Nitrous oxide reductase (nosZ) gene-specific PCR primers for detection of denitrifiers and three nosZ genes from marine sediments


*Corresponding author. Tel.: +1 (732) 932-6555 x306; Fax: +1 (732) 932-6520; E-mail:


Two PCR primer sets for the nitrous oxide reductase gene (nosZ) were developed. The initial primers were based on three sequences in GenBank and used to amplify nosZ from continental shelf sediments and from two denitrifiers in culture, Thiosphaera pantotropha and Pseudomonas denitrificans. Three unique marine sediment nosZ genes were identified and sequenced. The marine nosZ genes were most closely related to the nosZ genes of Paracoccus denitrificans or to Rhizobium meliloti. Alignment of all nosZ sequences currently available (n= 10) facilitated redesign of the PCR primers. Three new primer sets which amplify 1100 bp, 900 bp and 250 bp regions of the nosZ gene were designed and tested. The new primers robustly amplified nosZ fragments from samples in which the initial nosZ primers were only marginally successful.


Denitrification is the reductive respiration of nitrate or nitrite to N2 or N2O, and is carried out by a phylogenetically diverse group of bacteria, generally under anaerobic conditions. In continental shelf sediments, denitrification is particularly important since it can decrease the amount of nitrogen available to phytoplankton in the overlying waters, and may remove over 50% of the nitrogen inputs to the oceans as a whole [1, 2]. Beyond its importance in the oceanic nitrogen cycle, denitrification produces nitrous oxide, a gas implicated in both ozone destruction and global warming [3, 4].

Classical microbiological techniques are now considered insufficient for studying naturally occurring assemblages of bacteria since the majority of bacteria are widely believed to be unculturable by traditional techniques [5]. The current revolution in molecular biology has allowed microbiologists to overcome the limitations of culturability. However, molecular identification based on 16S rRNA genes does not necessarily correspond to metabolic function. Therefore, to study a phylogenetically widespread process such as denitrification, we have chosen to target a functional gene specific to this metabolic pathway.

In this report, we describe a technique for detection of denitrifier-specific DNA present in continental shelf sediments by analyzing for the nitrous oxide reductase (nosZ) gene. Nitrous oxide reduction is the final step in the denitrification pathway and represents loss of biologically available N. The nosZ gene is largely unique to denitrifying bacteria, although a few non-denitrifier species capable of reducing nitrous oxide have been identified [6], and has been studied in three model microorganisms [7–9]. This project required the initial development of PCR primers specific to the nosZ gene based on the three models, amplification of nosZ from environmental samples, sequencing, and the subsequent redesign of the PCR primers. The ultimate goal of this work is to develop a methodology which will allow the study of denitrifying bacteria in a wide variety of environmental settings.

2Materials and methods

2.1Environmental site

Marine sediments were obtained from a long-term ecosystem observatory site (LEO-15) established by the Mid-Atlantic Bight National Undersea Research Center. The LEO-15 site is operated in 15 m of water on the continental shelf and has been described in detail elsewhere [10]. Sediment cores from the LEO-15 site were hand retrieved by SCUBA divers and stored frozen at −20°C until analysis.

2.2Extraction of DNA

Total genomic DNA was extracted from approximately 100 mg (wet weight) of sediment, using slight modifications of a phenol-chloroform extraction protocol [11]. Modifications included increasing the amount of resuspension buffer from 200 to 225 μl, increasing the lysozyme incubation from 10 to 15 min, and decreasing the amount of EDTA added from 100 to 75 μl. PCR inhibitors were removed using CsCl density gradient ultracentrifugation [12]. For Thiosphaera pantotropha and Pseudomonas denitrificans, DNA was extracted from 1.5 ml of an overnight culture aerobically grown in Luria-Bertani medium to turbidity.

2.3Development of PCR primers

For development of the initial nosZ primer set, three nosZ sequences from GenBank were aligned using CLUSTAL [13] and a Microsoft Excel coloring macro [14], for identification of potential PCR priming sites. The initial primer set, Nos661F/Nos2230R, was designed to amplify a 1.6 kb fragment (Table 1). Oligonucleotide primers for sequencing were also designed as needed and are listed in Table 1. These primers were used to identify and sequence five new nosZ genes as described below.

Table 1.  Oligonucleotides used for PCR amplification and sequencing of nosZ gene fragments
  1. *The numbers in the primer names are relative to the nosZ sequence of P. stutzeri Zobell.

PCR primers 
Sequencing primers 

With all nosZ sequences known to date (n= 10), the nosZ PCR primers were redesigned. The redesigned primers are designated Nos1527F, Nos1527R and Nos1773R (Table 1). Primers were checked for specificity using the Blast and Fasta nucleotide database search tools [15, 16].

2.4PCR amplification

PCR amplifications were performed in a DNA thermal cycler (Perkin Elmer 2400) using ∼20 ng template DNA, 20 pmol primers, and 2.5 mM MgCl2 (reaction volume=50 μl). Reaction conditions were 1 cycle at 94°C for 5 min, followed by 35 cycles of 95°C for 0.5 min, 56°C for 1.5 min, 72°C for 2 min, and a final extension step at 72°C for 10 min. Identity of nosZ PCR product was confirmed by Southern hybridization, using Dig-dUTP (Boehringer Mannheim, Indianapolis, IN) labeled nosZ fragment from Pseudomonas stutzeri Zobell.

2.5Cloning and sequencing of PCR products

PCR products were cloned using the pCRScript cloning kit (Stratagene, San Diego, CA) as per the manufacturer's instructions. Unique clones were identified by HaeIII restriction digestion. Plasmid DNA from unique transformants was purified using the FlexiPrep Kit (Pharmacia, Piscataway, NJ) and sequenced on an ABI 373 automated sequencer (Perkin Elmer/ABI). Sequences were analyzed using the Auto Assembler and SeqNavigator ABI software programs, as well as BLASTN [15]. The nosZ gene sequences from three marine clones obtained from an 11/95 continental shelf sediment sample, S321195A, S321195B, and S321195C, as well as those from T. pantotropha and Pseudomonas denitrificans have been given GenBank accession numbers AF016055–59, respectively. Sequence lengths are 1597 bp, 1597 bp, 1585 bp, 1539 bp, and 1567 bp respectively.


3.1PCR amplification with Nos661F and Nos2230R

Amplification of the nosZ gene from LEO-15 genomic DNA was attempted for three sediment samples (collected 11/95, 5/96, and 6/96) using the Nos661F/Nos2230R primer set. Weak amplification was achieved for the 11/95 and 6/96 sediment samples, while no amplification was observed in the 5/96 sample (Fig. 1, lanes D–F). This primer pair was also used to amplify the nosZ gene from T. pantotropha and Pseudomonas denitrificans.

Figure 1.

Agarose gel showing PCR amplification of nosZ gene fragments from continental shelf sediment DNA, using Nos661F/Nos2230R (lanes B–F) and Nos661F/Nos1773R (lanes H–L). Lanes: A, λHindIII DNA marker; B, H, no DNA control; C, I, P. stutzeri Zobell DNA (positive control); D, J, 11/95 sample; E, K, 5/96 sample; F, L, 6/96 sample; G, empty lane.

3.2Sequence analysis of nosZ gene fragments

The nosZ gene PCR product from the 11/95 sample was cloned, and restriction fragment length polymorphism (RFLP) analysis revealed the presence of three unique copies of the nosZ gene in the clonal library. The five nosZ gene sequences generated in this study (three environmental+two from cultures) and five nosZ sequences currently available from GenBank were aligned (Fig. 2). Approximately 30% of the 1600 bp are conserved across all ten nosZ genes, with conserved and variable regions dispersed throughout the alignment. For example, the first 100 nucleotides of the alignment are extremely variable (∼15% conserved nucleotides), while the 900–1000 bp and 1100–1200 bp regions are highly conserved (∼50% conserved nucleotides).

Figure 2.

A condensed DNA alignment of ten 1.6 kb nosZ gene fragments from various denitrifiers. The actual sequence for each gene has been omitted for clarity. Shading indicates a conserved nucleotide across all ten genes, and each of the four nucleotides is represented by a different shading style. The alignment is read from left to right, in blocks of 100 nucleotides for each of the ten sequences. Location and direction of the PCR priming sites are indicated by the arrows. S32A, S32B, and S32C: marine clones S321195A, S321195B, and S321195C, respectively; ACHR: Achromobacter cycloclastes; ALEU: Alcaligenes eutrophus; PSTU: Pseudomonas stutzeri Zobell; PSDN: Pseudomonas denitrificans; RHIZ: Rhizobium meliloti; THPA: Thiosphaera pantotropha; PADN: Paracoccus denitrificans.

Generation of a similarity matrix for all ten nosZ genes indicates that the marine sediment nosZ clones are no more than 63% similar to any of the nosZ genes of pure cultures (Table 2). Clones S321195A and S321195B were 97% similar to each other, and ranged from 38% similarity to Alcaligenes eutrophus to 63% for Paracoccus denitrificans. Clone S321195C ranged from 41% similarity to A. eutrophus to 61% for Rhizobium meliloti. The S321195C clone was 62% similar to the S321195A/B group.

Table 2.  Similarity matrix of ten nosZ gene sequences from different denitrifiers; the matrix was generated using the Genetic Data Environment [21] and the neighbor-joining distance method
Species% Similarity
Paracoccus denitrificans61.362.854.9      
Rhizobium meliloti60.459.861.458.7     
Achromobacter cycloclastes59.259.954.082.656.3    
Pseudomonas denitrificans59.057.752.661.561.661.3   
Thiosphaera pantotropha56.254.952.559.861.859.597.5  
Pseudomonas stutzeri Zobell55.955.654.060.061.357.686.687.1 
Alcaligenes eutrophus39.737.841.447.151.344.455.855.458.6

Phylogenetic tree reconstruction indicated that clones S321195A/B cluster with Paracoccus denitrificans and Achromobacter cycloclastes, while S321195C is nearest to R. meliloti. The T. pantotropha/P. denitrificans nosZ genes cluster with P. stutzeri Zobell (Fig. 3).

Figure 3.

Phylogenetic tree of ten nosZ genes from different species of denitrifiers. The tree was reconstructed on the Genetic Data Environment [21] using the neighbor-joining distance method based on the alignment of 1370 bp of the nosZ gene fragments. Accession numbers for the nosZ gene sequences are as follows: Achromobacter cycloclastes (X94977), Alcaligenes eutrophus (X65278), Paracoccus denitrificans (X74792), Pseudomonas stutzeri Zobell (X65277), and Rhizobium meliloti (U47133).

3.3Amino acid analysis

Alignment of the derived amino acid sequences of the ten nosZ genes indicates several highly conserved regions, some of which correspond to the location of the nosZ gene-specific PCR primers (Fig. 4). The overall conservation of amino acid residues is roughly 30%, similar to the amount of conservation at the nucleotide level.

Figure 4.

Figure 4.

Alignment of the derived amino acid sequence from ten nosZ genes. Shading indicates a conserved amino acid across all ten protein sequences. Conserved histidine residues are highlighted by the darker shading. Locations of the PCR priming sites are indicated by the boxed regions, and their direction is given by arrows. The four letter sequence abbreviations are the same as in Fig. 2.

Figure 4.

Figure 4.

Alignment of the derived amino acid sequence from ten nosZ genes. Shading indicates a conserved amino acid across all ten protein sequences. Conserved histidine residues are highlighted by the darker shading. Locations of the PCR priming sites are indicated by the boxed regions, and their direction is given by arrows. The four letter sequence abbreviations are the same as in Fig. 2.

3.4PCR amplification with redesigned nosZ primers

The redesigned primer set Nos661F/Nos1773R successfully amplified the appropriately sized nosZ gene fragment from all environmental samples tested (Fig. 1, lanes J–L). The new primer pairs Nos661F/Nos1527R, and Nos1527F/Nos1773R also achieved successful amplification (data not shown). The redesigned primers produced significantly more robust amplifications than the original Nos661F/Nos2230R pair, for all time points tested (Fig. 1).


The ability to denitrify has been documented in a diverse group of prokaryotes including the α, β, γ Proteobacteria, the Gram-positives, the Cytophaga/Flavobacter, and the Archaea [6]. Our initial denitrifier-specific primer set, Nos661F/Nos2230R, was based on the three nosZ gene sequences available in GenBank in 1995 (all members of the Proteobacteria). Using this primer set, we amplified and cloned the nosZ gene from DNA extracted from the 11/95 continental shelf sediment sample.

The full alignment of all ten nosZ sequences known to date (Fig. 2) indicates that the Nos661F priming site remains relatively robust, with most of the nucleotides conserved. A few ambiguities are located in the middle of the primer, and do not appear to adversely affect its ability to amplify from all of our sediment samples at LEO-15 (Fig. 1, lanes J–L). The Nos2230R priming site did not prove to be as robust. The ambiguities at the nucleotide level are spread throughout the priming site. We therefore conclude that the inability of the original primer set to amplify all time points is based on the general lack of conservation of the Nos2230R primer.

Previous research described a set of nosZ-specific PCR primers for amplifying and labeling a fragment of the nosZ gene of P. stutzeri Zobell[17]. These primers were also based on limited sequence data, and when compared to the full alignment of all ten nosZ sequences, are found to be located in variable regions (amino acids 126–136 and 317–328 in Fig. 4). This set would have the same limitations as our initial primers.

The inclusion of seven additional nosZ gene sequences into the PCR primer design process led to the development of primers demonstrably better than the original. The Nos661F/Nos1773R and Nos661F/Nos1527R primer sets provided robust amplification and yielded a large part of the nosZ gene (1.1 kb and 0.9 kb respectively), including several regions of high variability (see Fig. 2). Either priming set can be used to easily differentiate between nosZ variants (e.g. via RFLP analyses) in natural samples. An added benefit of the Nos661F/Nos1527R set is that its relatively modest size can facilitate double-stranded sequence coverage, using modern sequencing equipment. The Nos1527F/Nos1773R primer set amplifies a ∼250 bp fragment containing sufficient variability (∼60% of the bases are variable in this region) and of an appropriate size to be useful for denaturing gradient gel electrophoresis (DGGE) analyses, which can be used to estimate the microbial diversity of environmental samples [18].

Of the ten nosZ gene sequences used in the PCR primer redesign, seven are known members of the Proteobacteria. Fig. 3 indicates that the marine clones may also belong to the Proteobacteria. Therefore, although the improved nosZ primers amplify very well, they are probably based solely on sequence data from members of the Proteobacteria, which may result in biased PCR amplification. While most currently known denitrifiers are members of the Proteobacteria [6], denitrifying members of the Gram-positive bacteria (such as the Bacillus spp.) are also numerically abundant. Therefore, the inclusion of Bacillus spp. and Cytophaga/Flavobacter representatives into subsequent nosZ gene PCR primer design is warranted.

The derived amino acid sequence alignment of the ten nosZ genes in this analysis reveals conserved residues important to the structure and function of the nitrous oxide reductase protein. Eleven conserved histidine residues have been identified between the nosZ genes of Paracoccus denitrificans, A. eutrophus, P. stutzeri and Pseudomonas aeruginosa[7]. These residues remain conserved with the inclusion of six additional nosZ gene sequences (Fig. 4). Our data support the implication of several of these histidine residues in copper binding by the mature protein [9].

Finally, a surprising result of this study was the near identity of the nosZ gene sequences from Pseudomonas denitrificans and T. pantotropha. Our bacterial stocks were examined and found to be free of contamination. In addition, partial sequencing of the 16S rRNA gene of both isolates indicated the appropriate small ribosomal subunit for each strain (data not shown). Prior research has suggested that T. pantotropha is actually Paracoccus denitrificans, based on the similarity of their 16S rRNA sequences [19]. However, we found the nosZ gene sequences of these two organisms to be only 63% similar. Other researchers [9] also found sequence differences between the nosZ gene of Paracoccus denitrificans and T. pantotropha based on N-terminal amino acid sequence analysis. Additionally, differences in the cytochrome c-550 gene between these two microbes has been reported [20]. This information implies that T. pantotropha has acquired either a 16S rRNA gene from Paracoccus denitrificans or eliminated a functional nosZ gene (if it is Paracoccus denitrificans) in favor of a nosZ gene from Pseudomonas denitrificans.

In conclusion, our study of denitrifiers in the Mid-Atlantic Bight sediments has revealed the presence of nosZ gene sequences substantially different from those present in GenBank. Given the generally low percentage of cultivable marine microbes, it is not surprising to discover that the nosZ gene sequences present in the environment are not closely related to those in culture collections. The fact that the isolated sequences are at most only 63% related to known denitrifier nosZ sequences, indicates a substantial in situ diversity of denitrifiers, not reflected in the cultivable fraction of the population. The newly designed nosZ gene-specific PCR primer sets have proven to be capable of tracking a larger subset of the denitrifier population. The further application of these new primers should serve to enhance our knowledge of the microbial ecology of denitrifying bacteria.


This work was supported in part from funds from the Institute of Marine and Coastal Sciences and NOAA grant #NA46RU0149 to L.J.K. and Sybil P. Seitzinger. The authors thank Rose Petrecca, Robert DeKorsey, and the IMCS divers for their assistance in obtaining sediment samples and SPS, Mary Voytek, and an anonymous reviewer for helpful insights/comments.