Antonie van Leeuwenhoek

, Volume 99, Issue 4, pp 837–844

Rapid detection and identification of the free-living nitrogen fixing genus Azospirillum by 16S rRNA-gene-targeted genus-specific primers


  • Shih-Yao Lin
    • Department of Soil and Environmental SciencesCollege of Agriculture and Natural Resources, National Chung Hsing University
  • Fo-Ting Shen
    • Department of Soil and Environmental SciencesCollege of Agriculture and Natural Resources, National Chung Hsing University
    • Department of Soil and Environmental SciencesCollege of Agriculture and Natural Resources, National Chung Hsing University
Original Paper

DOI: 10.1007/s10482-011-9558-1

Cite this article as:
Lin, S., Shen, F. & Young, C. Antonie van Leeuwenhoek (2011) 99: 837. doi:10.1007/s10482-011-9558-1


The modern agricultural practice utilizing plant growth promoting rhizobacteria (PGPR) has brought great benefits in the promotion of crop growth. Among PGPR, Azospirillum is considered as an important genus which is not only closely-associated with plants but also shows potential in the degradation of organic contaminants. However, lack of media for selective isolation or techniques for specific detection or identification limit the exploration of these rhizobacteria. This motivated us to design a genus-specific oligonucleotide primer pair which could assist in rapid detection of species of the genus Azospirillum by means of PCR-specific amplification. The sensitivity and specificity of the newly designed primer pair Azo494-F/Azo756-R were tested against 12 Azospirillum type strains and other closely-related genera. The Azospirillum-specific 16S rRNA gene fragment (263 bp) was successfully amplified for all the reference Azospirillum species with the designed primer pair. No amplification was noted for closely-related species from other genera. The genus specificity was validated with 18 strains including environmental isolates. Interestingly, two strains assigned earlier as Azospirillum amazonense (DSM 2787T) and Azospirillum irakense (DSM 11586T) failed to produce an Azospirillum-specific fragment with this primer pair. Further phylogenetic analysis of these two isolates based on 16S rRNA gene sequences shows that these two strains might belong to other genera rather than Azospirillum. Preliminary screening of isolates and soil samples with the Azospirillum-specific primers was successful in terms of the rapid detection of Azospirillum isolates. By using real-time PCR analysis the minimum limit of Azospirillum detection was 102 CFU g−1 in the seeded soil sample. The newly designed primers can be used to study the diversity of Azospirillum in ecosystems and aid in the exploration of novel species.


AzospirillumGenus specific primer16S rRNA genePhylogenetic analysisPlant growth promoting rhizobacteria


Agricultural manipulation by using plant growth promoting rhizobacteria (PGPR) has become more practical in many countries (Okon and Vanderleyden 1997; Okon and Itzigsohn 1992). Among the free-living nitrogen-fixing PGPR, Azospirillum strains have been well recognized as biofertilizers owing to their plant growth promoting activities such as nitrogen fixation, phosphate solubilization, and production of the phytohormones auxin, cytokinin, zeatin, regulatory substances or siderophores (Bashan et al. 2004; Hartmann and Baldani 2003; Saxena et al. 1986; Seshadri et al. 2000; Steenhoudt and Vanderleyden 2000; Thuler et al. 2003; Tien et al. 1979). Besides, the genus Azospirillum also shows versatile C- and N-metabolism and better colonizing capability in the rhizosphere (Steenhoudt and Vanderleyden 2000). Until now a total of 15 species has been described in the genus Azospirillum: A. amazonense (Falk et al. 1985), A. brasilense (Helsel et al. 2006), A. canadense (Mehnaz et al. 2007a), A. doebereinerae (Eckert et al. 2001), A. halopraeferens (Reinhold et al. 1987), A. irakense (Khammas et al. 1989), A. largimobile (Ben Dekhil et al. 1997), A. lipoferum (Tarrand et al. 1978), A. melinis (Peng et al. 2006), A. oryzae (Xie and Yokota 2005), A. palatum (Zhou et al. 2009), A. picis (Lin et al. 2009), A. rugosum (Young et al. 2008), A. thiophilum (Lavrinenko et al. 2010) and A. zeae (Mehnaz et al. 2007b). They are distributed mainly from soils and frequently associated with grasses, cereals and crops (Kirchhof et al. 1997). Some species were isolated from oil-contaminated soil and discarded road tar (Lin et al. 2009; Young et al. 2008).

The introduction of rRNA-targeted oligonucleotide probes is a milestone for microbial ecological studies (Stahl et al. 1988). Molecular markers such as 16S, 23S rRNA sequences or intergenic sequences are useful for distinguishing microorganisms among various genera, species and even to differentiate among strains (Woese et al. 1990; Shen and Young 2005). By using the PCR-based techniques the detection, identification and quantification of microorganisms can be performed and complemented with culture-dependent and biochemical methods. The development of Azospirillum species-specific probes labeled with fluorescence has been worked out by Stoffels et al. (2001) and used in fluorescence in situ hybridization analysis. However, this method does only detect physiologically active bacteria and fails to detect all members belong to the genus Azospirillum in the sample. Here we have attempted to design oligonucleotide primers specific for Azospirillum based on the conserved sequence of Azospirillum 16S rRNA gene. By conducting PCR amplification the robustness of the primer set proved to rapidly identify members of this genus from either pure isolates or from soil samples.

Materials and methods

Bacterial strains, soil samples and DNA extraction

A total of 12 reference strains of Azospirillum and 4 closely-related species were purchased from Bioresource Collection and Research Center (BCRC), Taiwan. We are not able to obtain A. largimobile ACM 2041T either from the author deriving this novel species (Ben Dekhil et al. 1997) or the collection center where the type strain was deposited. The culture condition for all these strains were followed by the recommendations from collection center. These bacteria can be cultivated on nitrogen free agar, nutrient agar or R2A agar plate (Lin et al. 2009). Strains of Azospirillum were isolated from agricultural soil and rotten wood. The tested soils for the detection of Azospirillum were sampled from the heavy oil contaminated sites near the oil refinery located in Kaohsiung County, Taiwan. Besides, a non-sterile clay loam soil sample was seeded with strain A. picis DSM 19922T, and the inoculated cell number ranged from 1.5 × 102 to 1.5 × 107 CFU g−1 soil. These soils containing different cell number of strain DSM 19922T was used to obtain the minimum detection limit under the given PCR condition. All the reference strains and environmental isolates used in the present study were listed in Table S2. Bacterial genomic DNA was isolated using UltraCleanTM Microbial Genomic DNA Isolation Kits (MO BIO, USA) according manufacturer’s instructions. DNA from various soil samples (500 mg) was extracted using a bead beating method with the PowerSoilTM DNA Isolation Kits (MO BIO, USA). The quality of the extracted DNA was checked by agarose gel electrophoresis [0.8% (w/v)] after staining with ethidium bromide.

Sequencing of 16S rRNA gene and phylogenetic analysis

Owing to many ambiguous sequences of the Azospirillum type strain deposited in NCBI GenBank, the 16S rRNA gene of various Azospirillum species were re-sequenced and re-submitted. 16S rRNA gene (1,532 bp) was amplified with bacterial universal primers 1F and 9R (Young et al. 2005) in a GeneAmp System 9700 thermal cycler (Applied Biosystem, USA). 16S rRNA gene cycle sequencing was performed using the Bigdye terminator kit (Heiner et al. 1998) and determination of the nucleotide sequence by genetic analyzer (ABI PRISM 310, Applied Biosystems, CA, USA) (Watts and MacBeath 2001). Distances and clustering with the neighbor-joining method were performed using the software package MEGA (Molecular Evolutionary Genetics Analysis) version 4 (Tamura et al. 2007). Bootstrap values based on 1,000 replications are listed as percentages at the branching points and the phylogenetic tree was reconstructed.

Design of the specific primers for the genus Azospirillum

The 16S rRNA gene sequences of 13 Azospirillum type strains and 19 closely-related genera belonging to the family Rhodospirillaceae were obtained from NCBI GenBank and used for primer design. After multiple sequence alignment by using the CLUSTAL_X (1.83) program (Thompson et al. 1997), a primer pair namely Azo494-F/Azo756-R with sequences flanking to the conserved regions in Azospirillum 16S rRNA gene was designed. The forward primer Azo494-F, 5′-GGC CYG WTY AGT CAG RAG TG-3′ (corresponding to 494–513 in A. picis IMMIB TAR-3T) and the reverse primer Azo756-R, 5′-AAG TGC ATG CAC CCC RRC GTC TAG C-3′ (corresponding to 732–756 in A. picis IMMIB TAR-3T) were used to amplify a fragment 263 bp in length. The genus specificity of the designed primer Azo494-F/Azo756-R was tested against all the DNA sequences available in GenBank.

PCR amplification of genomic DNA and soil DNA with primers specific for the genus Azospirillum

DNA extracted by the above methods was used as template for PCR amplification. The PCR reactions were performed in a final volume of 25 μl containing 0.2 mM each of the four dNTPs, 20 pmol of each primer, 3 μl extracted DNA and 2 units of Taq DNA polymerase with appropriate reaction buffer. All the amplifications were performed in a Veriti 96 well thermal cycler (Applied Biosystem, USA). Cycling conditions for primers Azo494-F/Azo756-R were: initial denaturation for 5 min at 94°C followed by 35 cycles of 1 min at 94°C, 1.5 min at 68°C and 0.5 min at 72°C, with a final extension of 7 min at 72°C. Amplification products were separated on 1% agarose gels and stained with ethidium bromide.

In the screening of Azospirillum isolates, the colony PCR was adopted. Briefly, colonies on nitrogen free agar plate (Reinhold et al. 1987) were picked and suspended in PBS buffer (pH 6.8). After treating with 200 μl lysozyme buffer (100 mM Tris, 50 mM EDTA, pH 8.0; lysozyme final concentration is 4 mg ml−1) at 37°C for 30 min, colony PCR was carried out by using 3 μl of bacterial suspensions as templates.

The DNA extracted from heavy oil contaminated soil or seeded soil was used as template for the detection of Azospirillum by using PCR-DGGE analysis. The soil DNAs were diluted 5–10 folds and used as templates. The forward primer used in the PCR-DGGE is clamped with 40 mer GC at the 5′ end (5′-CGC CCG CCG CGC GCG GCG GGC GGG GCG GGG GCG CGG GGG G-3′). PCR was conducted using an annealing temperature of 68°C in order to improve the detection of this group from environmental samples and to minimize sample contaminant interference.

DGGE analysis of the genus Azospirillum

The PCR amplicons obtained from the above reactions were purified from agarose gel using the QIA quick gel extraction kit (Qiagen, Inc., Chatsworth, CA, USA) and further used for DGGE analysis. The PCR-generated amplicons were separated on a DCode universal mutation detection system (Bio-Rad Laboratories Inc., USA); 10% (w/v) polyacrylamide gel with a 40–60% parallel denaturing gradient was prepared with a Hoefer SG100 gradient maker. Denaturant (100%) contained 7 mol/L urea and 40% (v/v) deionized formamide). Approximately 25 μl of PCR products of the expected size was loaded in each well. The gel was run for 17 h at 60 V in 1× TAE buffer (20 mM Tris, 10 mM acetic acid, 0.5 mM EDTA, pH 8.0) at 60°C. Selected bands were cut from the gel and recovered by using electro-elutor and sent for sequencing. The sequence similarities of these isolates to their closely-related neighbors were obtained after using BLAST program ( The image of the gel was obtained with a Kodak imaging system attached to a Kodak DC 290 zoom digital camera and analyzed with Kodak 1D image capture software.

Quantitative analysis of the genus Azospirillum by real-time PCR (qPCR)

Strain Azospirillum sp. CC-Nfb-7 was first incubated in R2A broth for 48 h. Bacterial culture with known cell number (6.6 × 106 CFU μl−1) was seeded into the non-sterile clay loam soil, mixed thoroughly and further used for DNA extraction. Soil DNA was extracted using the bead beating method with the PowerSoilTM DNA Isolation Kits (MO BIO, USA). The DNAs after serially diluted were used as templates in real-time PCR detection, meanwhile DNA concentrations were determined by using spectrophotometer and OD260 value was calculated to obtain the actual DNA concentration. The PCR reactions were performed in a final volume of 25 μl containing 20 pmol of each designed genus-specific primer, 1 μl extracted soil DNA and 1× GoTaq® qPCR Master Mix with the SYBR Green as fluorescence dye. Real-time PCR reaction was performed in a smart cycler® system II (TaKaRa Bio Inc., Japan) with the following cycling conditions: hot-start activation for 2 min at 95°C followed by denaturation and annealing/extension 40 cycles of 15 s at 95°C, and 1 min at 68°C, with a final dissociation from 60 to 95°C. The determination of the detection limit was based on the appearance of Ct (cycle threshold) value and recognizable linear correlation between all the data (Ct value) obtained. All the tests were done in triplicate.

Nucleotide sequence accession number

In the present study the newly obtained 16S rRNA gene sequences were used to substitute the ambiguous 16S rRNA gene sequences (with many unidentified bases) of A.amazonense DSM 2787T (X79735), A.canadense LMG 23617T (DQ393891), A. halopraeferens DSM 3675T (Z29618), A. irakense DSM 11586T (Z29583), A. lipoferum DSM 1691T (M59061) present in the GenBank database. The 16S rRNA gene sequences obtained here were deposited in GenBank under accession number GU256437-GU256444, HM636056, and HQ189391.

Results and discussion

Development and assessment of the primers specific for Azospirillum

In the present work a primer pair namely Azo494-F/Azo756-R was designed based on the conserved regions of the Azospirillum 16S rRNA genes. The regions containing Azospirillum-specific signature nucleotides were selected to design the genus-specific primers (Fig. 1). Sequences of the forward primer Azo494-F and reverse primer totally contained 45 nucleotides. When both these sequences were blasted in NCBI GenBank, a total of 140 identical sequence matches were found. Out of these 140 sequences, 125 belonged to the genus Azospirillum. The remaining 15 sequences were mainly from the strains with invalid names, which were ignored here for comparison. The blast results signifying the specificity of primer pair Azo494-F/Azo756-R for the recognition of Azospirillum members only. When considered for the strains with invalid names they also showed higher 16S rRNA gene sequence similarities with the type strain of Azospirillum, ranging from 94.3 to 99.7% (Table S1). These values were higher than the 16S rRNA gene sequence similarity (86.7–98.9%) at the interspecies level of Azospirillum.
Fig. 1

The highly conserved regions of 16S rRNA gene from Azospirillum spp. were selected and shown, for which a forward primer (Azo494-F) and a reverse primer (Azo756-R) were designed

Sensitivity and specificity of the Azospirillum specific primers

The effectiveness of the primers was tested on 14 Azospirillum strains representing 12 species and other closely-related species, which include R. pekingensis JCM 11669T, R. centenum JCM 21060T, S. aerolata DSM 18479T and I. limosus LMG 20952T (Table S2). PCR amplification was carried out at various annealing temperatures for the optimization of the condition. At an annealing temperature of 68°C, 10 type strains of Azospirillum showed a prominent band (263 bp) (Fig. 2). When the genomic DNAs of non-Azospirillum species were used, there was no amplification of this Azospirillum specific fragment (Fig. 3). This primer pair was able to distinguish Azospirillum strains from other genera to a very clear degree when 68°C was used as annealing temperature. In these closely-related species, the number of the different nucleotides flanking to the primer sequence ranges is 11 (I. limosus LMG 20952T), 11 (R. centenum JCM 21060T), 10 (R. pekingensis JCM 11669T), and 12 (S. aerolata DSM 18479T), signifying the differential ability of this newly designed primer pair (Fig. 1).
Fig. 2

Gel electrophoresis of the Azospirillum specific fragments after PCR amplification. M 100 bp DNA ladder marker; 1 A. picis DSM 19922T; 2A. rugosum DSM 19657T; 3 A. canadense LMG 23617T; 4 A. doebereinerae DSM 13131T; 5 A. irakense DSM 11586T; 6 A. zeae LMG 23989T; 7A. brasilense DSM 1690T; 8A. melinis LMG 24250T; 9A. halopraeferens DSM 3675T; 10A. oryzae JCM 21588T; 11A. lipoferum DSM 1691T; 12 A. amazonense DSM 2787T; BK negative control
Fig. 3

Gel electrophoresis of the Azospirillum specific fragments after PCR amplification. M 100 bp DNA ladder marker; 1Azospirillum sp. CC-Nfb-7; 2Azospirillum sp. CC-RW20-51; 3Skermanella aerolata DSM 18479T; 4Inquilinus limosus LMG 20952T; 5Rhodocista pekingensis JCM 11669T; 6Rhodocista centenum JCM 21060T; BK negative control

In the present study, the type strains of A. amazonense DSM 2787T and A. irakense DSM 11586T failed to give a positive amplification of the Azospirillum specific fragment. By comparing their 16S rRNA gene sequences with Azospirillum and other genera, it showed that these two strains share higher sequence similarity with the genus Rhodocista (Table 1). From the phylogenetic analysis based on 16S rRNA gene sequence these two strains cluster closely with Rhodocista (Fig. 4), which are in consistence with the results published by Stoffels et al. (2001). We have proposed that A. amazonense DSM 2787T and A. irakense DSM 11586T might belong to other genus rather than Azospirillum. The primer pair designed here has been proved to differentiate Azospirillum from other closely-related genera, and was further used in rapid detection and identification of isolates. More than 50 isolates were obtained from the nitrogen-free agar plate and screened for Azospirillum specific fragment by PCR. Among these isolates, only strain CC-Nfb-7 and CC-RW20-51 showed positive reaction in the amplification of Azospirillum specific fragment. Based on their 16S rRNA gene sequences these two isolates were assigned to the genus Azospirillum, and their closely-related neighbor was A.brasilense DSM 1690T, with sequence similarities 97.4 and 97.1%, respectively (Fig. 4).
Table 1

16S rRNA gene sequence similarities between A. amazonense DSM 2787T, A. irakense DSM 11586T and other closely-related genera


Azospirillum (13)a

Inquilinus (1)

Rhodocista (2)

Skermanella (2)

A. amazonense DSM 2787T



91, 93

89, 93

A. irakense DSM 11586T



95, 97

89, 91

aThe value in the parenthesis represents the number of species belonging to the respective genus
Fig. 4

Phylogenetic analysis of Azospirillum strains and other closely-related species based on 16S rRNA gene sequences. Distances and clustering were performed by using neighbor-joining method with the software package MEGA version 4. Bootstrap values based on 1,000 replications are listed as percentages at the branching points

Detection of Azospirillum by using PCR-DGGE and qPCR analysis

The denaturing gradient gel electrophoresis (DGGE) of the PCR amplified fragments was conducted to detect Azospirillum species either from pure cultures or soil samples. Genomic DNA isolated from pure culture of 11 Azospirillum species was used as template for the amplification of Azospirillum specific fragments. The DGGE condition was optimized by using the PCR amplicons from type strains of Azospirillum. By recognizing the position of the amplicon derived from 11 Azospirillum species they can be grouped into 4 clusters (Figure S1).

The specific PCR amplification along with DGGE was also used in the detection of Azospirillum from soil samples. DNA extracted from heavy oil contaminated soil gave positive amplification of Azospirillum specific fragments, indicating that members of this genus present in the soil samples. After sequencing the bands in denaturing gradient gel the sequences were assigned to Azospirillum species, with the similarity of 99.5% to A. oryzae JCM 21588T. The same condition was also used in the detection of Azospirillum in the seeded soil, with a non-seeded soil as control. Soil DNA extracted from control treatment gave no amplification of the PCR amplicon, demonstrated that there was no detectable Azospirillum strains in the original soil. In the seeded soil sample the minimum limit of Azospirillum detection was 1 × 104 CFU g−1.

In the real-time PCR detection, the sensitivity was much higher than that in the regular PCR amplification. The primer set Azo494-F/Azo756-R was successfully used in the detection of the inoculant (Azospirillum sp. CC-Nfb-7) after real-time PCR analysis was conducted, and the detection limit was 2.7 pg μl−1, corresponds to cell number 6.6 × 102 CFU g−1 in the seeded soil sample. The qPCR can be used to detect Azospirillum species when the cell number is low in the soil samples, providing higher resolution in the detection of this specific genus from soil or aquatic environments (Figure S2, S3).


The Azospirillum-specific primer pair Azo494-F/Azo756-R developed in the present study was successful in differentiating the genus Azospirillum from other genera under the given set of PCR conditions. The primer set, which amplifies a 263 bp fragment of the 16S rRNA gene of Azospirillum species, can also be used to detect the presence of Azospirillum species in soil samples either in DGGE or qPCR analyses. This genus-specific PCR-based technique will aid in the rapid detection and identification of members of the genus Azospirillum from pure cultures or soil samples, and can be used in the studies of the diversity of Azospirillum or exploration of novel species in the ecosystem.


This research work was kindly supported by grants from National Science Council, Council of Agriculture, Executive Yuan and in part by the Ministry of Education, Taiwan, R.O.C under the ATU plan.

Supplementary material

10482_2011_9558_MOESM1_ESM.doc (211 kb)
Supplementary material 1 (DOC 210 kb)

Copyright information

© Springer Science+Business Media B.V. 2011