Burkholderia genome mining for nonribosomal peptide synthetases reveals a great potential for novel siderophores and lipopeptides synthesis

Abstract Burkholderia is an important genus encompassing a variety of species, including pathogenic strains as well as strains that promote plant growth. We have carried out a global strategy, which combined two complementary approaches. The first one is genome guided with deep analysis of genome sequences and the second one is assay guided with experiments to support the predictions obtained in silico. This efficient screening for new secondary metabolites, performed on 48 gapless genomes of Burkholderia species, revealed a total of 161 clusters containing nonribosomal peptide synthetases (NRPSs), with the potential to synthesize at least 11 novel products. Most of them are siderophores or lipopeptides, two classes of products with potential application in biocontrol. The strategy led to the identification, for the first time, of the cluster for cepaciachelin biosynthesis in the genome of Burkholderia ambifaria AMMD and a cluster corresponding to a new malleobactin‐like siderophore, called phymabactin, was identified in Burkholderia phymatum STM815 genome. In both cases, the siderophore was produced when the strain was grown in iron‐limited conditions. Elsewhere, the cluster for the antifungal burkholdin was detected in the genome of B. ambifaria AMMD and also Burkholderia sp. KJ006. Burkholderia pseudomallei strains harbor the genetic potential to produce a novel lipopeptide called burkhomycin, containing a peptidyl moiety of 12 monomers. A mixture of lipopeptides produced by Burkholderia rhizoxinica lowered the surface tension of the supernatant from 70 to 27 mN·m−1. The production of nonribosomal secondary metabolites seems related to the three phylogenetic groups obtained from 16S rRNA sequences. Moreover, the genome‐mining approach gave new insights into the nonribosomal synthesis exemplified by the identification of dual C/E domains in lipopeptide NRPSs, up to now essentially found in Pseudomonas strains.

Two regions of the NRPS gene bamb_6472 were amplified using primers Up6472-F and Up6472-R (1015 bp amplicon) and primers Down6472-F and Down6472-R (887 bp amplicon), respectively. These primers contained recombination sites for the plasmid. The PCR mix consisted of 25 μl of PCR Master Mix (Thermo Scientific Fermentas), 10 μl Q-solution (Qiagen), 1.25 μl of each primer (each at 20 μM), 7.5 μl water and 5 μl of genomic DNA. The reaction mixture was subjected to the following thermal cycles: one cycle at 94°C for 3min; 30 cycles (94°C, 30 s; 60°C, 45 s; 72°C, 2 min) and a final extension at 72°C for 10 min. Plasmid pMQ30 was purified using GeneJet plasmid miniprep kit (Thermo Scientific Fermentas). A 5 µl sample was checked on a 1% agarose gel. Subsequently, the purified plasmid was digested by a mix of 9.5 μl milliQ water, 4 μl Tango yellow buffer (2x), 5 μl plasmid pMQ30, 1 μl BamHI and 0.5 μl EcoRI. The mix was incubated for 2h at 37°C. 5 μl of the plasmid digest was checked on 1% agarose and the concentration of the digest was also measured.

2-In vivo cloning in S. cerevisiae InvSc1
The PCR products of bamb_6472 were cloned flanking each other through in vivo homologous recombination in the yeast, S. cerevisiae InvSc1, grown overnight in yeast peptone dextrose (YPD) at 30°C. Then 0.5 ml were centrifuged for 1 min at 3000 rpm. The cell pellet was washed with 0.5 ml lazy bones solution and the following was added to the mix: 20 μl of carrier DNA (2mg/ml), 45 μl of each PCR products UP and DOWN, 5 μl digested pMQ30 plasmid. The mix was homogeneized for 1 min on a Vortex and incubated overnight at room temperature. The mixture was subjected to heat shock for 12 min at 42°C. Then, cells were centrifuged for 1 min at 3000 rpm, washed with 0.6 ml TE buffer and the resulting pellet was further redissolved in 0.6 ml of TE buffer and the cells were plated on SD-uracil medium. After 2 to 3 days of incubation, colonies were activated with 6 ml of SD-uracil medium. Further, plasmid was isolated using miniprep kit (Fermentas) and confirmed by PCR using primers Up6472-F and Down6472-R. 10 μl of the eluted plasmid were checked on 1% agarose gel.

3-Construction of AMMD-deficient mutant
Plasmid pMQ30-Δ6472 was introduced into E. coli WM3064 by electroporation. 2 μl of pMQ30 Δ6472 was added to 50 μl of thawed E. coli WM3064 competent cells. Mix was transferred to an electroporation cuvette and electroporated at 2.5 kV/200 ohms/25 µF. Electroporated cells were transferred into 1 ml of LB containing 2% of glucose and incubated in a rotary shaker for 1 h at 37°C. The cells were plated on LB containing 20 μg/ ml of gentamycin sulfate and 100mg/L of DAB (Diaminopimelic acid). Colony PCR was conducted using primers Up6472-F and Down6472-R to confirm the success of electroporation.

The plasmid was mobilized into B. ambifaria AMMD by conjugation. The donor strain E.coli
WM3064+pMQ30-Δ6472 was grown in LB broth containing Gm 20 μg/ml and DAB 100mg/L. The acceptor strain, AMMD, was grown without antibiotics. The cells were centrifuged for 2 min at 5000 rpm, the supernatant was discarded and the cells were washed with 1. 5 ml LB. For conjugation, 200 μl of acceptor strain was mixed with 200 μl of donor strain. This mix was spotted on LB plates containing 100 mg/L DAB, dried, then plates were incubated overnight at 37 °C. Cell mass were then collected and suspended in 1 ml LB. For each conjugate, dilutions up to 10 -3 were performed, plated on LB containing gentamycin 300 μg/ml and incubated overnight at 37 °C.
Colonies were streaked on new LB plates with gentamycin 300 μg/ml. Colony PCR was performed to confirm the presence of the plasmid in merodiploid strain (cell containing plasmid) using primers Up6472-F and Down6472-R. One colony was cultured on LB and incubated overnight at 37 °C in a rotary shaker. Culture was diluted up to 10 -5 and 100 μl were plated on LB containing 10% sucrose. Selected colonies were streak out on new LB plates without gentamycin.