Legume Nodulating Bacterium, Achromobacter xylosoxidans Found in Tropical Shrub Agroecosystem, Sumatera, Indonesia

Legume nodulating bacteria (LNB), known also as rhizobia, are soil bacteria, which are able to form root nodules and fi x nitrogen in the leguminous plants. The LNB availability in the soil depends on the type of agroecosystem, where plant grows. In this study, we isolated LNB from the shrub agroecosystem in Sumatera, Indonesia, and obtained four selected bacterial strains. Among them, the isolate UGM48a formed root nodule in Macroptilium atropurpureum and showed highest number of nitrogenase activity. UGM48a also contains nifH and nodA genes. An analysis of the PCR-amplifi ed 16S rDNA and BLASTn analysis showed that UGM48a displayed 96% similarity with Achromobacter xylosoxidans. In addition, UGM48a were successfully nodulated Glycine max (L.) merr var. wilis. This is the fi rst report detecting A. xylosoxidans as nodule-forming species for Glycine max possesing the positive copy of nodA gene.


Introduction
Legume nodulating bacteria (LNB), known also as rhizobia, are soil bacteria able to form root nodules and fi x nitrogen in the leguminous plants. Leguminous plants usually establish a symbiosis with rhizobia and benefi t from its ability to fi x atmospheric nitrogen, which allows them to grow more effi ciently on nutrient-limited soils.
LNB availability in the soil may change due to history of land use. Various agroecosystems in Sumatera, Indonesia have different land use, with 87% used for coffee plantation. Shrub agroecosystem means the soil was in fallow for about 5 years after coffee plantation (Evizal et al., 2013). This condition may infl uence the availability of LNB in the soil, because there are shrubbery that dominated this area.
LNB or Rhizobia are divided into several groups, which include alphaproteobacteria comprised of 13 genera of Rhizobium,  (Lin et al., 2008) and Microvirga (Ardley et al., 2012). Nowadays, there are many studies showing that LNB does not always belong to Alphaproteobacteria. Betaproteobacteria are comprised of three genera of Burkholderia (Sheu et al., 2012), Cupriavidus and Herbaspirillum (Euzeby's, 2006). In addition to alphaproteobacteria and betaproteobacteria, there are bacteria belonging to the class gammaproteobacteria (Benhizia et al., 2004;Shiraishi et al., 2010), but their symbiont properties have not been yet demonstrated.
This study reported four LNB based on cell and colony's morphology, as well as the nodulation assay in Macroptilium atropurpureum. Based on sequence analysis of 16S rDNA gene and its nitrogenase activity in M. atropurpureum, it was identified as strain of Achromobacterium xylosoxidans. This bacterium is one of the genera belonging to betaproteobacteria, which have been reported as legume nodulating bacteria for the fi rst time by Benata et al. (2008). A. xylosoxidans formed root nodule in Prosopis julifl ora (Benata et al., 2008); Mucuna bracteata (Salwani et al., 2012); and cowpea (Guimarães et al., 2012). In this study, we also confi rmed the existence of nifH and nodA copy genes. Our fi nding also observed the nodulation ability of A. xylosoxidans in soybean

Bacterial Isolates
Four isolates used in this study were screened from shrub agroecosystem in Sumatera, Indonesia. The pure cultures were maintained on yeast-extract mannitol agar (YEMA) slants at 4 o C using standard procedures (Somasegaran, 1994).

Nodulation and N 2 -fi xation Assay
Four isolates were examined through nodulation and N 2 -fixation assay using Macroptilium atropurpureum. Surfacesterilized seeds of M. atropurpureum were germinated in petridishes. The seedling were transferred to the test chamber called minirhizotron which consists zeolite based medium (Joachim et al., 2006). The seedlings were watered daily with N-free nutrient solution as described by Somasegaran (1994). At the 40 th day, the plants were analyzed for the number of fresh nodules, and their nitrogenase activity was determined using Acetylene Reduction Assay (ARA).
The best isolate showing highest number of nitrogenase activity were subsequently observed in various plants. In this report, soybean (Glicyne max [L.] merr var. wilis) was used as tested plant, to observe the nodulation ability of the selected isolate. The purpose of using this plant was investigate nodulating ability in soybean, since soybean can only be infected by few bacteria. Furthermore, Glicyne max (L.) merr var. wilis is one of the superior types of soybean released in Indonesia since 1983. Surface-sterilized seeds of wilis were germinated in petridishes after soaking the seeds overnight in a sterile water. Observation was performed 40 days after cultivation.

Morphological Observation
Bacterial colonies and colours were observed by using standard microbiological method as described by Yang et al. (2008), Vincent (19721982), and Somasegaran and Hoben (1985). The colonies of pure cultures were maintained on YEMA slants at 4 o C and also stored at 4 o C.

Genomic DNA Isolation
Total DNAs were extracted from 5 ml bacterial cultures grown in Luria Bertani. The cultures were centrifuged at 11000 rpm for 8 min and washed in TE buffer (10mM Tris, 1 mM EDTA; pH 8). The pellets were suspended in 30 μl SDS 10%, incubated at 37 o C for 30 min, then 100 μl NaCl 5 M and 200 μl CTAB were added, and incubated at 65 o C for 15 min. The mixtures were centrifuged at 11000 rpm for 8 min. The top layer was transferred to a new tube. This aqueous phase containing DNA was precipitated with 500 μl isopropanol and then centrifuged at 11000 rpm for 8 min. The precipitated DNA was washed with 100 μl ethanol, vacuum dried, and subsequently dissolved in TE buffer. This method was modifi cation as described method by Ausubel (1995).

PCR amplifi cation of nodA and nifH genes
To amplify nifH gene fragments, degenerate primers were used, forward primer 5'-GCI WTI TAY GGN AAR GGN GG-3' and reverse primer 5'-GCR TAI ABN GCC ATC ATY TC-3' as described by Burgmann and coworkers (2004). In this study, nodA-1 and nod A-2 were used to amplify nodA gene fragments. Forward primer (NodA-1) : 5'-TGC RGT GGA ARN RNN CTG GG<3' and reverse primer (nodA-2) : 5'-GGN CCG TCR TCR AAW GTC ARG-3' were used for PCR amplification of the nodA gene (Kaisa et al., 1998). These were also degenerate primer with notation W=A or T; B=C, G or T; R=A or G; N=A, C, G, or T; and I=inosine. To amplify both gene fragments, each reaction was performed on a 25 μl sample of the PCR reaction mixture contained 21,5 μl Vivantis Kit PCR Mix, the template genomic DNA (50 ng. μl -1 ), 0,5 μl Taq DNA Polymerase, and 25 pmoles each DNA primers. PCR was performed using PCR thermal cycler (T100 TM Thermal Cycler, Bio-Rad). PCR condition for amplifi cation of nifH gene was set as follows : pre denaturation at 95 o C for 5 min, denaturation at 94 o for 30 second, annealing at 56 o C for 30 second, elongation at 72 o C for 1 min (35 cycles), and fi nal elongation at 72 o for 10 min. In addition PCR cycling program for NodA amplifi cation was set as follows : 5 min pre denaturation at 94 o C, followed by 30 cycles of 30 seconds denaturation at 94 o C, 1 min annealing at 55 o C, and 1 min polymerization at 68 o C. Final elongation was 10 min at 68 o C. Amplifi ed bands were resolved by electrophoresis in a 1,5% (w/v) agarose gel in TBE buffer and visualized by ethidium bromide staining in UV-transiluminator. Size was compared by using 100 kb Vivantis DNA Ladder.

PCR amplification and Analysis of 16S rDNA Sequences
The 16S rDNA sequences were amplifi ed from the genomic DNA of the isolates using the universal primer 27F (5'-AGA GTT TGA TCC TGG CTC AG-3') and 1492R primer (5'-GGT TAC CTT GTT ACG ACTT-3') as described by Lane (1991). PCR was performed using PCR thermal cycler (T100 TM Thermal Cycler, Bio-Rad) with total volume 25μl PuRe Taq Ready-To-Go™ PCR Beads consist of : 22μl dH 2 O (nuclease free water), 1μl DNA template (50 ng. μl -1 ), 1 μl each DNA primers (25 pmoles). The PCR condition was set as follows : pre denaturation at 94 o C for 4 min, denaturation at 94 o for 30 second, annealing at 55 o C for 1 minute 30 second, elongation at 72 o C for 30 seconds (30 cycles), and fi nal elongation at 72 o for 10 min. Amplifi ed bands were resolved by electrophoresis in a 1,5% (w/v) agarose gel in TBE buffer and visualized by ethidium bromide staining in UV-transiluminator. Size was being compared by using 1 kb Vivantis DNA Ladder.

Morphological observation
Based on morphological observations of both cells and colonies, the four isolates (Table 1) had similar colonies morphology on YEMA plates (picture not shown). They were transparent to creamy (white) colonies with 0,5-2 mm in diameter after 3 days incubation at room temperature. All bacterial isolates were Gram negative and rod-shaped with different shape of edge, elevation, and inner cell's structure (Table 1).

Nodulation and N-fixation assay of four isolates on M. atropurpureum
The nodulation assay of four isolates on legume M. atropurpureum from shrub agroecosystem in Sumatera, Indonesia were obtained by using minirhizotron method. All bacterial isolates successfully formed nodule with different number of nodules. The fresh nodules were used for measuring the nitrogenase activity. Among them, UGM48a isolates showed the highest number of nitrogenase activity in 0,806 mmol C 2 H 4 .g nodule -1 .h -1 with the highest nodule dry weight (Table 2). Minirhizotron is zeolite based medium in the chamber (Joachim et al., 2006). The seedlings in the chamber were watered daily with N-free nutrient solution. Based on fi ndings by Busch et al. (2006), Rhizotron method is the easiest reliable method to plant and to harvest the belowground biomass in distinct soil layers because of the simple access to the intact soil and root system at harvest. The tip of lateral root spreads out to all part of the chamber of rhizotron, include the surface. It is easier for isolates to initiate fl avanoid (compound that triggers the secretion of Nod factors, which in turn are recognized by the host plant and can lead to root hair deformation and several cellular responses, such as ion fl uxes and the formation of a root nodule). At harvest, which was the 40 th day after cultivation, the plants were analyzed for the number of fresh nodules and their nitrogenase activity was determined using Acetylene Reduction Assay (ARA). The result showed that UGM48a isolates have the highest number of nitrogenase activity in 0,806 mmol C 2 H 4 .g nodule -1 .h -1 . This isolate was used as the selected isolate, which we observed thoroughly.

Analysis of the 16S rDNA sequences
Pure DNA genome of selected isolate, UGM48a, was used in 16s rDNA sequence analysis. This is one of the most effective tools for identifying bacteria. The 16S rDNA fragments of UGM48a was successfully amplifi ed. Based on BLASTn result at NCBI (http://blast.ncbi), UGM48a has 97% in similarity with Achromobacter xylosoxidans with identity 1272 bp query length. A. xylosoxidans is one of genera belong to betaproteobacteria that has been reported as legume nodulating bacteria for the fi rst time by Benata et al. (2008).

NifH and NodA amplifi cation
The existence of NifH and NodA gene copy were also amplified. The molecular weight of the amplified NifH and NodA gene were about 300 bp and 200 bp ( Figure  1), whereas the molecular weight of the amplifi ed NifH genes was between 371-464 bp (Burgmann et al., 2004). The amplifi cation proved that the bacterium has the potential to fi x up free nitrogen. The other publications also mentioned that extensively NifH gene amplification in nitrogen-fixing bacteria, non-symbiotic, or associative. Achouak et al. (1999) successfully amplify NifH genes in bacteria Paenibacillus polimyxa, P. macerans, and P. azotofi xans with a molecular weight of about 370 bp. In the other hand, Mirza et al. (2006) successfully amplify NifH genes in Pseudomonas sp strain K1 with a molecular weight of 360 bp.
UGM48a isolates were amplifi ed using primers degenarated NodA : NodA-1 (5' -TGC RGT GGA ARN RNN CTG GG -3') and primer NodA-2 (5'-GGN CCG TCR TCR AAW GTC ARG -3') with a molecular weight of 200 bp. According to Haukka et al. (1998) the expected amplifi cation product is a band with a molecular weight of 666 bp. But the results showed other bands besides the expected band. Generally, the additional band does not preclude the measurement between the bright and the smear band. However, degenerate primer performed in the attachment of the primary side is not suitable because of the low annealing temperature. In many experiments with degenerate primer, the expected molecular weight from amplifi ed products of PCR can not be predicted with certainty (McPherson and Møller, 2006). Orthologs of the nodA gene, one of the key nod genes encoding an acyl transferase (Kamst et al., 1998), have been discovered in symbiotic Methylobacterium sp. (Sy et al., 2001) and Burkholderia sp (Moulin et al., 2001). Major nod factor-triggered responses include the formation and deformation of root hairs, intra-and extracellular alkalinization, membrane potential depolarization, changes in ion fluxes, induction of early nodulin gene expression, and formation of nodule primordia (Broughton et al., 2000;Perret et al., 2000). So that, the NodA gene has been shown to be a good nodulation marker indeed (Boivin and Giraud, 1999).