Diversity Analysis of Diazotrophic Bacteria Associated with the Roots of Tea (Camellia Sinensis (L.) O

Home , Azospira, Azospira oryzae, Azospira restricta, Delftia lacustris, Diazotroph, Herbaspirillum

J. Microbiol. Biotechnol. (2011), 21(6), 545–555 doi: 10.4014/jmb.1012.12022 First published online 18 April 2011

Diversity Analysis of Diazotrophic Bacteria Associated with the Roots of Tea (Camellia sinensis (L.) O. Kuntze)

Arvind, Gulati1*, Swati Sood1, Praveen Rahi1, Rishu Thakur1, Sunita Chauhan1,2, and Isha Chawla nee Chadha1

1Plant Pathology and Microbiology Laboratory, Hill Area Tea Science Division, Institute of Himalayan Bioresource Technology, Council of Scientific and Industrial Research, P.O. Box 6, Palampur (H.P.)-176 061, India 2Department of Microbiology, College of Basic Sciences, CSK Himachal Pradesh Krishi Vishvavidyalaya, Palampur (H.P.)-176 062, India Received: December 16, 2010 / Revised: February 27, 2011 / Accepted: March 21, 2011

The diversity elucidation by amplified ribosomal DNA of atmospheric N into utilizable forms by plants and restriction analysis and 16S rDNA sequencing of 96 indirectly through the secretion of phytohormones that associative diazotrophs, isolated from the feeder roots of enhance plant growth [53]. Nitrogen-fixing ability is widely tea on enriched nitrogen-free semisolid media, revealed distributed among phylogenetically diverse bacteria such as the predominance of Gram-positive over Gram-negative Acetobacter, Arthrobacter, Azoarcus, Azospirillum, Azotobacter, bacteria within the Kangra valley in Himachal Pradesh, Bacillus, Burkholderia, Enterobacter, Herbaspirillum, India. The Gram-positive bacteria observed belong to two Klebsiella, and Pseudomonas associated with some taxonomic groupings; Firmicutes, including the genera agronomically important crops [11, 13, 15, 23, 47, 61]. Bacillus and Paenibacillus; and Actinobacteria, represented Diazotrophic bacteria are considered to be possible by the genus Microbacterium. The Gram-negative bacteria alternatives to inorganic nitrogen fertilizers for promoting included α-Proteobacteria genera Brevundimonas, Rhizobium, crop growth and yield [25, 28, 51]. and Mesorhizobium; γ-Proteobacteria genera Pseudomonas The sustained productivity of the world’s most popular and Stenotrophomonas; and β-Proteobacteria genera Azospira, non-alcoholic beverage, tea, relies heavily upon the use of Burkholderia, Delftia, Herbaspirillum and Ralstonia. The nitrogen [37, 50]. Leaching, the harvesting of shoots, and low level of similarity of two isolates, with the type strains the skiffing and pruning of bushes as an essential husbandry Paenibacillus xinjiangensis and Mesorhizobium albiziae, practice, are the major limiting factors for the maintenance suggests the possibility of raising species novum. The of nitrogen levels suitable for the sustainability of tea bacterial strains of different phylogenetic groups exhibited productivity [38]. The intensive use of chemical fertilizers distinct carbon-source utilization patterns and fatty acid creates a highly selective environment and adversely affects methyl ester profiles. The strains differed in their nitrogenase microbial diversity. This phenomenon is even more pronounced activities with relatively high activity seen in the Gram- in the monocultural conditions of tea plantations, where the negative strains exhibiting the highest similarity to Azospira perennial nature of bushes also exerts a strong rhizosphere oryzae, Delftia lacustris and Herbaspirillum huttiense. effect on the soil microflora [20, 29, 39]. The introduction Keywords: Camellia sinensis, diazotrophic bacteria, acetylene of non-resident microorganisms into such environments is reduction assay, 16S rDNA sequencing, FAME analysis, problematic and necessitates the selection of indigenous carbon-source utilization microorganisms well-adapted to the environmental conditions [49, 55]. Information about diazotrophs in tea is limited to knowledge of the presence of Azospirillum, Azotobacter, Achromobacter, Bacillus, Mycobacterium, Clostridium, and Nitrogen (N), considered an essential macronutrient for plant Beijerinkia, reported in the Republic of Georgia [34]. The growth, is often a yield-limiting factor in crop productivity aim of the present study is to broaden the current knowledge owing to the loss of a major component of mineral nitrogen base through an assessment of the genetic diversity of from the soil through run-off and leaching [8]. Diazotrophic associative tea diazotrophs in the Kangra valley, with a bacteria influence plant growth directly through the fixation view towards developing eco-friendly bioformulations based on the native microorganisms. The Kangra valley’s century- *Corresponding author Phone: +91-1894-230411; Fax: +91-1894-230433; old tea plantations, with bushes of nearly 60 years average E-mail: [email protected]; [email protected] age, cover about 2,300 hectares and are located between 546 Arvind et al.

32.03o and 32.20oN and 76.37o and 76.80oE at 1,000- water. The restriction mixtures were incubated for 2 h with AluI, o o 1,700 m above mean sea level in the mid-hills of the HpaII, and HinfI at 37 C and at 65 C with TaqI. The reaction Western Himalayas in Himachal Pradesh in India. The products of ARDRA resolved on 2% (w/v) agarose gel, prepared in regional soil has been characterized as acidic with a low 1× TBE buffer, were stained with ethidium bromide (10 µg/ml). The availability of nutrients [54]. The valley endures somewhat banding patterns were visualized and photographed with Digidoc (BioRad, CA, USA) under UV light. The band sizes were estimated hot and humid summers, with a mean temperature maximum through a comparison with a 100 bp DNA ladder (Fermentas, Vilnius, of 36oC and a minimum of 16.5oC, and cool winters, with a o o Lithuania). Fingerprints generated by restriction analysis were recorded mean temperature maximum of 20 C and a minimum of 0 C. in a binary form, where 1=presence of a band and 0=absence of the The annual potential transpiratory demand of 1,100 mm is band. The dissimilarity dendrogram was generated using the hierarchic met fully by an annual mean rainfall of 2,500 mm [57]. Unweighted Pair Group Means Average (UPGMA) method employing TREECON software ver. 1.3b [59].

MATERIALS AND METHODS 16S rDNA Sequence Analysis The representative strains of various rDNA types generated by Sample Collection ARDRA were subjected to sequence elucidation of 16S rDNA. The The feeder roots of tea bushes under cultivation were sampled from DNA from the gel slice was eluted by employing a Quick Gel 17 different locations: Baijnath, Banoi, Ballah, Bir, Choubin, Chauntra, Extraction Kit (Invitrogen, CA, USA). The gel-purified 16S rDNA Drang, Gopalpur, Joginder Nagar, Kathak, Pathiar, Palampur, Sakri, was ligated into the pGEM-T easy vector and transformed into Rajpur, Sungal, Yol and 61 Miles, which cover the various tea chemically competent Escherichia coli strain DH5-α cells. Plasmid growing zones of the Kangra valley. The soil was dug to a 30 cm DNA for sequencing was purified by using a Plasmid GenElute depth and young feeder roots were identified by tracing them through Miniprep Kit (Sigma-Aldrich Corp, MO, USA). The quality of the their main roots to their respective bushes. Fifty-one samples of purified plasmid DNA was checked on 1% (w/v) agarose gel these young feeder roots were then collected, in sum representing prepared in a 1× TAE buffer. The sequences of the inserts were three samples for each location. determined using a Big-Dye Terminator Cycle Sequencer and ABI Prism 310 Genetic Analyzer (Applied Biosystems, CA, USA). The Enrichment and Isolation of Diazotrophs sequencing PCR reaction, of a total volume of 5 µl, included 1 µl of N-free semisolid malate (NFb) enrichment media and Rojo Congo 5× sequencing buffer, 1 µl of Big-Dye Terminator Premix, 1 µl of agar were employed for the isolation of diazotrophs from the root primer (5 pmol), and 2 µl of plasmid. Thermal cycling conditions samples [5, 44]. The feeder roots were washed with sterile water consisted of an initial denaturation at 96oC for 3 min, followed by and treated with 1% chloramine-T for 2 min and again washed with 30 cycles of 94oC for 10 s, 50oC for 40 s, and 60oC for 4 min. The sterile water, cut into 1-2 mm length pieces and placed individually unincorporated dye terminators were removed from the sequencing o into tubes containing a 5 ml NFb medium and incubated at 28±1 C reaction using a Montage SEQ96 Sequencing Reaction clean up kit for 72 h. A loopful of white, dense, undulating and diffuse pellicles, (Millipore Corp, Billerica, MA, USA). The purified sequencing developing 1-4 mm undersurface, were streaked on Rojo Congo products were then transferred into the injection plate for sequence agar and incubated at 28±1oC. The light pink or colorless colonies elucidation. which turned scarlet after 72 h were purified by repeated subculturing Sequences were compared with the GenBank database using the on Rojo Congo agar. Blastn (http://www.ncbi.nlm.nch.gov) facility of the National Center for Biotechnology Information and aligned to the nearest neighbors in 16S rDNA Amplification of Diazotrophic Bacteria ClustalW (http://align.genome.jp/). The evolutionary distances amongst Genomic DNA was extracted using a Qiagen DNeasy Plant Mini the bacterial strains and their related taxa were calculated with a Kit (Qiagen, Valencia, CA, USA). The quality of genomic DNA was TREECON software package, ver. 1.3b [59], using Kimura’s two- checked on 0.75% agarose gel prepared in 1× TAE buffer. Amplification parameter model [24]. The 16S rDNA sequences of the diazotrophs of 16S rDNA was performed using fD1 (5-AGAGTTTGATCCTGG were then submitted to the GenBank (Accession No. in Table 1). CTCAG-3) and rP2 (3-ACGGCTACCTTGTTACGACGT-5) primers [60]. The 50 µl PCR reaction mixture contained 200 µM dNTPs, Carbon-Source Utilization Pattern (BIOLOG) 50 µM of each primer, 1× PCR buffer, 3U Taq DNA polymerase Carbon-source utilization pattern for 95 carbon sources, by the (Invitrogen, CA, USA) and 100 ng of genomic DNA. The thermocycling diazotrophs representing rDNA types, was studied using a BIOLOG procedure involved an initial denaturation at 94oC for 3 min, followed system (Microstation, Microbial Identification System, 1998, Biolog by 35 cycles at 94oC for 1 min, 52oC for 1 min, and 72oC for 2 min, Inc, CA, USA). Cultures were grown on Tripticase soya agar (TSA) and a final extension at 72oC for 8 min. PCR products were analyzed for 48 h at 30±1oC, resuspended in 20 ml of inoculation fluid on 1.2% agarose gel and their sizes estimated using a 1 kb DNA (BIOLOG), and their inoculum density was adjusted to a transmittance ladder (Fermentas, Vilnius, Lithuania). of 52% (Gram-negative non-enteric), 28% (Gram-positive sporulating rods), and 20% (Gram-positive non-sporulating rods and cocci) using Amplified Ribosomal DNA Restriction Analysis (ARDRA) a BIOLOG Turbidimeter. The BIOLOG GN and GP microplates Four tetra-cutter restriction endonucleases, AluI, HpaII, HinfI and were inoculated with 150 µl of the bacterial suspension per well and TaqI, were employed for the restriction analysis of PCR products. incubated at 30±1oC for 24 h. The development of color was read The restriction reaction consisted of 1.5 µl of 10× restriction enzyme on a 595 nm filter in a MicroStation Reader between 6 and 24 h of buffer, 1 U restriction enzyme, 10 µl of PCR product and 18.2 mΩ incubation. The substrate utilization profiles were evaluated with DIVERSITY OF DIAZOTROPHS ASSOCIATED WITH TEA ROOTS 547 ) -1 h -1 mgprotein 4 H 2 Nitrogenase activity (nmol C (%) Similarity b (AJ542508) 99.4 173.32 GH06 (DQ365561) 98.6 115.65

(EU558284) 97.9 243.85 IAM12423 (AB294553) 98.9 190.95

BCRC17757 (EU179327) 99.6 219.17 B538 (AY839868) 96.5 141.29 CCUG57438 (FJ477040) 98.9 49.83 WatG (AB117953) WatG 99.7 243.88 ATCC14670 (NR_024698) ATCC14670 99.7 412.69 LMG23507 (DQ311088) 94.9 74.40 DSM8801 (AM747813) 99.7 198.20 Ab10b LMG 19425 ( NR_028800) 98.6 0 LMG 12614 (AF311970) 99.4 163.31 DSM 11031 (AB021198)DSM 11031 99.7 167.57 X19 LMG24727(DQ358736) 97.7 244.08 NRRL NRS666 (D78473) 99.0 233.82 B8W22 (EF114313) 99.9 121.14 LMG21833 AsdM5-2B (FM955870) 99.7 158.37 IAM13153 (NR_025831) IAM13153 99.2 322.95 WH22 (AJ310083) WH22 99.5 337.39 SEMIA4088 (EU399915) 99.6 198.60 Highest similarity 332 (EU888308) 99.3 536.00 GR-1 (AY277622) 99.7 470.65 ATCC25621 (AY795568)ATCC25621 99.4 222.21 Stenotrophomonas maltophilia Stenotrophomonas Pseudomonas aeruginosa Bacillus altitudinus Bacillus Microbacterium phyllosphaerae Microbacterium tropici Rhizobium Mycoplana bullata albiziaeMesorhizobium Burkholderia cepacia Herbaspirillum huttiense taiwanensisRalstonia Lysinibacillus xylanilyticus Lysinibacillus Paenibacillus xinjiangensis Paenibacillus tundrae Bacillus vallismortisBacillus Azospira oryzae Delftia lacustris Bacillus fusiformisBacillus Paenibacillus taichungensis Paenibacillus sonchi Paenibacillus lautus Bacillus frigoritoleransBacillus Bacillus bataviensisBacillus Bacillus cereus cereus Bacillus Bacillus aryabhattai ) a rDNA type (Accession No. Representative isolate (isolates) XX (1) IHB B 4036 (HM234000) XIII (7)XIII IHB B 4035 (HM233999) XVI (1) IHB B 2267 (HM234006) XIX (1)XIX XXII (2) IHB B 2274 (HM233968) IHB B 2263 (HM233962) XXIII XXIII (1) IHB B 2272 (HQ686041) XXIV (6) IHB B 2268 (HM233964) VII VII (2) (2)VIII IHB B 4019 (GQ411067) IHB B 2257 (HM233959) VI VI (7) IHB B 4033 (HM233997) XII XII (6) IHB B 2276 (HM233969) XV (1) IHB B 4037 (HM234001) V (2) IHB B 4012 (HM233988) IX (1)IX IHB B 2283 (HM233974) X (1) XI (3) IHB B 2284 (HM233975) IHB B 2310 (HQ686042) IV (3)IV IHB B 2287 (HM233977) III (2)III IHB B 2269 (HM233965) II (3)II IHB B 4042 (HM234003) rDNA type rDNA

Taxonomic position and Taxonomic nitrogenase activity of diazotrophic bacteria from tea rhizoplane. position Taxonomic Taxonomic -Proteobacteria (2) XIV IHB B 4026 (HM233995) -Proteobacteria XVIII (2) IHB B 2273 (HM233967) -Proteobacteria XXI (1) IHB B 4040 (HM234002) ActinobacteriaGram-negative XVII (3)α IHB B 4017 (HM233991) γ β Gram-positive Firmicutes I (36) IHB B 4007 (HM233987) Validly published species in NCBI GenBank. in species published Validly NCBI GenBank. b Table 1. a 548 Arvind et al.

MicroLog version 4.2 database software. The reactions were scored as negative (0), positive (1) or borderline (0.5). A dendrogram was constructed using the STATISTICA data analysis software system, ver. 7 (StatSoft Inc., Tulsa, OK, USA).

Whole-Cell Fatty Acid Methyl Ester (FAME) Analysis The composition of the whole-cell fatty acids of the isolates was studied by gas chromatographic analysis using a Sherlock Microbial Identification System (MIDI, Inc., Newark, DE, USA) following the previously described procedure [52]. Cultures streaked on TSA at 30±1oC for 24 h were harvested from the third quadrant. The samples derivatized to methyl esters were subjected to GC analysis using an Ultra 2 phenyl methyl silicone-fused silica capillary column of 25 m×0.2 mm on GC 6890 (Agilent Technologies, CA, USA). Hydrogen was used as the carrier gas, nitrogen as the “make up” gas, and air to support the flame. The GC oven temperature was programmed to increase incrementally from 170 to 270oC at a 5oC rise per minute, with a 2 min hold at 300oC. Fatty acids were identified and quantified by a comparison with retention times and peak areas of calibration standards obtained from the Sherlock Microbial Identification System. The dendrogram was generated on the basis of whole-cell fatty acid composition using MIDI Sherlock analysis software, ver. 6.0.

Acetylene Reduction Assay (ARA) The nitrogenase activity of diazotrophic bacteria was determined by the reduction of acetylene to ethylene [21]. Bacterial cultures grown in 10 ml of Trypticase soya broth (TSB) for 24 h at 28±1oC were then centrifuged for 10 min at 8,000 rpm. The pellets were washed twice with N-free minimal medium, resuspended into 10 ml of NFb-medium and incubated for 48 h at 28±1oC. The cultures were transferred to 60 ml air-tight bottles and 10% of air (v/v) from each bottle was replaced using an air-tight hypodermic syringe with pure acetylene gas (Sigma Gases, New Delhi, India). The bottles were incubated at ambient temperature for 1 h, and 1 ml of gas mixture was assayed for ethylene concentrations using a GC6890 (Agilent Technologies, CA, USA), equipped with an FID detector and a Carboxen-fused silica capillary column, 30 m×0.53 mm (Supelco, PA, USA). Nitrogen was used as the carrier gas at a 5 ml/min flow rate. The GC oven temperature was programmed to increase from 70 to 200oC at 20oC/min with a 2 min hold at 70oC and a 5 min hold at 200oC. The standard curve was prepared using different concentrations of pure ethylene (Sigma Gases, New Delhi, India). The protein content of the cultures was determined after lysing the cells by NaOH-SDS solubilization and the utilization of the Bradford protocol [9, 18]. Different concentrations of bovine serum albumin (BSA) were used to prepare the standard curve.

RESULTS

Isolation of Bacteria A total of 96 diazotrophic bacteria associated with tea roots were isolated on NFb enrichment media. Colonies Fig. 1. UPGMA cluster analysis based on the restriction patterns with different observable morphological features on Rojo of amplified 16S rDNA of diazotrophic bacteria from tea Congo agar were selected, subcultured, purified, stocked rhizoplane. in 25% glycerol and stored at -70oC for further studies. rDNA types at 90% similarity level are indicated in roman numerals. DIVERSITY OF DIAZOTROPHS ASSOCIATED WITH TEA ROOTS 549

ARDRA glycerol, adenosine, methyl pyruvate and L-arabinose were PCR amplification of the 16S rDNA of various isolates the common carbon sources utilized by all Gram-positive resulted in a single fragment of approximately 1.5 kb, bacteria (Supplementary Table 1). All Gram-negative bacteria, corresponding to the predicted size of the amplification excepting IHB B 2273 and IHB B 2263, utilized methyl product of 16S rDNA. The purified PCR products restricted pyruvate, monomethyl succinate, formic acid, β-hydroxy with the endonucleases AluI, HinfI, HpaII and TaqI produced butyric acid, D,L-lactic acid, L-alanine, L-asparagine, L- polymorphic restriction banding patterns (Supplementary aspartic acid, L-glutamic acid, L-histidine and L-proline Fig. 1). AluI and HpaII generated 2-5 bands, whereas TaqI (Supplementary Table 2). and HinfI produced 2-4 bands each. AluI and HpaII were Cluster analysis, based on carbon-source utilization found to be more discriminating with 16S rDNA-amplified patterns, formed two clusters at a 4.0 linkage distance for products of the isolates. ARDRA cluster analysis generated the Gram-positive diazotrophs (Fig. 3): Cluster I contained 24 rDNA types, which included 16 groups with 2-36 4 strains, IHB B 2287, IHB B 4012, IHB B 4042 and IHB isolates under various groupings, and 8 independent branches B 4033 of Bacillus; and Cluster II contained separately in at the 90% similarity level (Fig. 1). two subclusters three strains, IHB B 2269, IHB B 2276 and IHB B 4035 of Bacillus, and five strains, IHB B 2257, 16S rDNA Sequencing IHB B 2283, IHB B 2284, IHB B 2310 and IHB B 4019 of Nearly full-length 16S rDNA sequences were obtained for Paenibacillus. The strain IHB B 4007 of Bacillus and the 24 strains representing their rDNA types. A BLASTn strain IHB B 4017 of Microbacterium stood independently search of these sequences revealed the presence of bacteria outside the two clusters. The Gram-negative diazotrophs belonging to diverse genera showing a >97.5% similarity also grouped at a 4.0 linkage distance into two clusters with their nearest relatives in the NCBI GenBank database (Fig. 3): IHB B 2273, IHB B 2263, IHB B 2274 and IHB (Table 1). The presence of Gram-positive and Gram-negative B 2272 under Cluster I, showing the maximum similarity bacteria was recorded in discrete groups showing a >97.5% with Azospira oryzae, Ralstonia taiwanensis, Herbaspirillum similarity to their nearest relatives in the NCBI database. huttiense and Mesorhizobium albiziae; and IHB B 2267 Only two isolates exhibited a 96.5% and a 94.9% similarity and IHB B 2268 under Cluster II showing the maximum to Paenibacillus xinjiangensis strain B538T and Mesorhizobium similarity with Brevundimonas bullata and Burkholderia albiziae strain LMG23507T, respectively. cepacea. IHB B 4026, IHB B 4037, IHB B 4040 and IHB Phylogenetic analysis based on 16S rDNA sequences of B 4036, with a maximum similarity to Rhizobium tropici, Delftia the representative rDNA types and their closely related lacustris, Pseudomonas aeruginosa and Stenotrophomonas taxa revealed 5 groups representing different phyla: Firmicutes, maltophilia, stood independently outside of the clusters. Actinobacteria, α-Proteobacteria, β-Proteobacteria and γ- Proteobacteria (Fig. 2). The first grouping, under Firmicutes, FAME Analysis was further divided into 2 subgroups, belonging to Bacillus, The diazotrophs differed in the composition of whole-cell with 8 strains, and Paenibacillus, with 5 strains. The strain fatty acids (Supplementary Table 3). Fatty acids observed IHB B 4017 was grouped with the validly published species in the majority of Gram-positive and Gram-negative strains of Microbacterium under Actinobacteria. Three strains, were C14:0, C16:0 iso, C16:0 and C18:0. The Gram- IHB B 4026, IHB B 2272, and IHB B 2267, under α- positive strains also exhibited a predominance of C14:0 Proteobacteria were found to be closely related to iso, C14:0, C15:0 iso, C15:0 anteiso, C16:0 w7c alcohol, Rhizobium multihospitium CCBAU83401T, Mesorhizobium C16:0 iso, C16:1 w11c, C16:0 and C18:0. However, C15:0 albiziae LMG23507T, and Brevundimonas bullata IAM13153T. iso and C15:0 anteiso were the major fatty acids detected γ-Proteobacteria included two strains, IHB B 4040 and for the strains belonging to Bacillus and Paenibacillus. IHB B 4036, closely related to Pseudomonas aeruginosa IHB B 4017 with the maximum similarity to Microbacterium LMG1242T and Stenotrophomonas maltophilia ATCC13637T. phyllosphaerae, mainly displayed the presence of C15:0 Five strains, IHB B 2273, IHB B 4037, IHB B 2268, IHB anteiso, C16:0 iso and C17:0 anteiso. The majority of B 2274, and IHB B 2263, showing close phylogenetic Gram-negative isolates recorded high quantities of C16:0 relationships with Azospira oryzae 6a3T, Delftia tsuruhatensis and unresolved fatty acids, sum in feature 3 (C16:1 w7c/ ATCCBAA554T, Burkholderia arboris LMG24066T, C16:1 w6c), sum in feature 8 (C18:1 w7c) and summed Herbaspirillum putei ATCCBAA806T and Ralstonia oxalatica feature 8 (C18:1 w7c, C18:1 w6c). However, IHB B 4036 DSM1105T were included under β-Proteobacteria. with the maximum similarity to Stenotrophomonas maltophilia, displayed the presence of C15:0 iso, C16:0 and sum in Carbon-Source Utilization Patterns feature 3 (C16:1 w7c/C16:1 w6c) as major fatty acids All Gram-positive and Gram-negative strains were noted exhibiting unusual composition when compared with other to be positive for L-arabinose utilization and negative for diazotrophs (Supplementary Fig. 2). Cluster analysis, based glucose-1-phosphate utilization. D-Ribose, pyruvic acid, on the FAME profiles of diazotrophic bacteria, revealed 2 550 Arvind et al.

Fig. 2. Phylogenetic tree based on 16S rDNA sequences, drawn using the neighbor-joining method with evolutionary distances computed using Kimura’s two-parameter method, showing the relationship of diazotrophic bacteria from tea rhizoplane with validly published sequences of related genera. 16S rDNA accession numbers are within brackets. Bootstrap values are given at nodes. Scale bar is of 0.1 substitutions per nucleotide position. DIVERSITY OF DIAZOTROPHS ASSOCIATED WITH TEA ROOTS 551

Fig. 3. Dendrogram of diazotrophic bacteria representing rDNA types based on BIOLOG carbon-source utilization constructed using STATISTICA data analysis software system, ver. 7. Similarities to the strains are assigned by 16S rDNA sequencing.

-1 -1 major groups at 75 euclideans: Cluster I included all the C2H4 mg protein h . Eight isolates were negative for Gram-positive bacteria in addition to the Gram-negative acetylene reduction. strain IHB B 4036, whereas Cluster II included the other Gram-negative bacterial strains (Fig. 4). DISCUSSION Acetylene Reduction Assay Among the representative strains of rDNA types, the highest ARDRA generation of 24 rDNA types, and 16S rDNA -1 -1 nitrogenase activity of 536.0 nmol C2H4 mg protein h sequences of bacteria isolated on N-free enrichment media, was recorded for IHB B 4037, showing the maximum revealed a high diversity in diazotrophic bacteria associated similarity with Delftia lacustris. A high nitrogenase activity with young roots in tea (Fig. 1). The representative strains was also exhibited by IHB B 2273, IHB B 2274, IHB B of rDNA types belong to 13 diverse genera (Table 1). 4012, and IHB B 2267, with maximum similarities to Phylogenetic analysis confirms the predominance of Azospira oryzae, Herbaspirillum huttiense, Bacillus fusiformis, Gram-positive bacteria, including Bacillus and Paenibacillus, and Brevundimonas bullata, respectively (Table 1). The ARA under the phylum Firmicutes (Fig. 2). Sixty-six bacterial activity for other isolates ranged from 74.4 to 284.6 nmol strains were clustered under 8 rDNA types representing the 552 Arvind et al.

Fig. 4. Dendrogram of diazotrophic bacteria representing rDNA types based on whole-cell fatty acid composition by GC-FAME constructed using MIDI Sherlock analysis software, ver. 6.0. Similarities to the strains are assigned by 16S rDNA sequencing. genus Bacillus. The preponderance of Bacillus corroborates rhizoplane has been reported for its N-fixing ability from earlier reports regarding the microflora dominating the rhizoplane of several crops, but not for tea [4, 10, 14]. rhizosphere of established tea bushes [40, 41, 56]. Several Similarly, β-Proteobacteria genera Azospira, Burkholderia, species of Bacillus and Paenibacillus have previously been Delftia, Herbaspirillum and Ralstonia, isolated from tea reported as diazotrophs in the rhizoplane of different plants rhizoplanes in the present studies have not previously been [1, 2, 17, 46]. The diazotrophic strains of Bacillus but not recorded in tea rhizospheres/rhizoplanes, although the of Paenibacillus have also been reported from tea plants diazotrophic strains of these genera have been reported in [34]. The Gram-positive strain from tea roots exhibiting the rhizosphere soils/rhizoplanes of other plants [19, 33, the closest similarity with Microbacterium phyllosphaerae, 42, 45]. However, Azospirillum, Azotobacter, Achromobacter, under the phylum Actinobacteria recorded in the present Mycobacterium, Clostridium and Beijerinckia not recorded studies, has previously been reported in association with in the Kangra valley have been reported in tea from the Lasiurus sindicus roots [10]. Republic of Georgia [34]. This is possibly due to the The prevalence of β-Proteobacteria was observed amongst differences in the geoclimatic conditions between the two Gram-negative diazotrophic bacteria associated with tea regions, which are known to influence microbial diversity roots (Fig. 2). The strains grouped under α-Proteobacteria [35]. The low similarity of the two bacterial strains recorded, exhibited the highest similarity of 16S rDNA sequences with their most closely affiliated species, Paenibacillus with Rhizobium tropici, Mesorhizobium albiziae and xinjiangensis (96.5%) and Mesorhizobium albiziae (94.9%), Brevundimonas bullata. Species of the genera Rhizobium indicates the possibility of the future discovery of undescribed and Mesorhizobium are well known for N-fixation in non- taxa as species novum amongst the associative diazotrophs legumes [27, 30, 31, 48], whereas Brevundimonas aurantiaca of tea (Table 1). The random distribution and occurrence of has been reported as an endophytic diazotroph from diazotrophic bacterial strains from different locations into cultivated rice [43]. γ-Proteobacteria represented by the various ARDRA groups indicate the lack of location- genera Pseudomonas and Stenotrophomonas from tea specific diversity in the tea plantations of the Kangra valley. DIVERSITY OF DIAZOTROPHS ASSOCIATED WITH TEA ROOTS 553

The representative strains of the ARDRA grouping, In conclusion, tea rhizoplanes exhibit a rich inter- identified as the diverse taxonomic units by 16S rDNA generic and intra-generic diversity of diazotrophic bacteria, sequencing, exhibited a distinctive carbon-source utilization including Azospira, Bacillus, Brevundimonas, Burkholderia, pattern as different phenotypes. The grouping of genetically Delftia, Herbaspirillum, Paenibacillus, Pseudomonas, Rhizobium, related strains in different metabolic clusters indicated that Mesorhizobium, Microbacterium, Stenotrophomonas and carbon-source utilization patterns do not correspond to Ralstonia. The dominance of Bacillus and Paenibacillus phylogenetic relationships. Phenotypic characters are known strains indicates their greater potential for survival, adaptability, to be influenced by genetic as well as environmental factors and application as native diazotrophic strains for the [36]. The utilization of maximum carbon sources by IHB improvement of the availability of nitrogen for tea plants B 2269 and IHB B 2287, showing the highest similarity in the Kangra valley and other tea-growing areas with with Bacillus bataviensis and Bacillus frigoritolerans similar geographic conditions. amongst Gram-positive bacteria, and IHB B 4026, IHB B 2267, IHB B 2268 and IHB B 4036, exhibiting maximum similarity with Rhizobium tropici, Brevundimonas bullata, Acknowledgments Burkholderia arboris and Stenotrophomonas maltophilia amongst Gram-negative bacteria, respectively, indicated The authors would like to acknowledge the generous their high metabolic efficiencies. The utilization of a wide assistance of Dr. P. S. Ahuja, the Director of the Institute of range of carbon sources has been considered as a major Himalayan Bioresource Technology (CSIR), for the provision factor for the successful colonization and attainance of of the laboratory facilities. The Council of Scientific and dominance in the rhizosphere [26]. However, the utilization Industrial Research, Government of India, also deserves of relatively fewer carbon sources, by the representative credit for the financial support provided under the Supra- strains of dominant rDNA types, indicates the contribution institutional Project “High Value Products from Agro- of other factors in the colonization of tea roots by diazotrophic forestry Resources from the Himalayan Region and Improving bacteria (Supplementary Table 1). Productivity and Quality of Product Development, including FAME analysis has revealed a specific fatty acid composition Nutraceuticals/Value-added Products” (Supra-Institutional as the “microbial fingerprint” and the clustering of strains Project SIP-003). Thanks are also due to Mr. Khushal according to the respective individual Gram reactions (Fig. 4). Chand and Mr. V. S. Dhadwal for their hard work and Gram-negative bacteria exhibited the largest quantity of assistance in the sampling process of different genera from C16:0, whereas Gram-positive bacteria showed an abundance a wide variety of locations in the Kangra valley. of branched C15:0 [6]. The Gram-negative bacterial strain Stenotrophomonas maltophilia IHB B 4036, which displayed a fatty acid composition comparable to the clinical strains of Stenotrophomonas maltophilia [22], was clustered with REFERENCES the Gram-positive bacterial strains, owing to the presence of C15:0 iso as the major fatty acid; sum in feature 8 in a 1. Adhikari, T. B., C. M. Joseph, G. P. Yang, D. A. Phillips, and L. minor amount and the complete absence of summed feature M. Nelson. 2001. Evaluation of bacteria isolated from rice for 8 common to Gram positive bacteria (Supplementary Table 3). plant growth promotion and biological control of seedling 47: - The specific fatty acid-profiles corroborate the results of disease of rice. Can. J. Microbiol. 916 924. 2. Albino, U., D. P. Saridakis, M. C. Ferreira, M. Hungria, P. 16S rDNA sequencing and carbon-source metabolic profiling Vinuesa, and G. Andrade. 2006. High diversity of diazotrophic that various rDNA types are distinct, taxonomically. bacteria associated with the carnivorous plant Drosera villosa Diazotrophic bacteria varied in acetylene reduction, var. villosa growing in oligotrophic habitats in Brazil. Plant Soil which is an indirect measure of N-fixing potential while 287: 199-207. also being a specific means for monitoring of functional 3. Andrade, G., E. Esteban, L. Velascol, J. L. Maria, and E. nitrogenase activity [3, 42]. The high acetylene reduction J. Bedmar. 1997. Isolation and identification of N2-fixing by tea diazotrophic bacterial strains closely related to microorganisms from the rhizosphere of Capparis spinosa (L.). Delftia tsuruhatensis, Azospira oryzae and Herbaspirillum Plant Soil 197: 19-23. huttiense reinforces the findings of a high nitrogenase 4. Bae, H. S., B. A. Rash, F. A. Rainey, M. Fernanda Nobre, I. activity reported for the latter bacteria [11, 19, 58]. However, Tiago, M. S. da Costa, and W. M. Moe. 2007. Description of eight isolates grown on NFb media were found to be Azospira restricta sp. nov., a nitrogen-fixing bacterium isolated from groundwater. Int. J. Syst. Evol. Microbiol. 57: 1521-1526. negative for acetylene reduction, possibly due to the 5. Baldani, V. L. D. and J. Doebereiner. 1980. Host-plant specificity unique requirement involving some specific conditions for in the infection of cereals with Azospirillum spp. Soil Biol. nitrogenase activity by these isolates [7, 16, 32], or due to Biochem. 12: 433-439. their ability to be nurtured only on nitrogen-fixed and 6. Basile, F., K. J. Voorhees, and T. L. Hadfield. 1995. Microorganism released by diazotrophs [10]. Gram-type differentiation based on pyrolysis-mass spectrometry 554 Arvind et al.

of bacterial fatty acid methyl ester extracts. Appl. Environ. 22. Hejnar, P., Z. Chmela, and M. Rypka. 2002. Fatty acid analysis Microbiol. 61: 1534-1539. of Stenotrophomonas maltophilia clinical strains showing different 7. Beauchamp, C. J., G. Levesque, D. Prevost, and F. P. Chalifour. susceptibility to antibiotics at 30 and 37oC. Folia Microbiol. 47: 2006. Isolation of free-living dinitrogen-fixing bacteria and their 742-746. activity in compost containing de-inking paper sludge. Bioresour. 23. Kennedy, I. R., A. T. M. A. Choudhury, and M. L. Keeskes. Technol. 97: 1002-1011. 2004. Non-symbiotic bacterial diazotrophs in crop farming 8. Bhattacharjee, R. B., A. Singh, and S. N. Mukhopadhyay. 2008. systems: Can their potential for plant growth promotion be Use of nitrogen-fixing bacteria as biofertilizer for non-legumes: better exploited? Soil Biol. Biochem. 36: 1229-1244. Prospects and challenges. Appl. Microbiol. Biotechnol. 80: 24. Kimura, M. 1980. A simple method for estimating evolutionary 199-209. rates of base substitutions through comparative studies of 9. Bradford, M. M. 1976. A rapid and sensitive method for the nucleotide sequences. J. Mol. Evol. 16: 111-120. quantitation of microgram quantities of protein utilizing the 25. Ladha, J. K. and P. M. Reddy. 2000. Steps for nitrogen-fixation principle of protein-dye binding. Anal. Biochem. 72: 248-254. in rice, pp. 359. In J. K. Ladha and P. M. Reddy (eds.). “The 10. Chowdhury, S. P., M. Schmid, A. Hartmann, and A. K. Tripathi. quest for nitrogen fixation in rice” Proceedings of the Third 2007. Identification of diazotrophs in the culturable bacterial Working Group Meeting on Assessing Opportunities for Nitrogen community associated with roots of Lasiurus sindicus, a Fixation in Rice, Makai City (Philippines). perennial grass of Thar desert, India. Microb. Ecol. 54: 82-90. 26. Lugtenberg, B. J. J. and L. C. Dekkers. 1999. What makes 11. Ding, L. and A. Yokota. 2004. Proposals of Curvibacter gracilis Pseudomonas bacteria rhizosphere competent? Environ. Microbiol. gen. nov., sp. nov. and Herbaspirillum putei sp. nov. for bacterial 1: 9-13. strains isolated from well water and reclassification of [Pseudomonas] 27. Martensson, L., B. Deiz, I. Wartianen, Z. Weiwen, El-S. Rehab, huttiensis, [Pseudomonas] lanceolata, [Aquaspirillum] delicatum and R. Ulla. 2009. Diazotrophic diversity, nifH gene expression and [Aquaspirillum] autotrophicum as Herbaspirillum huttiense and nitrogenase activity in a rice paddy field in Fujian, China. comb. nov., Curvibacter lanceolatus comb. nov., Curvibacter Plant Soil 325: 207-218. delicatus comb. nov. and Herbaspirillum autotrophicum comb. 28. Martinez, L., J. Caballero-Mellado, J. Orozco, and E. Martinez- Int. J. Syst. Evol. Microbiol. 54: 2223-2230. Romero. 2003. Diazotrophic bacteria associated with banana 12. Ding, Y., J. Wang, Y. Liu, and S. Chen. 2005. Isolation and (Musa spp.). Plant Soil 257: 35-47. identification of nitrogen-fixing bacilli from plant rhizospheres 29. Martinez-Romero, E. and J. Caballero-Mellado. 1996. Rhizobium in Beijing region. J. Appl. Microbiol. 99: 1271-1281. phylogenies and bacterial genetic diversity. Crit. Rev. Plant Sci. 13. Dutta, D. and R. Gachhui. 2006. Novel nitrogen-fixing Acetobacter 15: 113-140. nitrogenifigens sp. nov., isolated from Kombucha tea. Int. J. 30. Matiru, V. N. and F. D. Dakora. 2004. Potential use of rhizobial Syst. Evol. Microbiol. 56: 1899-1903. bacteria as promoters of plant growth for increased yield in 14. Eckford, R., F. D. Cook, D. Saul, J. Aislabie, and J. Foght. landraces of African cereal crops. Afr. J. Biotechnol. 3: 1-7. 2002. Free-living heterotrophic nitrogen-fixing bacteria isolated 31. Mehboob, I., M. Naveed, and A. Zahir. 2009. Rhizobial association from fuel-contaminated Antarctic soil. Appl. Environ. Microbiol. with non-legumes: Mechanisms and applications. Crit. Rev. 68: 5181-5185. Plant Sci. 28: 432-456. 15. Elmerich, C. and W. E. Newton. 2007. Associative and Endophytic 32. Montanez, A., C. Abreu, P. R. Gill, G. Hardarson, and M. Nitrogen-fixing Bacteria and Cyanobacterial Associations. Sicardi. 2009. Biological nitrogen fixation in maize (Zea mays Springer Verlag. L.) by 15N isotope-dilution and identification of associative 16. Estrada-De Los Santos, P., R. Bustillos-Cristales, and J. culturable diazotrophs. Biol. Fertil. Soils 45: 253-263. Cabellaro-Mellado. 2001. Burkholderia, a genus rich in plant- 33. Naher, U. A., O. Radziah, Z. H. Shamsiddin, M. S. Halimi, and associated nitrogen fixers with wide environmental and geographic M. Razi. 2009. Isolation of diazotrophs from different soils of distribution. Appl. Environ. Microbiol. 67: 2790-2798. Tanjong Karang rice growing area in Malaysia. Int. J. Agric. 17. Gardener, B. B. M. 2004. Ecology of Bacillus and Paenibacillus Biol. 11: 547-552. spp. in agricultural systems. Phytopathology 94: 1252-1258. 34. Natsvaladze, M. Y., L. K. Nitse, and S. N. Sakharova. 1992. 18. Guerlava, P., V. Izac, and J. L. Tholozan. 1998. Comparison of Free-living and associative diazotrophs in the rhizocoenosis of different methods of cell lysis and protein measurements in tea. Izvestia Timiryazevskoi Sel’ sokhozyaistvennoi Akademii 2: Clostridium perfringens: Application to the cell volume 95-102. determination. Curr. Microbiol. 36: 131-135. 35. Ndifon, W., J. B. Plotkin, and J. Dushoff. 2009. On the accessibility 19. Han, J., L. Sun, X. Dong, Z. Cai, X. Sun, H. Yang, Y. Wang, of adaptive phenotypes of a bacterial metabolic network. PLoS and W. Song. 2005. Characterization of a novel plant growth- Comput. Biol. 5: 1-11. promoting bacteria strain Delftia tsuruhatensis HR4 both as a 36. O’Donnell, A. G., M. Seasman, A. Macrae, I. Waite, and J. T. diazotroph and a potential biocontrol agent against various plant Davies. 2001. Plants and fertilisers as drivers of change in pathogens. Syst. Appl. Microbiol. 28: 66-76. microbial community structure and function in soils Plant Soil 20. Han, W., S. J. Kemmitt, and P. C. Brookes. 2007. Soil microbial 232: 135-145. biomass and activity in Chinese tea gardens of varying stand 37. Oh, K., T. Kato, Z. P. Li, and F. Y. Li. 2006. Environmental age and productivity. Biochemistry 39: 1468-1478. problems from tea cultivation in Japan and a control measure 21. Hardy, R. W. F., R. D. Holsten, E. K. Jackson, and R. C. Burns. using calcium cyanamide. Pedosphere 16: 770-777.

1968. The acetylene-ethylene assay for N2 fixation: Laboratory 38. Okano, K., K. Chutani, and K. Matsuo. 1997. Suitable level of and field evaluation. Plant Physiol. 43: 1185-1207. nitrogen fertilizer for tea (Camellia sinensis L.) plants in relation DIVERSITY OF DIAZOTROPHS ASSOCIATED WITH TEA ROOTS 555

to growth, photosynthesis, nitrogen uptake and accumulation of 51. Sarita, S., U. B. Priefer, J. Prell, and P. K. Sharma. 2008. free amino acids. Jpn. J. Crop Sci. 66: 279-287. Diversity of nifH gene amplified from rhizosphere soil DNA. 39. Pandey, A. and L. M. S. Palni. 1996. The rhizosphere effect of Curr. Sci. 94: 109-115. tea on soil microbes in a Himalayan monsoonal location. Biol. 52. Sasser, M. 1990. Tracking a Strain Using the Microbial Identification Fertil. Soils 21: 131-137. System. Technical Note 102. MIS, Newark, DE. 40. Pandey, A. and L. M. S. Palni. 1997. Bacillus species: The 53. Sato, A., T. Watanabe, Y. Unno, E. Purnomo, M. Osaki, and T. dominant bacteria of the rhizosphere of established tea bushes. Shinano. 2009. Analysis of diversity of diazotrophic bacteria Microbiol. Res. 152: 359-365. associated with the rhizosphere of a tropical arbor, Melastoma 41. Pandey, A., L. M. S. Palni, and N. Coulomb. 1997. Antifungal malabathricum L. Microbes Environ. 24: 81-87. activity of bacteria isolated from rhizosphere of established tea 54. Sharma, D. K., S. Thakur, and K. L. Sharma. 2005. Effect of bushes. Microb. Ecol. 152: 105-112. method, time and levels of phosphorous application on yield 42. Park, M., C. Kin, J. Yang, H. Lee, W. Shin, S. Kim, and. T. Sa. and quality of tea (Camellia sinensis). Indian J. Agron. 50: 2005. Isolation and characterization of diazotrophic growth 324-326. promoting bacteria from rhizosphere of agricultural crops of 55. Soares, R. A., L. F. R. Roesch, G. Zanatta, F. A. O. Camargo, Korea. Microbiol. Res. 160: 127-133. and L. M. P. Passaglia. 2006. Occurrence and distribution of 43. Prakamhang, J., K. Minamisawa, K. Teamtaisong, N. Boonkerd, nitrogen fixing bacterial community associated with oat (Avena and N. Teaumroong. 2009. The communities of endophytic sativa) assessed by molecular and microbiological techniques. diazotrophic bacteria in cultivated rice (Oryza sativa L.). Appl. Appl. Soil Ecol. 33: 221-234. Soil Ecol. 42: 141-149. 56. Sood, A., S. Sharma, V. Kumar, and R. L. Thakur. 2007. Antagonism 44. Rodriguez Caceres, E. A. 1982. Improved medium for isolation of dominant bacteria in tea rhizosphere of Indian Himalayan of Azospirillum spp. Appl. Environ. Microbiol. 44: 990-991. regions. J. Appl. Sci. Environ. Manage. 11: 63-66. 45. Roesch, L. F. W., F. A. O. Camargo, F. M. Bento, and E. W. 57. Sud, R. G. and A. Baru. 2000. Seasonal variations in theaflavins, Triplett. 2008. Biodiversity of diazotrophic bacteria within the thearubigins, total colour and brightness of Kangra orthodox tea soil, root and stem of field-grown maize. Plant Soil 302: 91- (Camellia sinensis (L.) O. Kuntze) in Himachal Pradesh. J. Sci. 104. Food Agric. 80: 1291-1299. 46. Rozycki, H., H. Dahm, E. Strzelcyk, and C. Y. Li. 1999. 58. Tan, Z. and B. Reinhold-Hurek. 2000. Dechlorosoma suillum Diazotrophic bacteria in root free soil in the root of pine (Pinus Achenbach et al. 2001 is a later subjective synonym of Azospira sylvestris L.) and oak (Quercus robur L.). Appl. Soil Ecol. 12: oryzae Reinhold-Hurek and Hurek 2000. Int. J. Syst. Evol. 239-250. Microbiol. 53: 1139-1142. 47. Sachdev, D., V. Agarwal, P. Verma, Y. Shouche, P. Dhakephalkar, 59. Van de Peer, Y. and R. de Watcher. 1994. TREECON for and B. Chopade. 2009. Assessment of microbial biota Windows: A software package for the construction and drawing associated with rhizosphere of wheat (Triticum aestivum) during of evolutionary trees for the Microsoft Windows environment. flowering stage and their plant growth promoting traits. Internet Comput. Appl. Biosci. 10: 569-570. J. Microbiol. Accessible http://www.ispub.com/journal/the_internet_ 60. Weisburg, W. G., S. M. Barns, D. A. Pelletier, and D. J. Lane. journal_of_microbiology/volume_7_number_2_26. 1991. 16S Ribosomal DNA amplification for phylogenetic study. 48. Saikia, S. P. and V. Jain. 2007. Biological nitrogen fixation with J. Bacteriol. 173: 697-703. non-legumes: An achievable target or a dogma? Curr. Sci. 92: 61. Zhang, G. X., G. X. Peng, E. T. Wang, H. Yan, H. Y. Qing, W. 317-322. Zhang, X. Lou, H. Wu, and Z. Y. Tan. 2008. Diverse endophytic 49. Saleena, L. M., S. Rangarajan, and S. Nair. 2002. Diversity of nitrogen-fixing bacteria isolated from wild rice Oryza rufipogon Azospirillum strains isolated from rice plants grown in saline and description of Phytobacter diazotrophicus gen. nov. sp. nov. and nonsaline sites of coastal agricultural ecosystems. Microb. Arch. Microbiol. 189: 431-439. Ecol. 44: 271-277. 50. Saravanakumar, D., C. Vijayakumar, N. Kumar, and R. Samiyappan. 2007. PGPR-induced defense responses in the tea plant against blister blight disease. Crop Prot. 26: 556-565.