Bacterial Analogs of Plant Tetrahydropyridine Alkaloids Mediate Microbial Interactions in a Rhizosphere Model System

Abstract

Plants expend significant resources to select and maintain rhizosphere communities that benefit their growth and protect them from pathogens. A better understanding of assembly and function of rhizosphere microbial communities will provide new avenues for improving crop production. Secretion of antibiotics is one means by which bacteria interact with neighboring microbes and sometimes change community composition. In our analysis of a taxonomically diverse consortium from the soybean rhizosphere, we found that Pseudomonas koreensis selectively inhibits growth of Flavobacterium johnsoniae and other members of the Bacteroidetes grown in soybean root exudate. A genetic screen in P. koreensis identified a previously uncharacterized biosynthetic gene cluster responsible for the inhibitory activity. Metabolites were isolated based on biological activity and were characterized using tandem mass spectrometry, multidimensional nuclear magnetic resonance, and Mosher ester analysis, leading to the discovery of a new family of bacterial tetrahydropyridine alkaloids, koreenceine A to D (metabolites 1 to 4). Three of these metabolites are analogs of the plant alkaloid γ-coniceine. Comparative analysis of the koreenceine cluster with the γ-coniceine pathway revealed distinct polyketide synthase routes to the defining tetrahydropyridine scaffold, suggesting convergent evolution. Koreenceine-type pathways are widely distributed among Pseudomonas species, and koreenceine C was detected in another Pseudomonas species from a distantly related cluster. This work suggests that Pseudomonas and plants convergently evolved the ability to produce similar alkaloid metabolites that can mediate interbacterial competition in the rhizosphere. IMPORTANCE The microbiomes of plants are critical to host physiology and development. Microbes are attracted to the rhizosphere due to massive secretion of plant photosynthates from roots. Microorganisms that successfully join the rhizosphere community from bulk soil have access to more abundant and diverse molecules, producing a highly competitive and selective environment. In the rhizosphere, as in other microbiomes, little is known about the genetic basis for individual species’ behaviors within the community. In this study, we characterized competition between Pseudomonas koreensis and Flavobacterium johnsoniae, two common rhizosphere inhabitants. We identified a widespread gene cluster in several Pseudomonas spp. that is necessary for the production of a novel family of tetrahydropyridine alkaloids that are structural analogs of plant alkaloids. We expand the known repertoire of antibiotics produced by Pseudomonas in the rhizosphere and demonstrate the role of the metabolites in interactions with other rhizosphere bacteria.

Draft genome sequence of a cold-adapted phosphorous-solubilizing Pseudomonas koreensis P2 isolated from Sela Lake, India

“…Primary genome assembly generated 40 contigs with 59.8% GC content and a scaffold N50 value of 438,962 bp. Similar reports of genome assembly of P. koreensis strains from soybean and rice rhizosphere have been published earlier (Lozano et al 2019;Lin et al 2016). In this study, genome validation and quality control using Bowtie software revealed 29 million reads mapped concordantly and 6 million reads mapped discordantly with an overall alignment rate of 97.82%.…”

Identifying the Compounds of the Metabolic Elicitors of Pseudomonas fluorescens N 21.4 Responsible for Their Ability to Induce Plant Resistance

“…They have a wide range of biological activities, including insecticidal [55], antibacterial and antifungal [56][57][58]. Literature does mention other alkaloid compounds produced by P. fluorescens, such as pyrrolnitrin or hydropyridine-type compounds that also show antibiotic and antifungal activity [28,59]. However, hapalindole-type natural products have never been described in P. fluorescens; hence, this result would require further investigation to be more conclusive.…”

Microbial model communities: To understand complexity, harness the power of simplicity

“…Model communities could, for example, be used to elucidate how microorganisms share metabolites and partition resources [20] , [21] , [22] , under what conditions communities are dominated by competitive or cooperative interactions [23] , [24] , [25] , which specific genes that govern community interactions through e.g. secondary metabolites [26] , [27] , and to what extent interactions are dependent on environmental changes [28] . Furthermore, model systems can aid in understanding the links between taxonomic and functional diversity [29] , what factors that contributes to community stability in the face of invasion from non-resident strains or other stressors [30] , [31] , highlight important keystone species for functional stability of a community [32] , [33] , and shed light on the evolution of mutualism between microbes [34] , [35] .…”

“…That said, they comprise completely non-overlapping sets of members; SXMP consists of Stenotrophomonas rhizophila , Xanthomonas retroflexus , Microbacterium oxydans and Paenibacillus amylolyticus , while THOR is built on Pseudomonas koreensis , Flavobacterium johnsoniae and Bacillus cereus . Arguably, the members of THOR would be easier to genetically manipulate using standard protocols, but that has not prevented the SXMP community to be used in a variety of studies [30] , [42] , [62] , [63] , [64] , [65] , compared to THOR which was more recently introduced as a model [26] , [52] . The ease of genetic manipulation and the many known interactions between the species in these model communities make them ideal for identifying which genes that are responsible for interaction behaviors between microbial species.…”

“…The endeavor to tease apart interactions in complex microbial communities needs to start small. Microbial model communities of different scales have contributed to our understanding of microbial predator–prey relationships [23] , [30] , [55] , resource partitioning [35] , [48] , [53] , [57] , pathogenicity [37] , [47] , antibiotic tolerance [28] , [89] , [90] , resistance to stress [28] , [31] , as well as the conditions that trigger the production of secondary metabolites [26] , [46] , [52] . This, however, is just scratching the surface of what model communities could reveal about community dynamics; it is apparent that there is a plethora of biological phenomena related to interactions between microbes to explore, and model communities provide a means to start assessing these interactions and the mechanisms behind them.…”