Integrated roles of BclA and DD-carboxypeptidase 1 in Bradyrhizobium differentiation within NCR-producing and NCR-lacking root nodules

Abstract

Legumes harbor in their symbiotic nodule organs nitrogen fixing rhizobium bacteria called bacteroids. Some legumes produce Nodule-specific Cysteine-Rich (NCR) peptides in the nodule cells to control the intracellular bacterial population. NCR peptides have antimicrobial activity and drive bacteroids toward terminal differentiation. Other legumes do not produce NCR peptides and their bacteroids are not differentiated. Bradyrhizobia, infecting NCR-producing Aeschynomene plants, require the peptide uptake transporter BclA to cope with the NCR peptides as well as a specific peptidoglycan-modifying DD-carboxypeptidase, DD-CPase1. We show that Bradyrhizobium diazoefficiens strain USDA110 forms undifferentiated bacteroids in NCR-lacking soybean nodules. Unexpectedly, in Aeschynomene afraspera nodules the nitrogen fixing USDA110 bacteroids are hardly differentiated despite the fact that this host produces NCR peptides, suggesting that USDA110 is insensitive to the host peptide effectors and that nitrogen fixation can be uncoupled from differentiation. In agreement with the absence of bacteroid differentiation, USDA110 does not require its bclA gene for nitrogen fixing symbiosis with these two host plants. Furthermore, we show that the BclA and DD-CPase1 act independently in the NCR-induced morphological differentiation of bacteroids. Our results suggest that BclA is required to protect the rhizobia against the NCR stress but not to induce the terminal differentiation pathway.

Figure 1

The USDA110 bclA gene confers sensitivity to the antibiotic bleomycin and the antimicrobial peptide Bac7. (a) Bleomycin sensitivity in B. diazoefficiens strain USDA110 and its bclA mutant derivative and in E. coli strain BW25113ΔsbmA and S. meliloti strain Sm1021ΔbacA carrying an empty vector or a vector expressing USDA110 bclA. Blue bars are for strains expressing bclA and red bars are for strains lacking the gene. Bleomycin concentrations were applied as indicated and growth was determined by optical density measurement at 600 nm with a plate reader after 72 h, 48 h or 24 h incubation for USDA110, Sm1021 and BW25113 respectively. (b) Survival of E. coli strain BW25113ΔsbmA (top) and S. meliloti strain Sm1021ΔbacA (bottom) derivatives carrying the empty pRF771 plasmid or the USDA110 bclA gene located on plasmid pRF771 after treatment with the peptide Bac7 at the indicated concentration. The surviving bacteria were counted and expressed as % from the water control treatment. Error bars in all panels are standard deviations.

Figure 2

The USDA110 bclA gene confers resistance to antimicrobial NCR peptides and mediates the uptake of FITC-NCR247. (a) S. meliloti strain Sm1021ΔbacA derivatives expressing no bacA-related gene (empty vector) or the USDA110 bclA gene were incubated with NCR247, NCR335 or NCR035 or with water (control) and the surviving bacteria were counted and expressed as % from the control treatment. Error bars in all panels are standard deviations. (b) FITC-NCR247 uptake by S. meliloti strain Sm1021ΔbacA or its derivative expressing the USDA110 bclA gene was measured by flow cytometry in the presence of trypan blue to quench extracellular fluorescence. FITC-positive cells are marked with a green box. (c) FITC is not taken up by S. meliloti derivatives. FITC-treated bacteria were analyzed by flow cytometry in the presence of trypan blue to quench extracellular fluorescence. No FITC-positive cells were detected.

Figure 3

The bclA gene of B. diazoefficiens USDA110 complements the bclA mutation in Bradyrhizobium sp. ORS285ΔbclA and the bacA mutation in S. meliloti Sm1021ΔbacA. (a,c,e) Phenotype of A. indica nodules at 14 dpi, infected with the indicated strains. Scale bars are 1 mm. (b,d,f) Bacteroid viability determined by live/dead and calcofluor white staining of nodule sections and confocal microscopy in A. indica nodules induced by the indicated strains. Scale bars are 10 µm. (g,i,k) Phenotype of M. sativa nodules at 28 dpi, infected with the indicated strains. Scale bars are 1 mm. (h,j,l) Bacteroid viability determined by live/dead staining of nodule sections and confocal microscopy in M. sativa nodules induced by the indicated strains. Scale bars are 10 µm.

Figure 4

The B. diazoefficiens bclA mutant is not affected in symbiosis with soybean. (a) Plant growth and nodule phenotype of G. max plants inoculated with B. diazoefficiens USDA110 wild type and the bclA mutant at 14 dpi. Scale bars are 3.5 cm (top panels) and 1 mm (bottom panels). (b) The nitrogen fixation activity of G. max plants inoculated with B. diazoefficiens USDA110 wild type and the bclA mutant, measured by the acetylene reduction assay per mg of fresh nodule weight (nmol ethylene produced per hour incubation and per mg nodule weight). Box-plots represent in the rectangle the first quartile to the third quartile, divided by the median value, whiskers above and below the box show the minimum and maximum measured values. (c,d) Toluidine blue stained thin sections (two top rows) and confocal microscopy of bacteroid viability determination by live/dead staining (bottom row) of G. max nodules induced by USDA110 or USDA110ΔbclA. Scale bars are 100 µm for the top panels and 10 µm for the other panels. (e) Flow cytometry analysis of DNA content by DAPI fluorescence (top) and of cell size by forward scatter (bottom) in free-living B. diazoefficiens USDA110 (pink) or bacteroids isolated from G. max nodules infected with USDA110 wild type (blue) or USDA110ΔbclA (orange). The forward scatter and DAPI fluorescence profiles are completely overlapping for the 3 samples.

Figure 5

The B. diazoefficiens bclA, DD-CPase1, bclA/DD-CPase1 mutant phenotypes in symbiosis with A. afraspera. (a) Plant growth and nodule phenotype of A. afraspera plants inoculated with B. diazoefficiens USDA110 wild type and indicated mutants at 14 dpi. Scale bars are 2 cm (top panels) and 1 mm (bottom panels). (b) Confocal microscopy of bacteroid viability determination by live/dead and calcofluor white staining of A. afraspera nodules induced by USDA110 wild type and indicated mutants. Scale bars are 10 µm. (c) Flow cytometry analysis of DNA content by DAPI fluorescence and of cell size by forward scatter (FS) in free living B. diazoefficiens USDA110 (blue) or bacteroids isolated from A. afraspera nodules infected with USDA110 wild type and indicated mutants (pink). (d) Relative nitrogen fixation activity (% of maximum/h/plant) of A. afraspera plants inoculated with B. diazoefficiens USDA110 wild type and indicted mutants, measured by the acetylene reduction assay per plant. Box-plots represent in the rectangle the first quartile to the third quartile, divided by the median value, whiskers above and below the box show the minimum and maximum measured values. (e) Bleomycin sensitivity in B. diazoefficiens strain USDA110 and its bclA, DD-CPase1, bclA/DD-CPase1 mutant derivatives at 0,0375 µg bleomycin/ml. Growth was determined by optical density measurement at 600 nm with a plate reader after 72 h. Error bars in panels (d) and (e) are standard deviations.

Figure 6

The Bradyrhizobium strain ORS285 ΔbclA, DD-CPase1, ΔbclA/DD-CPase1 mutant phenotypes in symbiosis with A. afraspera and A. indica. (a) Confocal microscopy of bacteroid viability determination by live/dead staining of A. afraspera and A. indica nodules induced by ORS285 wild type and the indicated mutants. Scale bars are 10 µm. (b) Relative nitrogen fixation activity (% of maximum/h/plant) of A. afraspera and A. indica plants inoculated with ORS285 wild type and the indicted mutants, measured by the acetylene reduction assay per plant. Box-plots represent in the rectangle the first quartile to the third quartile, divided by the median value, whiskers above and below the box show the minimum and maximum measured values.