodopirellula baltica (indicated by **), are shown as the out-group on the top of the tree. The NrfA sequence of Rhodopirellula baltica is quoted by Glöckner et al. [23]. The bar indicates 10% sequence divergence.

Download figure to PowerPoint

3.5Diversity of nrfA genes in Anammox community DNA obtained from Santiago de Compostella

Community DNA from a second Anammox reactor was next analysed for nrfA diversity to determine whether species detected in the Delft sample were also present in other Anammox reactors. Twenty-one of the 25 clones sequenced produced 490 bp fragments with primers F1 and R1, and these were identified as nrfA fragments. The corresponding amino acid sequences were essentially similar but on the basis of minor variations five nrfA sequence types could be identified. None of these sequences were identical to those obtained from the Delft samples, indicating that different strains were present in this community. These are represented by ANASD1, ANASD2, ANASD5, ANASE3 and ANASF2 sequences in Fig. 3. The remainder of the clones produced 570 bp fragments with sequences that corresponded with an identity of 44% to the c-type cytochrome detected previously in the Delft sample (Fig. 4; represented by clone ANASF7 in Fig. 5).

3.6Diversity of nrfA genes in a sulphate reducing sludge

Sludge samples from a sulphate reducing system obtained from Paques, Balk, NL, were next analysed to determine whether the primers could amplify a diverse range of nrfA sequences from a fully functional industrial wastewater treatment plant and, if so, to compare the sequences obtained with those from the laboratory Anammox reactors. After PCR amplification of community DNA with primers F1 and R1, only 490 bp products were obtained. Subsequent cloning and sequencing of 16 clones showed them all to be nrfA fragments that encode three different amino acid sequences. These again were not the same as the Delft and Santiago samples, and are represented by sequences PA1, PB2 and PB11 in Fig. 3. No other c-type cytochrome fragments were amplified from this DNA.

3.7Phylogenetic analysis of nrfA clones

A phylogenetic tree was constructed using nrfA sequences retrieved from GenBank as well as from web sites of genomes currently being sequenced (Fig. 5). The cloned sequences, excluding the primers used to amplify the cloned fragments, were also included in this tree. Most of the cloned sequences grouped together with Bacteroides thetaiotaomicron in the Bacteroidetes phylum. Exceptions were two clones (PB2 and PB8) from the sulphate reducing sludge, which grouped with Geobacter, and one clone (ANASF2), which grouped with Desulfotobacterium. The 570 bp c-type cytochrome fragments and the related Rhodopirellula baltica RapC sequence are shown as an outgroup.


  1. Top of page
  2. Abstract
  3. 1Introduction
  4. 2Materials and methods
  5. 3Results
  6. 4Discussion
  7. Acknowledgements
  8. References

There were two primary objectives in this study. The first was to design degenerate oligonucleotide primers suitable for detecting nrfA homologues encoding cytochrome c nitrite reductases in a variety of bacteria. These primers would be used to amplify, by PCR, fragments of the nrfA genes from mixed cultures of bacteria, including bacteria that cannot be cultured, from wastewater treatment plants and other environmental sources. A second objective was to determine whether the Anammox community contained bacteria capable of reducing nitrite to ammonia.

Six primers were designed from an alignment of known nrfA sequences. Some of these primers were shown to detect nrfA fragments from a wide variety of bacteria that spanned the various bacterial divisions. Some of the sequences amplified share little sequence identity, for example, nrfA from Campylobacter and Desulfovibrio. However, the ability to detect such extensive sequence divergence of nrfA genes from various genera and to detect low abundance sequences in environmental samples required several compromises, so caution must be exercised in their use. The primers are of necessity highly degenerate, so a touch-down PCR protocol was necessary to decrease the amplification of genes unrelated to nrfA. Cloning and sequencing of PCR products was essential to confirm the presence of nrfA in DNA isolated from mixed cultures. Note also that the use of highly degenerate primers coupled with 30 PCR cycles will bias the amplification of some sequences compared to others [24], so although the protocols described are effective in amplifying low abundance species, they are unsuitable for determining the relative abundance of different species in a mixed community. With these caveats, the primer pair F1 + R1 appear to be excellent universal primers for detecting nrfA in environmental samples.

Primer F1 covers the codons for the first haem-binding motif of NrfA, which in all six sequences initially available in the databases includes a conserved lysine rather than a histidine as a ligand to the haem iron. When the primers were designed, this C–X–X–C–K motif was believed to be an invariant signature motif typical of the nitrite-binding site of NrfA [25,26]. The nrfEFG genes are required for haem to be attached to this motif [26,27]. However, subsequent genome sequencing revealed the presence of nrfA in some bacteria, for example, Campylobacter jejuni (Fig. 1) that lack the specialised haem lyase encoded by nrfEFG. In these bacteria, the C–X–X–C–K motif in the first haem-binding site is replaced by a conventional C–X–X–C–H motif. It was significant, therefore, that primers F1 + R1 amplified a single nrfA fragment from C. jejuni DNA.

Fragments of DNA related to nrfA were amplified from community DNA isolated from two Anammox reactors. The Anammox reaction includes a nitrite reduction step, so it is possible that this reaction is catalysed by a planctomycete nitrite reductase related to NrfA. If so, the product of this reaction is likely to be hydroxylamine rather than ammonia [1,6]. The Anammox process requires strictly anaerobic conditions, so an alternative possibility is that the Anammox community includes contaminating bacteria that are anaerobes capable of reducing nitrite to ammonia. If so, this reaction would occur only if organic matter or another powerful reducing agent such as sulfide became available, conditions that are less favourable for the efficient operation of the Anammox process. If this alternative explanation is correct, it carries the important implication that there are contaminating bacteria in the Anammox community waiting to displace the planctomycetes required for efficient nitrogen removal from wastewater.

To date no nrfA gene has been detected in the partial genome assembly of Candidatus“Kuenenia stuttgartiensis”. The sequences derived from the reactor samples therefore probably reflect the contaminants that are present within the Anammox community. Phylogenetic analysis of the NrfA sequences obtained from the reactor samples demonstrated that most of them were affiliated to the Bacteroidetes phylum. This implies a degree of homogeneity of the communities studied. The association of Anammox bacteria with Bacteroidetes highlights an interesting question regarding their role. Bacteroidetes are present in diverse environments, play an important role in the turnover of organic matter and are enriched in bio-films [28,29]. This raises the intriguing possibility that a mutualism might occur between the Anammox organism and the Bacteroidetes. The former probably provides the organic matter in terms of dead cells for Bacteroidetes to flourish. The latter might contribute towards the formation of granules establishing the anaerobic environment for the Anammox bacteria as well as providing the ammonium, which is required for the Anammox reaction.

The Anammox communities were expected to be similar to each other, but the tight clustering of NrfA sequences from the Anammox reactors with those from the sulphate reducing sludge was surprising. No Desulfovibrio species were detected in this sample. However, the Bergey's taxonomic outline for Prokaryotes divides the Bacteroidetes phylum into 12 families [30], so the possibility remains that diversity might exist within this clustered group. This possibility can only be verified once nrfA sequences of some of the known members of the Bacteroidetes become available. On the other hand, the Anammox bacteria were originally isolated from a waste water system in the Netherlands [6] and the similarity of the sequences might be explained by the carry over of the Bacteroidetes community from the waste water system into the Anammox reactors.

The nrfA primers also detected a novel c-type cytochrome in the Anammox community. A homolog, which was annotated as rapC, was detected in a genome of the planctomycete, Rhodopirellula baltica[23]. This organism also has a nrfA gene that clustered between the Epsilonproteobacteria and Bacteroidetes (Fig. 5). Hence the possibility that Candidatus“Kuenenia stuttgartiensis” genome might also reveal a nrfA gene cannot be ruled out until the whole genome sequence has been assembled.

In conclusion, oligonucleotide primers are now available that are able to detect nrfA sequences from a very wide range of bacteria. This is significant for several reasons. First, it provides a method for detecting genes required for a short circuit in the biological nitrogen cycle, the rapid, respiratory reduction of nitrite to ammonia by fermentative bacteria (a process that was first described from environmental samples more than 25 years ago [16,31]). Secondly, the primers used successfully to amplify nrfA sequences were more highly degenerate than many used to probe other branches of the nitrogen cycle (see, for example, [11]). Significantly more successful was the use of Taq polymerase rather than several DNA polymerases with proof-reading functions: nevertheless, identical sequences were obtained repeatedly using the touch-down programme described. With the possible exception of clone ANADB07, the sequences listed in Fig. 3 are therefore due to genuine variations in strains because each sequence was detected repeatedly. Finally, cytochrome c nitrite reductases located in the bacterial periplasm provide an alternative to denitrification, but have been relatively little studied. They dominate many types of electron-rich environments, especially where only traces of nitrate are available [15,18,32]. The primers described will enable the full extent of this branch of the nitrogen cycle to be revealed even amongst bacteria that cannot be cultured in the laboratory.


  1. Top of page
  2. Abstract
  3. 1Introduction
  4. 2Materials and methods
  5. 3Results
  6. 4Discussion
  7. Acknowledgements
  8. References

The authors are grateful to G. Voordouw, University of Calgary, Canada for the chromosomal DNA from Desulfovibrio desulfuricans NCIMB8307, Desulfovibrio desulfuricans G200, D. vulgaris Hildenborough and D. salexigens NCIMB 8365; to Ligia Saraiva and Ines Pereira, ITQB, Oeiras, Portugal for D. desulfuricans ATCC 27774 DNA and D. fairfieldensis cells, respectively; to J. Shaw, University of Sheffield, UK for chromosomal DNA from Aeromonas caviae strain Sch; to C. Penn for Campylobacter jejuni DNA; to N.-P. Revsbech for cultures of Stenotrophomonas nitritireducens; to Anuska Mosquera, University of Santiago de Compostella, Spain for the Anammox community cells; to K. van de Pas-Schoonen for cultures of Alcaligenes defragrans and Alcaligenes faecalis and to Marga Breeuwsma, Paques, Balk, NL for the sulphate reducing sludge. This project was supported by EC Programme Framework V contract EVKI-CT2000–00054.


  1. Top of page
  2. Abstract
  3. 1Introduction
  4. 2Materials and methods
  5. 3Results
  6. 4Discussion
  7. Acknowledgements
  8. References
  • [1]
    Schmidt, I., Hermelink, C., van de Pas-Schoonen, K., Strous, M., op den Camp, H.J., Kuenen, J.G., Jetten, M.S. (2002) Anaerobic ammonia oxidation in the presence of nitrogen oxides (NO(x)) by two different lithotrophs. Appl. Environ. Microbiol. 68, 53515357.
  • [2]
    Fries, M.R., Zhou, J., Chee-Sanford, J., Tiedje, J.M. (1994) Isolation, characterization, and distribution of denitrifying toluene degraders from a variety of habitats. Appl. Environ. Microbiol. 60, 28022810.
  • [3]
    Velasco, L., Mesa, S., Delgado, M.J., Bedmar, E.J. (2001) Characterization of the nirK gene encoding the respiratory, Cu-containing nitrite reductase of Bradyrhizobium japonicum. Biochim. Biophys. Acta 1521, 130134.
  • [4]
    Robertson, L.A., Kuenen, J.G. (1984) Aerobic denitrification – old wine in new bottles. Antonie van Leeuwenhoek 50, 525544.
  • [5]
    Strous, M., Fuerst, J.A., Framer, E.H.H., Logemann, S., Muyzer, G., van de Pas-Schoon, K.T., Webb, R., Kuenen, J.G., Jetten, M.S.M. (1999) Missing lithotroph identified as new planctomycete. Nature 400, 446449.
  • [6]
    Kuenen, J.G., Jetten, M.S.M. (2001) Extraordinary anaerobic ammonium oxidising bacteria. Am. Soc. Microbiol. News 67, 456463.
  • [7]
    Schmid, M.C., Walsh, K., Webb, R., Rijpstra, W.I.C., van de Pas Schoonen, K., Verbruggen, M.J., Hill, T., Moffett, B., Fuerst, J., Schouten, S., Sinninge Damsté, J.S., Harris, J., Shaw, P., Jetten, M., Strous, M. (2003) Candidatus“Scalindua brodae”, sp. nov., Candidatus“Scalindua wagneri”, sp. nov., two new species of anaerobic ammonium oxidizing bacteria. Syst. Appl. Microbiol. 26, 529538.
  • [8]
    Fujii, T., Sugino, H., Rouse, J.D., Furukawa, K. (2002) Characterization of the microbial community in an anaerobic ammonium-oxidizing biofilm cultured on a nonwoven biomass carrier. J. Biosci. Bioeng. 94, 412418.
  • [9]
    Kuypers, M.M.M., Sliekers, A.O., Lavik, G., Schmid, M., Jørgensen, B.B., Kuenen, J.G., Sinninghe Damste, J.S., Strous, M., Jetten, M.S.M. (2003) Anaerobic ammonium oxidation by anammox bacteria in the Black Sea. Nature 422, 608611.
  • [10]
    Braker, G., Fesefeldt, A., Witzel, K.-P. (1998) Development of PCR primer systems for amplification of nitrite reductase genes (nirK and nirS) to detect denitrifying bacteria in environmental samples. Appl. Environ. Microbiol. 64, 37693775.
  • [11]
    Flanagan, D.A., Gregory, L.G., Caret, J.P., Karakas-Sen, A., Richardson, D.J., Spiro, S. (1999) Detection of genes for periplasmic nitrate reductase in nitrate respiring bacteria and in community DNA. FEMS Microbiol. Lett. 177, 263279.
  • [12]
    Gregory, L.G., Karakas-Sen, A., Richardson, D.J., Spiro, S. (2000) Detection of genes for membrane-bound nitrate reductase in nitrate respiring bacteria and in community DNA. FEMS Microbiol. Lett. 183, 275279.
  • [13]
    Hallin, S., Lindgren, P.E. (1999) PCR detection of genes encoding nitrite reductase in denitrifying bacteria. Appl. Environ. Microbiol. 65, 16521657.
  • [14]
    Nogales, B., Timmis, K.N., Nedwell, D.B., Osborn, A.M. (2002) Detection and diversity of expressed denitrification genes in estuarine sediments after reverse transcription-PCR amplification from mRNA. Appl. Environ. Microbiol. 68, 50175025.
  • [15]
    Potter, L., Angove, H., Richardson, D., Cole, J. (2001) Nitrate reduction in the periplasm of Gram-negative bacteria. Adv. Microbiol. Physiol. 45, 51112.
  • [16]
    Cole, J.A., Brown, C.M. (1980) Nitrite reduction to ammonia by fermentative bacteria: a short circuit in the Biological Nitrogen Cycle. FEMS Microbiol. Lett. 7, 6572.
  • [17]
    Forsythe, S.J., Dolby, J.M., Webster, A.D.B., Cole, J.A. (1988) Nitrate and nitrite-reducing bacteria in the achlorhydric stomach. J. Med. Microbiol. 25, 253259.
  • [18]
    Jorgensen, B.B. (1982) Ecology of the bacteria of the sulphur cycle with special reference to anoxic-oxic interface environments. Philos. Trans. R. Soc. Lond. B: Biol. Sci. 298, 543561.
  • [19]
    Potter, L.C., Millington, P., Griffiths, L., Thomas, G.H., Cole, J.A. (1999) Competition between Escherichia coli strains expressing either a periplasmic or a membrane-bound nitrate reductase: does Nap confer a selective advantage during nitrate-limited growth. Biochem. J. 344, 7784.
  • [20]
    Hussain, H., Grove, J., Griffiths, L., Busby, S., Cole, J. (1994) A seven gene operon essential for formate-dependent nitrite reduction to ammonia by enteric bacteria. Mol. Microbiol. 12, 153163.
  • [21]
    Sambrook, J., Fritsch, E., Maniatis, T. (1989) Molecular cloning: a laboratory manual. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY..
  • [22]
    Purkhold, U., Pommerening-Röser, A., Juretschko, S., Schmid, M.C., Koops, H.-P., Wagner, M. (2000) Phylogeny of all recognized species of ammonia oxidizers based on comparative 16S rRNA and amoA sequence analysis: implications for molecular diversity surveys. Appl. Environ. Microbiol. 66, 53685382.
  • [23]
    Glöckner, F.O., Kube, M., Bauer, M., Teeling, H., Lombardot, T., Ludwig, W., Gade, D., Beck, A., Borzym, K., Heitmann, K., Rabus, R., Schlesner, H., Amann, R., Reinhardt, R. (2003) Complete genome sequence of the marine planctomycete Pirellula sp. Strain. Proc. Natl. Acad. Sci. 100, 82988303.
  • [24]
    Polz, M.F., Cavanaugh, C.M. (1998) Bias in template-to-product ratios in multitemplate PCR. Appl. Environ. Microbiol. 64, 37243730.
  • [25]
    Darwin, A., Hussain, H., Griffiths, L., Grove, J., Sambongi, Y., Busby, S., Cole, J. (1993) Regulation and sequence of the structural gene for cytochrome c552 from Escherichia coli: not a hexahaem but a 50 kDa tetrahaem nitrite reductase. Mol. Microbiol. 9, 12551265.
  • [26]
    Grove, J., Busby, S., Cole, J. (1996) The role of the genes nrfEFG and ccmFH in cytochrome c biosynthesis in Escherichia coli. Mol. Gen. Genet. 252, 332341.
  • [27]
    Eaves, D.J., Grove, J., Staudenmann, W., James, P., Poole, R.K., White, S.A., Griffiths, L., Cole, J.A. (1998) Involvement of products of the nrfEFG genes in the covalent attachment of haem c to a novel cysteine-lysine motif in the cytochrome c552 nitrite reductase from Escherichia coli. Mol. Microbiol. 28, 205216.
  • [28]
    Cottrell, M.T., Kirchman, D.L. (2000) Natural assemblages of marine proteobacteria and members of Cytophaga-Flavobacter cluster consuming low and high molecular weight dissolved organic matter. Appl. Environ. Microbiol. 66, 16921697.
  • [29]
    O'Sullivan, L.O., Weightman, A.J., Fry, J.C. (2002) New degenerate Cytophaga-Flavobacter-Bacteroides-specific 16S ribosomal DNA-targeted oligonucleotide probes reveal high bacterial diversity in river Taff epilithon. Appl. Environ. Microbiol. 68, 201210.
  • [30]
    Garrity, G.M., Bell, J.A., Lilburn, T.G. (2003) Taxonomic outline of Prokaryotes. Bergey's manual of Systematic Bacteriology, Release 4.0, second ed. Springer-Verlag, New York..
  • [31]
    Simon, J. (2003) Enzymology and bioenergetics of respiratory nitrite ammonification. Mol. Microbiol. 26, 285309.
  • [32]
    Sorensen, J. (1978) Capacity for denitrification and reduction of nitrate to ammonia in a coastal marine sediment. Appl. Environ. Microbiol. 35, 301330.