Evidence of Autoinducer-Dependent and -Independent Heterogeneous Gene Expression in Sinorhizobium fredii NGR234

Populations of genetically identical Sinorhizobium fredii NGR234 cells differ significantly in their expression profiles of autoinducer (AI)-dependent and AI-independent genes. Promoter fusions of the NGR234 AI synthase genes traI and ngrI showed high levels of phenotypic heterogeneity during growth in TY medium on a single-cell level. However, adding very high concentrations of N-(3-oxooctanoyl-)-l-homoserine lactone resulted in a more homogeneous expression profile. Similarly, the lack of internally synthesized AIs in the background of the NGR234-ΔtraI or the NGR234-ΔngrI mutant resulted in a highly homogenous expression of the corresponding promoter fusions in the population. Expression studies with reporter fusions of the promoter regions of the quorum-quenching genes dlhR and qsdR1 and the type IV pilus gene cluster located on pNGR234b suggested that factors other than AI molecules affect NGR234 phenotypic heterogeneity. Further studies with root exudates and developing Arabidopsis thaliana seedlings provide the first evidence that plant root exudates have strong effects on the heterogeneity of AI synthase and quorum-quenching genes in NGR234. Therefore, plant-released octopine appears to play a key role in modulation of heterogeneous gene expression.     

Rhizobium meliloti nodD genes mediate host-specific activation of nodABC.

To differentiate among the roles of the three nodD genes of Rhizobium meliloti 1021, we studied the activation of a nodC-lacZ fusion by each of the three nodD genes in response to root exudates from several R. meliloti host plants and in response to the flavone luteolin. We found (i) that the nodD1 and nodD2 products (NodD1 and NodD2) responded differently to root exudates from a variety of hosts, (ii) that NodD1 but not NodD2 responded to luteolin, (iii) that NodD2 functioned synergistically with NodD1 or NodD3, (iv) that NodD2 interfered with NodD1-mediated activation of nodC-lacZ in response to luteolin, and (v) that a region adjacent to and upstream of nodD2 was required for NodD2-mediated activation of nodC-lacZ. We also studied the ability of each of the three R. meliloti nodD genes to complement nodD mutations in R. trifolii and Rhizobium sp. strain NGR234. We found (i) that nodD1 complemented an R. trifolii nodD mutation but not a Rhizobium sp. strain NGR234 nodD1 mutation and (ii) that R. meliloti nodD2 or nodD3 plus R. meliloti syrM complemented the nodD mutations in both R. trifolii and Rhizobium sp. strain NGR234. Finally, we determined the nucleotide sequence of the R. meliloti nodD2 gene and found that R. meliloti NodD1 and NodD2 are highly homologous except in the C-terminal region. Our results support the hypothesis that R. meliloti utilizes the three copies of nodD to optimize the interaction with each of its legume hosts.