Nutrient Broth culture (10mL) of Arthrobacter nicotianae strain PR was grown at 30 °C, by picking a single colony from a nutrient agar stock plate prepared earlier by streak-plate technique. Genomic DNA was isolated using DNA bactozol kit (Molecular Research Center, OH, USA). The genomic DNA isolation was confirmed by resolving the bacterial lysate in agarose gel before use in PCR.

CER1 and EMT1 primers (8) were used in this study. Primers were synthesized at Integrated DNA Technologies, Coralville, IA, USA. The PCR reactions were carried out as standardized earlier(9). Thermocycler (Eppendorf Mastercycler personal, Eppendorf) was used to cycle the PCR. The following conditions: 95 ºC (5 min), 95 ºC (1 min), 55 ºC (1 min), and 72 ºC (3 min) for 30 cycles were adopted.

Amplicons obtained from PCR were sub cloned in E. coli, using manufacturer’s protocol instructions (Invitrogen USA). Positive clones were screened using IPTG/X-Gal (Fermentas, USA) with Ampicillin (Sigma, USA). Positive colonies were grown in ampicillin /LB liquid culture followed by plasmid isolation using Pure Yield Plasmid Miniprep System (Promega, USA). The clones were confirmed by EcoR1 restriction before sequencing using a T3 and or T7 primer at the Center for Genetics and Molecular Medicine (CGeMM) DNA Core facility of the University of Louisville, KY, USA. Further analysis of the DNA sequence was done using NCBI-blast and MegAlign software (lasergene, DNAstar).

After confirmation of the DNA sequence using NCBI vector contamination software (vecscreen), further computational analysis was conducted. Homologous nucleotide and or amino acid sequence search was performed using NCBI blast search of protein databases (http://www.ncbi.nlm.nih.gov/BLAST/). The newly determined Glutamate Synthase sequence was deposited in GenBank and an accession number FJ979920 was obtained using the BankIT:GenBank (www.ncbi.nlm.nih.gov/BankIT/) submission program. To confirm the specificity signature sequence conserved amino acid residue search was performed during multiple sequence alignment. MegAlign software was then used to generate the phylogenetic tree.

To determine the domain structure of the clone, Cn3D 4.1 (NCBI) software was used. The newly cloned sequence was compared to 3D domain model constructed from multiple sequence alignment of many peptides in the same class. Domain alignment was performed to demonstrate common and consensus regions. Images were exported from Cn3D software as JPEG.

In an effort to find a putative identity of the gene partial sequence, we analyzed the amplicon sequence by multiple sequence alignment using multiple sequence alignment software, MegAlign, DNAstar (Fig 2). Alignment was performed with potential matches obtained from NCBI blast search of microbe protein databases. Using ClustalW, we revealed the presents of conserved amino acid residues shown in figure 2. A thorough search of the conserved domains revealed that the newly cloned sequence matched in part to the TIM barrel of the GltS as well as having FMN binding consensus amino acid residues (Fig. 2). A comparison of the TIM barrel primary structure and our Arthrobacter nicotianae GltS sequence showed amino acids considered critical in binding FMN, a cofactor required for the function of the enzyme (10-12). These amino acids found in the new GltS sequence are residues; GGMSYGS- position 1165 to 1171, GEG- position 1058 to 1060, and K- position 1145. We noted a unique substitution of a conserved Q by P in position 1123. As shown in figure 3, there are other conserved amino acids revealed by this analysis that probably do not play a role in FMN binding. It is our opinion based on the structure of GltS that these residues may have a role in forming and maintaining the TIM barrel tertiary structure.

Figure 2

Fig 2. Multiple sequence alignment of Glutamate synthase reveals consensus amino acid residues. The homologous matches obtained from BLASTX search were aligned using Megalign program (Lasergene, DNAstar). Shown are only 8 of the 72 bacteria strains revealed by the NCBI blast search of microbe protein databases. ClustalW shows the presents of conserved amino acids residues in specific positions indicated

Tertiary structure analysis of the newly sequenced GltS was performed using Cn3D 4.1 software (NCBI) (13). Conserved amino acid residues of the structural alpha and beta sheets that form the TIM barrel are present in Arthrobacter nicotianae (Fig. 3). As shown in figure 3, on the left are three different views of the domains that form the TIM barrel (11) and on the right are the TIM barrel domains in the newly cloned GltS sequence. Included in the images is the putative binding site for the phosphate group of the cofactor i.e. FMN (Fig. 3).

Phylogenetic tree of partial glutamate synthase sequence cloned from arthrobacter nicotianae along with other microbial sequences and genomes was performed using MegAlign software (Fig. 4). Analysis included all 72 matches from the NCBI blast search (supplementary 1 data not shown). As shown in Figure 4, similar strains of bacteria clustered together. Notable is that Arthrobacter nicotianae did not cluster together with other Arthrobacter despite having common consensus amino acid residues. This suggests possible diversity and divergent evolutionary development in the bacterial species.

Figure 3

Fig 3. 3D analysis reveals conserved TIM barrel domain in GltS. The tertiary structure of the newly cloned GltS was compared with that of similar multiple sequences in protein databases. This was performed using Cn3D 4.1 software. The 3D model comparison revealed common conserved alpha and beta sheets that partly form the TIM barrel and the binding site of cofactor phosphate group. On the right are three views of the TIM barrel from multiple sequence alignment of sequences on NCBI link protein databases and in the left are three similar images showing domains in common with the newly cloned GltS sequence.

Figure 4

Fig 4. The phylogenetic tree constructed and based on published microbe Glutamate Synthase. Sequences of and other related microbes aligned in Fig. 3 and Supplementary 1 were compared and used in phylogenetic tree construction using MegAlign software. The tree is unrooted and each node and length of arm represents sequence divergence estimated as number of amino acid substitution as the tree branches.

Glutamate synthase is phylogenetically interesting as well as metabolically pivotal to the biology of bacteria and other organisms (15-20). Previous phylogenetic analysis revealed that GltS had undergone lateral gene transfer from bacteria to a common ancestor of animals, fungi, and plants (21). Our analysis demonstrates a close relationship between strains of the same type of bacteria as expected. Unique is that Arthrobacter nicotianae as well as Mycobacterium sp did not seem to cluster with other Arthrobacter or Mycobacterium bacteria. This suggests possible divergence in the phylogeny of Glutamate synthase in these bacteria. The expression of genes, such as GltS are dependent on environmental factors (22). The growth of Arthrobacter bacteria was previously shown to be in part dependent on the levels of ammonia and type of carbon source like methylamine (22). If this is true, then the evolutionary development of the Arthrobacter bacteria may be tied to the local environment, but that is beyond the scope of this study.

It is important to highlight a few bacteria that matched with the newly cloned GltS. In mycobacterium, the activity of GltS and Glutamine synthase have been shown to be dependent on nitrogen source and a levels of ammonia in the environment (23-25). Nocardia sp and Frankia similarly demonstrates enhanced expression of GS/GOGAT pathway as the major pathway for the assimilation of ammonia with glutamate as the source of nitrogen (26-28). GltS harbored in Arthrobacter nicotianae also shows similarity to those of soil inhabiting exiguobacteria suggesting influence from a possible common environmental factor.

The availability and use of software like Cn3D 4.1 has potential to positively impact analysis and discovery of life science data like we have not seen before (13, 29-33). By importing the newly cloned GltS sequence into the software we were able to align the primary structure as well as determine conserved domains and view them in their putative three dimensional forms. This approach to analysis of genomic data as we have seen has the potential to identify and rapidly unveil the metabolism as well as the phylogentic position of the organism.

This work was supported by National Science Foundation, USA (Grant # MCB-0639356 to Narayanan Rajendran). Authors thank ASM/JGI(DOE) and Drs. Brad Goodner, Cheryl Kerfeld, Kathleen Scott and Sam Donovan of ASM Spring-2009 Bioinformatics Institute for sequence alignment, phylogenetic and bioinformatics training.