Polyhydroxyalkanoate synthesis by Sinorhizobium meliloti drives a host-specific collapse in symbiosis with Medicago sativa

Polyhydroxyalkanoate synthesis by Sinorhizobium meliloti drives a host-specific collapse in symbiosis with Medicago sativa | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Article Polyhydroxyalkanoate synthesis by Sinorhizobium meliloti drives a host-specific collapse in symbiosis with Medicago sativa Barney Geddes, Garrett Levin, Chinh Luu, Natalie Visich, Scott Hoselton, and 5 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-7715224/v1 This work is licensed under a CC BY 4.0 License Status: Under Review Version 1 posted You are reading this latest preprint version Abstract Naturally occurring root-nodule bacteria (rhizobia) vary substantially in their effectiveness at promoting growth of different plant hosts via symbiotic nitrogen fixation. These variations in rhizobial partner quality have important implications for the productivity of nitrogen-fixing symbioses in natural and agricultural ecosystems, yet we have a limited understanding of the genetic basis for this variation. In a case of host-specific reduction in symbiotic effectiveness (N 2 -fixation) with Medicago sativa , we identified the causative genetic elements from the pSymA replicon of Sinorhizobum meliloti HM006 and show them to be involved in polyhydroxyalkanoate (PHA) production in nitrogen-fixing bacteroids. Transfer of this gene region to a strain that forms an effective symbiosis with Medicago sativa resulted in a complete loss of symbiotic N 2 -fixation. We showed the mechanism for symbiotic collapse is the diversion of succinate semialdehyde pools in the bacteroid to gamma-hydroxybutyrate (GHB) by an iron-containing dehydrogenase, GhbD. These findings reveal unexpected impacts of carbon metabolism changes in nodules on symbiont performance and provide a rare example of mechanism for variation in rhizobium partner quality, suggesting that host-specific metabolic incompatibility may be a key player in the variations in partner quality observed in nature. Biological sciences/Microbiology/Microbial genetics/Bacterial genetics Biological sciences/Plant sciences/Plant symbiosis/Rhizobial symbiosis Symbiosis Nitrogen Fixation Genetics Figures Figure 1 Figure 2 Figure 3 Figure 4 Figure 5 Figure 6 Figure 7 Introduction As the primary sustainable source of nitrogen nutrition in agricultural and natural ecosystems, the nitrogen-fixing root nodule symbiosis between rhizobia and legumes is a central player in global biogeochemical cycling and sustainable agricultural productivity. Variation in the symbiotic capacity of rhizobia is widespread within symbionts that are able to nodulate and fix nitrogen with a given host legume ( 1–3 ). Locally adapted native strains that are less effective at fixing nitrogen for their hosts limit the contribution of the symbiosis to agricultural systems by outcompeting elite inoculant strains for nodulation, dubbed “the rhizobium competition problem” ( 4–6 ). Though mechanisms remain elusive, hosts partially overcome this variation by selectively rewarding or punishing (sanctioning) rhizobia based on their productivity ( 7–10 ). Uncovering the genetics responsible for variation in microbial services to their hosts and elucidating what forces (e.g., “cheating”) maintain such widespread variation in symbioses despite host sanctioning mechanisms have emerged as key questions in microbial ecology ( 3 ). Recent studies point towards much of this variation being associated with mobile modular units such as plasmids and integrative conjugative elements (ICE) that are common features of rhizobium genomes ( 1, 11, 12 ). In the tripartite genome of the model rhizobium Sinorhizobium meliloti , the “symbiotic megaplasmid” pSymA contains many essential genes for symbiosis with its Medicago hosts ( nod, nif and fix ). These gene sets, which encode the ability to produce the “Nod Factor” that initiates nodulation of the host (the nod genes) and to fix nitrogen in root nodules (the nif and fix genes) ( 13 ), account for ~ 5% of its size, yet are sufficient for functional symbiosis in the absence of the remaining ~ 95% of pSymA ( 14 ). The chromid pSymB is more ancestral than pSymA and has been implicated in playing a primary role in saprophytic competence ( 15 ), while also carrying the dicarboxylate transport genes ( dct ) essential for fueling nitrogen fixation in the nodule and exopolysaccharide synthesis ( exo ) genes important for plant invasion ( 16, 17 ). Despite decades of genetic study, the vast majority of genes on these replicons remain undefined with respect to their contributions to symbiosis. In part, this is due to the challenging nature of performing genetic screens for more subtle phenotypes (e.g., partner quality or competition for nodule occupancy) than the nodule-minus phenotypes of nod mutants or nitrogen-fixation-minus phenotypes of nif and fix mutants ( 18, 19 ), or of genes associated with symbiotic incompatibility ( 20, 21 ). The rapid increase in genome sequencing of rhizobium isolates in recent decades has revealed that rhizobia, such as S. meliloti , contain large, open pangenomes ( 11, 22 ), further demonstrating the unexplored genetics that could contribute to variation in symbiosis outcomes. This wealth of genomic data, coupled with next-generation sequencing approaches to screen rhizobium symbiosis traits en masse ( 23, 24 ), enable the prediction of candidate genes that may impact the mutualism using quantitative genetics such as genome-wide association studies (GWAS) ( 12, 25–27 ). However, thus far, few candidates identified by these studies have been empirically validated for their role in symbiosis. Here, we used strains wherein extrachromosomal replicons have been cured ( 15, 28 ) as surrogate hosts to identify genetic determinants that define differences observed in the vast diversity of effectiveness of functional symbioses between Sinorhizobium strains and their Medicago hosts. First, we investigated the co-transfer of partner quality with extrachromosomal replicons from five natural strains. The pSymA replicons were found to have the greatest impact on symbiotic performance and in the most extreme case, the transfer of pSymA from strain HM006 resulted in a substantial reduction of symbiotic performance with Medicago sativa . Next, through a series of refined deletions and gain-of-function experiments, the locus responsible for limiting symbiotic nitrogen fixation was identified and shown to be dominant. Genes within the locus were found to be involved in carbon polymer synthesis during symbiosis, and their impact on symbiotic N 2 -fixation revealed an unanticipated aspect to carbon metabolism in root nodules and its impact on partner quality. Results Co-transfer of host-specific symbiotic effectiveness phenotypes with the Sinorhizobium replicons. We hypothesized that most genes contributing to variance in partner quality would be present on the extrachromosomal replicons in sinorhizobia and reasoned we could identify key genes involved in partner quality by transferring the extrachromosomal replicons (~ 40% of genome) to a surrogate S. meliloti background in which pSymA, pSymB, or both pSymA and pSymB were removed ( 14, 15, 29, 30 ). For replicon transfer, we used four genome-sequenced S. meliloti strains: KH46, HM006, USDA1021, and T073 and one Sinorhizobium medicae strain, WSM419 ( 31, 32 ). These genome-sequenced strains showed significant variation in partner quality (shoot-dry weight accumulation under nitrogen-limiting conditions) with Medicago sativa and Medicago truncatula ( 33 ) ( Supplemental Figure S1 ). The pSymA or pSymB replicons from the five strains were transferred independently or together to derivatives of the long-studied wildtype chromosomal background SU47 (Rm1021-derived background strain RmP110, or Rm5000) ( 33, 34 ) (Fig. 1A). The symbiotic productivity of the resulting hybrid strains with M. sativa revealed a clear positive correlation between the abilities of the natural strains to promote plant growth, and the relative abilities of hybrid strains containing their respective pSymA replicons (Fig. 1B, Supplemental Figures S2A and S3A ). In contrast, no significant difference in plant growth promotion was observed for pSymB transfer (Fig. 1C, Supplemental Figures S2A and S3A ). With M. truncatula as a host, increased growth relative to the wildtype was also observed by transfer of pSymAs from more effective strains (2- to 3-fold increase in SDW) and was enhanced with co-transfer of pSymB (~ 6-fold increase in SDW) ( Supplemental Figures S2B and S3B ). Figure 1. Delineating a Genetic Region Leading to Ineffective Symbiosis with Alfalfa. ( A ) Simplified scheme showing the transfer of pSymA from natural strains into a wildtype strain cured of its own pSymA. ( B ) Comparison between the relative abilities of natural strains versus their corresponding pSymA hybrid to promote shoot dry weight production by M. sativa . ( C ) Comparison between the relative abilities of natural strains versus their corresponding pSymB hybrid to promote shoot dry weight production by M. sativa . ( B and C ) Data are expressed as relative shoot dry weight compared to the mean maximum shoot dry weight of the best performing strain. ( D ) Diagram highlighting unique regions present in the HM006 pSymA replicon and absent in the pSymA of the wildtype RmP110 (blue) and other strains (KH46C, T073, USDA1106, and WSM419) (orange). ( E ) Representative images of M. sativa shoots and nodules inoculated with wildtype, pSymA hybrid, and Deletion 1. Scale bar represents 5 cm for the shoot panel and 0.5 cm for the nodule panel. ( F ) Quantitation of shoot dry weight accumulation by M. sativa. A genetic determinant on the S. meliloti HM006 pSymA leads to a collapse of symbiotic efficiency with M. sativa. As a platform to investigate genetics that underly variation in pSymA partner quality, we chose the pSymA that conferred the most dramatic phenotype; a collapse in the productivity of the symbiosis with M. sativa conferred by the pSymA from S. meliloti HM006 when transferred to the RmP110 background, which forms an effective symbiosis with M. sativa when bearing its native pSymA. We first compared the pSymA contents of HM006 and RmP110. While known symbiosis gene sets in the pSymAs were nearly identical ( Supplemental Fig. 4 ), two large regions from HM006 pSymA were found to be absent from the RmP110 pSymA (Fig. 1D, Supplemental Figure S5 ). Deletion of these regions from the HM006 pSymA-hybrid strain revealed that the loss of 223 kb in one region (Deletion 1, D1) completely restored symbiotic effectiveness with M. sativa , whereas the deletion of the second region (Deletion 2, D2) had no effect (Figs. 1E and 1F). Neither deletion significantly impacted symbiosis with M. truncatula ( Supplemental Figure S6 ). We concluded that dominant genetic determinant(s) within the D1 region of the HM006 pSymA led to the collapse in effective symbiosis with M. sativa . Additive inhibition of symbiotic effectiveness by a duplicated gene cluster on HM006 pSymA. To identify genes within the 223 kb region removed in D1 responsible for reduced symbiotic N 2 -fixation with M. sativa , we generated nine sub-deletions and screened them for their symbiotic productivity with M. sativa (Figs. 2A and 2B). While none of the deletions fully restored symbiotic productivity, two sub-deletions, D7 and D9, each resulted in a ~ 50% recovery in shoot dry weight production (Figs. 2A, 2B, and 2C). Further sub-deletions within the 60 kb region removed in D7 (D14-D16) and the 35 kb region removed in D9 (D17-D19) localized the genes responsible for the partial recovery of M. sativa shoot dry weight production to two gene sets deleted in D15 and D18 (Figs. 2A and 2D). Both genetic loci, which we name pha-1 (removed in D15, locus tags CDO22_RS21580-CDO22_RS21610) and pha-2 (removed in D18, locus tags CDO22_RS21900-CDO22_RS21925), show an operon-like configuration with a 99.9% identity to one-another over a majority of their length (Fig. 2E). Both clusters included genes annotated as encoding an iron-containing alcohol dehydrogenase (TIGR02638 E-value 1.70e-60) and two subunits of a Class III polyhydroxyalkanoic acid synthase: PhaE (TIGR01834 E-value 1.33e-08) and PhaC (TIGR01836 E-value 2.94e-113/3.25e-85) in an operon-like configuration. Homology between the loci extends 2783 bp upstream of these gene sets and contains two annotated pseudogenes along with, presumably, the promoter. A putative operon in pha-1 extends beyond the duplicated region and included genes predicted to encode a Coenzyme A transferase (TIGR02428 E-value 8.25e-11) and an acetyl-CoA synthetase (TIGR02188 E-value 0). Figure 2. Relative locations and symbiotic effectiveness phenotypes of sub-deletions spanning the 223 kb region removed in D1. ( A ) Map displaying the relative locations of genes and approximate lengths of deletions across the region. ( B ) Quantitation of shoot dry weight accumulation by M. sativa following inoculation with the wild type, the pSymA hybrid, D1, and the sub-deletions of D1: D3-D11. ( C ) Representative shoot and nodules images of the partial recovery phenotype resulting from Deletions 7 and 9. Scale bars represent 5 cm for shoot and 0.5 cm for nodule images. ( D ) Quantitation of shoot dry weight accumulation by M. sativa following inoculation with the HM006 pSymA hybrid, and sub-deletions of D7 (D14-D16) and D9 (D17-D19). ( E ) Genetic diagram of the duplicated region found in both D15 and D18 and the annotations of predicted open reading frames within the putative operons. Symbiotic carbon polymer synthesis leads to a host-specific inhibition of productive symbiosis with M. sativa. Based on their annotation, we hypothesized that the gene sets in pha-1 and pha-2 may be involved in the production of a polyhydroxyalkanoate (PHA) carbon polymer such as polyhydroxybutyrate (PHB) during symbiosis. PHB is visualized by transmission electron microscopy (TEM) as large granules within bacteroids of rhizobia that synthesize PHB during symbiosis ( 35 ). TEM of M. sativa and M. truncatula nodules revealed the presence of large granules in the strains bearing pSymA-HM006. These granules were absent from bacteroids of the wild type or the D1 strain lacking both pha-1 and pha-2 (Fig. 3 A, Supplemental Figure S7 ). While the small inefficient M. sativa nodules restricted further characterization, we also measured PHA production in effective M. truncatula nodules using flow cytometry, using PHA-bound Nile Red for quantitation ( 36 ) (Fig. 3 B). A significant increase in Nile Red fluorescence of bacteroids was observed in wildtype HM006 and the HM006-pSymA hybrid strains relative to wildtype RmP110 (Fig. 3 B). The increased fluorescence was significantly reduced in D1, which lacked both pha-1 and pha-2 , as well as D15 which lacked the expanded gene set of pha-1 (Fig. 3 B). To examine the expression of the PHA synthesis genes during symbiosis, we analyzed the transcriptome of the HM006 pSymA hybrid strain in M. sativa and M. truncatula root nodules. RNAseq showed a high level of transcription across the regions that include the PHA synthesis genes in both hosts (Fig. 3 C, Supplemental Figure S8 ). Several genes within them were among the top 50 most highly expressed genes from HM006 pSymA during symbiosis ( Supplemental Table S10 ). We further observed robust pha cluster gene expression in nitrogen-fixing bacteroids of nodules using gusA promoter fusion constructs (Fig. 3 D, Supplemental Figure S9 ). Altogether, these data indicate that the production of carbon polymers during symbiosis by bacteroids containing the newly discovered PHA genes drives a host-specific symbiotic incompatibility between S. meliloti and M. sativa. Expression of the pha genes alone leads to a complete symbiotic collapse in a naïve background that is relieved by local catabolism genes. To independently verify that the genes present in the pha loci are the sole components of the HM006 pSymA replicon responsible for the reduced symbiotic N 2 -fixation in M. sativa nodules, we applied a gain-of-function approach. We integrated the putative symbiotic PHA synthesis operon from pha-1 (Figs. 2A and 2E) into the genome of RmP110 with its native pSymA. Strikingly, RmP110 with pha-1 alone resulted in a more complete collapse of symbiosis than that associated with pSymA-HM006 (Fig. 4 A). Shoot dry weight production was similar to uninoculated control plants (Fig. 4 A), with few pink nodules (Fig. 4 E) and barely detectable acetylene reduction activity ( Supplemental Figure S10 ). TEM analysis showed that nodules were sparsely populated with bacteroids which showed signs of degradation by the host plant ( 37 ) ( Supplemental Figure S11 ). To rationalize the discordance between the partially reduced symbiotic efficiency conferred by the complete HM006 pSymA, and the near non-functional symbiosis rendered by the pha gene cluster alone, we hypothesized that other genetic determinants on HM006 pSymA alleviate the inhibition of symbiosis caused by the pha cluster. In search of such determinants, we identified other strains bearing phaC homologues by BLAST ( Supplemental Table S11 ) and compared the content of the local genetic region surrounding phaC (Fig. 4 C). These included the well-studied pea symbiont Rhizobium johnstonii (formerly Rhizobium leguminosarum ) Rlv3841 as well as S. meliloti USDA1963. We found that a set of carbon metabolism genes previously characterized in A. tumefaciens C58 as involved in N-acyl-homoserine lactone catabolism ( attKLM ) were consistently collocated with the pha cluster in other strains ( 38 ) (Fig. 4 C), and we considered they could be responsible for alleviating the inhibition of symbiosis caused by the pha cluster alone when they were transferred with HM006 pSymA. We cloned the catabolism gene set and introduced them to the wild type that contained the pha gene cluster. Introduction of the catabolism genes was sufficient to restore functional symbiosis (Fix + phenotype) to the wildtype background characterized by increased shoot dry weight and pink nodule number (Figs. 4 D, 4 E, and 4 F). Consistent with these data, Deletion 8 of HM006 pSymA which lacked these genes was more symbiotically impaired than other deletions (Fig. 2B). Evidence for a novel symbiotic poly(gamma)hydroxybutyrate cycle. We considered that the recovery of symbiosis mediated by the catabolism genes may be the result of a PHA cycle, wherein the pha synthesis genes produce a polymer, whereas the att catabolism genes are able to break it down. Based on the annotation of the function of homologues to the att genes in A. tumefaciens (Fig. 5 A), we predicted a novel pathway for polymer synthesis and catabolism based on the intermediate gamma-hydroxybutyrate (Fig. 5 B). The proposed pathway places the synthesis genes in a rational scheme for polymer production based on their annotations; oxidation of succinate semialdehyde to gamma-hydroxybutyrate by the iron-containing dehydrogenase, the addition of an acetyl-CoA group by the CoA transferase generating GHB-CoA (with acetyl-CoA substrate generated by acetyl-CoA synthetase), and polymerization of GHB-CoA monomers via the PhaE/PhaC PHB synthase. The breakdown of the P4HB product of the proposed synthesis pathway then fits with the characterized activities of homologues to AttL and AttK, wherein they break down GHB to succinate, fueling the TCA cycle and bacteroid nitrogen-fixation. In this scheme, the lactonase AttM could conceivably resolve monomers from a GHB polymer with a similar hydration reaction to the cleaving of the lactone ring in gamma-hydroxybutyrolactone ( 38 ). To provide evidence for the proposed pathway, we performed metabolomics of root nodules formed by rhizobia bearing or lacking these pathways, looking for the presence and abundance of pathway intermediates. Consistent with the pathway scheme above, we found significant increase of both the predicted pathway intermediates, SSA and GHB (Fig. 5 C, Supplemental Figure S12 ). Conversion of succinate semialdehyde to gamma-hydroxybutyrate by an iron-containing dehydrogenase in bacteroids is the mechanism for symbiotic collapse in M. sativa . To gain further insight into the role of individual genes in the pathway, we constructed knockout mutations within genes from each step of the predicted pathway ( ghbD , phaC2 , ghbT , and acs ), with double knockouts created for the duplicated genes ghbD and phaC2 . Strikingly, only a double knockout within the iron-containing dehydrogenase ghbD (the predicted first step of the synthesis pathway) was able to recover symbiosis (Figs. 6 A and 6 B). These data suggested that the first step of the predicted pathway, which we predicted to direct carbon from SSA away from succinate in a reductant consuming reaction, rather than polymer synthesis per se, was responsible for symbiotic collapse. To ensure these data were not an artifact of working in the HM006 pSymA hybrid background, we also constructed double-knockouts of the iron-containing dehydrogenase in HM006 and showed a similar recovery of symbiosis with M. sativa (Figs. 6 C and 6 D), coincident with a reduction in carbon polymer synthesis observed by flow cytometry (Fig. 6 E, Supplemental Figure S13 ). Since GhbD, the iron-containing dehydrogenase, appeared the critical element of symbiotic collapse, we wished to verify its hypothesized activity in the predicted P4HB synthesis pathway (Fig. 5 B). We purified the iron-containing dehydrogenase and demonstrated its activity with SSA as a substrate ( Supplemental Figures S14 and S15 ). Purified GhbD showed a Km of 0.1129 or 0.7708 and a Vmax of 1.891 or 2.469 in an NADH/NADPH linked spectrophotometric assay when incubated with SSA and NADH or NADPH, respectively, indicating a higher affinity for NADH as a cofactor (Figs. 7 A and 7 B). The loss-of-function data combined with the verified activity of GhbD suggested that the primary mechanism for symbiotic collapse is the biochemical conversion of SSA to GHB in bacteroids rather than polymer synthesis, since phaC double mutants retained symbiotic collapse phenotype (Figs. 6 A and 6 B). To further support this presumption, we expressed the dehydrogenase in the naïve wildtype background, without the downstream portions of the operon predicted to be involved in GHB polymerization (Figs. 7 D to 7 F). We found this construct was still sufficient for the collapse of M. sativa symbiosis previously observed by transferring the whole operon (Fig. 4 ). Based on the confirmed biochemical activity of GhbD (Figs. 7 A and 7 B), we further predicted we could rescue the symbiotic phenotype caused by ghbD expression by overexpressing attK from the catabolism gene set which in Agrobacterium is involved in the conversion of SSA into succinate in a reductant generating reaction ( 38 ) (Fig. 7 C). Such an activity would compete with the conversion of SSA to GHB in a reductant-consuming reaction by GhbD (Fig. 7 C). Indeed, we found that including the attK gene downstream of ghbD in the gain-of-function construct significantly restored symbiosis between S. meliloti and M. sativa (Figs. 7 D to 7 F). Discussion Replicon transplantation experiments led to the finding that the synthesis of GHB-based hydrocarbon polymers is linked to the collapse of symbiotic effectiveness in M. sativa nodules leading to a reduction in partner quality. Several differences are notable between the carbon polymer production genes discovered here (Fig. 4 H) and the well-characterized polyhydroxybutyrate synthesis genes ( phbABC ). First, annotations of genes in the operon (e.g., the inclusion of an iron-dependent dehydrogenase) are inconsistent with those in the well-characterized PHB synthesis pathway in rhizobia ( 40 ). Second, the high level of expression of these genes during symbiosis contrasts with the typically described pattern of PHB synthesis in S. meliloti , where PHB is produced in the free-living state and during infection thread invasion but not during symbiotic nitrogen fixation ( 36, 41 ). The inclusions in bacteroids observed in electron micrographs did not show the typical bright granules due to a lack of staining by lead citrate as is often observed for PHB ( 35 ). Finally, the observation of GHB accumulation in M. sativa nodules occupied by rhizobia containing the new pha synthesis genes (Fig. 5 C), and the verification of GhbD as a dehydrogenase that reduces SSA to GHB (Fig. 7 ) further highlight the unique nature of this pathway. Altogether, these data support its existence as a novel symbiotic PHA synthesis pathway in rhizobia. We further supply evidence that the att catabolism genes impact the symbiotic phenotype associated with symbiotic PHA synthesis and propose that the pha and att clusters act together in a symbiotic P4HB cycle (Fig. 5 B). A likely source of SSA, the substrate for the first step in the proposed synthesis pathway, is gamma-aminobutyrate (GABA) catabolism, which is provided to bacteroids from plant cells and catabolised to SSA via highly redundant transamination reactions ( 42–44 ). It has been suggested that the production of reduced carbon storage polymers such as PHB may negatively impact the symbiosis by diverting resources away from nitrogen fixation ( 7 ). If the M. sativa phenotypes we document in this study was the result of diversion of resources from symbiosis via PHA synthesis, they would be by far the most dramatic observation of this phenomenon to date. To investigate the plausibility of this, we utilized an available host/symbiont metabolic model (VINE ( 45 )) to investigate the degree to which the reduction in M. sativa partner quality could be explained by resource consumption (either reductant or carbon) by symbiotic PHA synthesis ( Supplemental Fig. 16 ). While the modelling was consistent with resource consumption negatively impacting nitrogen fixation, the magnitude of reduction in shoot dry weight accumulation predicted by the model was significantly less than observed in M. sativa experimentally. We thus conclude that the observed symbiotic collapse cannot be explain simply due to diversion of reductant or carbon away from nitrogen fixation. Instead, we hypothesize that the accumulation of GHB intermediates during PHA polymer synthesis in M. sativa nodules is toxic to Sinorhizobium bacteroids, resulting in a loss of nitrogen fixation capacity. This is supported by multiple observations. First, only mutations in the pha loci that abolish ghbD function enable the recovery of symbiosis in M. sativa (Fig. 6 A). Similarly, expressing GhbD alone (converting SSA to GHB) in a naïve background resulted in a substantial reduction in symbiosis in M. truncatula , which was partially rescued by co-expressing AttL which diverts SSA to succinate ( Supplemental Figure S18 ). Both of these observations suggest that synthesis of the pathway intermediates (rather than the polymer itself) was sufficient to collapse the symbiosis. Lastly, even though large PHA granules accumulate in bacteroids within both M. sativa and M. truncatula nodules, a substantial increase in GHB and SSA intermediates was only observed in M. sativa nodules ( Supplemental Figure S17 ), consistent with our observation of a strong symbiotic phenotype only in this host. The production of reduced carbon polymers such as PHB have been implicated in rhizobium “cheating”, in which the rhizobium benefits (storage of carbon for use once released from the nodule) at the expense of the host (reduced rate of nitrogen fixation) ( 7 ). However, it is not clear whether carbon polymer synthesis is a cheating behavior. Although PHB mutants showed increased nitrogen fixation in some rhizobia ( 46 ), limited effects were observed in others ( 35, 41, 46, 47 ). In fact, PHB production and other lipogenesis represent important reductant stores during symbiotic nitrogen fixation and therefore may facilitate rather than limit nitrogen-fixation capacity ( 48 ). In this study, removing the pha gene sets from HM006 pSymA did not significantly impact partner quality with M. truncatula ( Supplemental Figure S6 ), even though the bacteroids within M. truncatula nodules did accumulate large PHA granules (Figs. 3 B and 3 D, Supplemental Figures S7, S8 and S9 ), suggesting that the polymer synthesis was not at the expense of the host. While P4HB synthesis in bacteriods came at a large cost to M. sativa , the proposed toxicity of the intermediates to the bacteriods and the complete collapse of the symbiosis likely means that P4HB synthesis hurt, rather than benefited, S. meliloti . Collectively, these results suggest that at least on the tested plants, P4HB synthesis is unlikely to represent a cheating behavior. Presumably, the pha clusters confer some benefit to the rhizobia and/or the host to maintain them in rhizobium populations despite the potential for limiting partner quality on some hosts. PHA clusters have been theorized to benefit the symbiosis by providing a source of carbon to fuel SNF during carbon-limited scenarios ( 49 ), and the cycle we propose here would be particularly well suited to this purpose by engaging with central metabolism through dicarboxylates (unlike the canonical PHB catabolism pathway ( 50 )), which must allow SNF in the nodule to drive ammonium excretion to the plant ( 51 ). PHAs accumulated in nodules could also benefit rhizobia outside of the nodule where they often encounter carbon-limited soil environments ( 36, 51 ). However, as bacteriods are terminally differentiated in Medicago nodules and thus cannot regrow in the soil, it may be that PHAs accumulated in bacteriods do not benefit free-living rhizobia. Alternatively, selection for maintenance of the pha clusters may be independent of the symbiosis, while their expression in bacteriods is non-beneficial and the result of inappropriate activation following detection of plant-produced signals. Though the benefits conferred by the pha clusters to the microsymbiont and/or host remain to be elucidated, we expect that coevolution of these clusters in rhizobia with specific hosts can tune their activity to limit their negative impact on symbiosis, for example, through ensuring rapid flux through toxic pathway intermediates. Indeed, HM006 was isolated from M. truncatula nodules grown with soil from its native range ( 33, 52 ). When HM006 is present in M. truncatula , although the genes are highly expressed, we observes different gene expression patterns across the region relative to HM006 bacteroids in M. sativa nodules ( Supplemental Figure S8 ), which may reflect the ability to fine tune the pathways through adaptation of transcriptional control. In addition, other genetic elements in the loci may participate in limiting the impact to specific hosts (Fig. 4 C). Partner quality differences in nature are widespread in many if not all rhizobium/legume systems ( 53–55 ). The replicon transplantation experiments described in this report revealed a correlation of partner quality phenotypes between M. sativa and the pSymA replicons of the tested strains. Thus pSymA (and perhaps the mobile symbiotic element in many rhizobia) likely represents the primary reservoir for accessary genes influencing partner quality in Sinorhizobium , although accessory genes on pSymB or additional plasmids can also influence symbiotic outcomes ( 21, 56 ). Given the much higher degree of genetic variation on extrachromosomal replicons relative to the chromosome ( 11, 22, 32, 57, 58 ), it is unsurprising these elements contain important determinants of the partner quality differences observed across different strains. Certainly, many mechanisms exist for the diversity in partner quality phenotypes observed between rhizobia and their different hosts, distributed through a reshuffling of genetic modules that engender metabolic flexibility but also can render individual partnerships inefficient ( 12, 59 ). This study suggests that along with diversity in symbiotic and immune signaling ( 60–63 ), metabolic incompatibility may be a key theme governing the diversity of these interactions we observe in nature. Materials and Methods Genetic Manipulations Bacterial strains and plasmids used in this work are listed in Supplemental Table S1 . Media used for routine growth and selection of microbes included LB for Escherichia coli , LBmc for S. meliloti , and YPD or SC defined media for Saccharomyces cereviciae . Antibiotics or supplements were added at routine concentrations when appropriate and details are summarized in the Supplemental Methods . We routinely used a triparental mating protocol to transfer plasmids between donor and recipient strains by conjugation using the E. coli helper strain MT616 to drive mobilization ( 16 ). Transfer of extrachromosomal replicons between S. meliloti strains also utilized conjugation; either via an integrated oriT with triparental mating, or through the use of rctB in the donor strain to drive biparental conjugation ( 1 ). Deletions were made by FLP- FRT recombination ( 64 ), wherein FRT sites were integrated flanking a target region by homologous recombination, and subsequent excision of the region was catalyzed by introduction of FLP recombinase. Alternatively, region capture from S. meliloti involved a quadriparental mating with S. meliloti donor, E. coli helper, E. coli with a FLP-recombinase plasmid, and an E. coli recipient. FLP- FRT recombination was also used to introduce gene clusters into the genome by catalyzing recombination between an FRT site in a plasmid containing the cluster and a single FRT site in a landing pad in the genome ( 14 ). We used Φ M12 bacteriophage to transfer insertions between S. meliloti backgrounds using transduction ( 49 ). Selection of desired events involved the use of antibiotics as well as prototrophic selection. Further details of genetic manipulations are included in Supplemental Methods. Verification of genetic modifications was performed by PCR using primers in Supplemental Tables S4 to S7 or by whole genome sequencing. Plasmid Construction Plasmids for homologous recombination were routinely assembled using Level 1 (BsaI) Golden Gate cloning into suicide destination vectors that follow the BEVA architecture ( 66 ). Inserts were PCR amplified with primers found in Supplemental Table S3. Plasmids for gene cluster integration were assembled by yeast recombineering as previously described ( 66, 67 ). Primers for amplification of regions that were combined in the assembly are in Supplemental Table S8. Further details of plasmid construction are available in Supplemental Methods. Plasmids were verified by restriction digest or whole plasmid sequencing. Symbiotic Assays Plant materials used in this work included Medicago sativa cv. Iroquois (alfalfa) and Medicago truncatula A17 (barrel medic). Plant assays were conducted in Leonard jar assembles with a 1:1 (wt:wt) mixture of sand and vermiculite with 250 mL of Jensen’s medium ( 68 ). Four to six seedlings were added to each Leonard jar and inoculated two days after with 10 mL of sterile water containing 100 µL of dense overnight culture. M. sativa seeds were surface sterilized for five minutes with 95% ethanol and for 20 minutes with 2.5% sodium hypochlorite. Seeds were then rinsed with sterile water for one hour, with water replaced every 15 minutes. M. sativa seeds were placed on 1.5% water agar, placed in the dark at room temperature, and allowed two days to germinate. M. truncatula seeds were first scarified with sulfuric acid for 10 minutes. After scarifying, seeds were rinsed 10 times with sterile water and then proceeded to be surface sterilized with 2.5% sodium hypochlorite for two minutes. Seeds were rinsed another 10 times with sterile water. M. truncatula seeds were placed on 1.5% water agar and placed in the dark at 4°C for three days. The seeds were then moved to room temperature and allowed two more days to germinate. Four to six seedlings were added to each Leonard jar and inoculated two days after with 10 mL of sterile water containing 100 µL of dense overnight culture. Plants were grown in Conviron Gen1000/2000 growth chambers programmed with a day cycle of 18 hours at 21°C with maximum light followed by a night cycle of six hours at 17°C with no light. Plants were allowed 42 days post-inoculation to grow. After the 42-day period, shoots were cut from roots and placed in a drying oven for one week, after which they were weighed for shoot dry weight. Fix + (pink) nodules were picked from roots, counted, and weighed immediately for average nodule fresh weight. Collected nodules were placed in an air-tight container with desiccant packets and let dry for a week, after which they were weighed for nodule dry weight. In some experiments HP6890 gas chromatograph was used to perform gas chromatography for acetylene reduction assays as previously described ( 69 ). This assay was performed immediately after shoots were detached from the root system. A summary of all plant data collected is included in a Supplemental Data file. Flow Cytometry Nodules pooled from four plants from one pot were macerated in 750 µL of phosphate-ascorbate buffer (PAB) using a mortar and pestle. An additional 750 µL of PAB was used to rinse and collect any additional bacteroids from the mortar. After vortexing briefly, samples were spun at 200g for 5 minutes to spin down residual plant debris. The resulting supernatant was then pelleted and resuspended in 100 µL of fixative (30% molecular-grade ethanol). After 15 minutes of incubation, samples were pelleted to remove the fixative and resuspended in 500 µL of PBS. A portion of each sample was pooled to create the three control samples: unstained, DAPI-only, and Nile Red only. Prior to staining, samples were diluted 1:10 in PBS. 1 µL of 10% Nile Red in DMSO and 1 µL of DAPI (1 mg/mL in PBS) were added to samples and allowed 15 minutes to incubate in darkness. Following staining, samples were again diluted 1:10 in PBS, giving a final 1:100 sample dilution. Cells were analyzed on a Beckman Cytoflex S flow cytometer equipped with violet, yellow, red, and blue lasers. The 561 nm yellow laser was used to excite Nile Red, and fluorescence was detected using the 610/20 bandpass filter. DAPI was excited using the 405nm violet laser and detected with the 450/45 bandpass filter. DAPI-positive bacteroids were gated and analyzed for Nile Red fluorescence intensity. Details on the gating schemes are described in Supplemental Figures S19 and S20 . Electron Microscopy Specimens for transmission electron microscopy were fixed in 2.5% glutaraldehyde in sodium phosphate buffer, postfixed in 2% osmium tetroxide, dehydrated through a graded acetone series including saturated uranyl acetate in 70% acetone, and embedded in Spurr’s resin polymerized at 60 O C for 24 hours. Embedded samples were sectioned at 70–90 nm thickness on an RMC MT XL ultramicrotome. Ultrathin sections were collected on copper grids with a formvar-carbon supporting film, then stained with Reynolds’ lead citrate. Observation and imaging were performed on a JEOL JEM-1400 electron microscope operating at 120 kV and equipped with an AMT NanoSprint 15L bottom-mount camera. GUS Staining and Microscopic Observations The promoter regions of the pha clusters, nifH , and nodA were cloned via Golden Gate Cloning. Plasmids and strains used are listed in Supplemental Table S1 . Primers used for promoter region amplification are listed in Supplemental Table S3 . The nodules were fixed in 100 mM sodium-phosphate buffer with 0.1% Triton X-100 and then in 1.5% glutaraldehyde in 100 mM phosphate buffer under vacuum for 30 minutes and 90 minutes. After that, the nodules were embedded in 3% (w/v) low-melting agarose and sectioned into 100-µm slices using vibrating microtome VF-210-0Z (Precisionary, USA). The slices were then vacuum immersed in GUS Staining Solution for 30 min. The solution contains 2 mM 5-bromo-4-chloro-3-indoxyl-β-d-glucuronide cyclohexyl ammonium salt (X-Gluc), 0.2% Triton X-100, and 2 mM each of K3[Fe(CN)6] and K4[Fe(CN)6] in 50 mM sodium phosphate buffer (pH 7.2). After incubating at 37°C for 2 hours, the stained nodule slices were observed using a BioTek Cytation 5 Cell Imaging Multimode Reader (Agilent, USA). Enzyme Assays A NADPH/NADH colorimetric assay was used to determine the Km and Vmax values of GhbD reacting with succinate semialdehyde (SSA). Individual 100 µL reactions comprised of purified GhbD (1 µg), NADPH (1 mM) or NADH (1 mM), SSA (0, 0.125, 0.25, 0.5, 1, 2.5, 5, or 10 mM), and reaction buffer (50 mM Na 3 PO 4 pH 7.0, 200 mM NaCl, 10 mM MgCl 2 ). The 340 nm absorbance value of every reaction occurring at 28°C was measured in 1-minute intervals for 11 minutes starting immediately after the addition of GhbD. Absorbance values were read using a BioTek Cytation 5 Cell Imaging Multimode Reader by Agilent. Simple linear regression tests were completed with absorbance values over time for specific substrate concentrations. The resulting slopes, with a R 2 value > 0.95, were used to determine the rate of reaction per mg of GhbD present. These rates, at different substrate concentrations, were used in the Michaelis-Menten equation to determine the Km and Vmax values for GhbD with SSA and NADPH/NADH. GraphPad Prism Version 10.4.1 was used to perform simple linear regression and non-linear regression tests. Metabolomics Metabolomics analysis of plant tissue samples was performed using a 4-channel chemical isotope labeling (CIL) LC-MS method ( 70 ). Normalization and labeling kits were obtained from Nova Medical Testing Inc. (Edmonton, AB, Canada). Each tissue sample was homogenized using a bead beater with steel beads, followed by the addition of ice-cold methanol/water (4:1, v/v) for metabolite extraction. The resulting extracts were normalized and labeled according to the protocols provided with the CIL kits. LC-MS analysis was carried out using a Thermo Vanquish UHPLC system (ThermoFisher, Waltham, MA, US) coupled to a Bruker Impact II QTOF mass spectrometer (Bremen, Germany). Chromatographic separation was performed on an Agilent Eclipse Plus C18 reversed-phase column (150 × 2.1 mm, 1.8 µm particle size) maintained at 40°C. The mobile phases consisted of 0.1% (v/v) formic acid in water (A) and 0.1% (v/v) formic acid in acetonitrile (B). The gradient program was as follows: 0 min, 25% B; 10 min, 99% B; 15 min, 99% B; 15.1 min, 25% B; 18 min, 25% B. The flow rate was 400 µL/min. Mass spectra were acquired at 1 Hz over a mass range of m/z 220–1000. LC-MS data were processed using DataAnalysis (Bruker Daltonics) and IsoMS Pro (Nova Medical Testing Inc.). Manual verification of the key compound gamma-hydroxybutyrate (GHB) was performed using an authentic standard on an alternative instrumental platform. Specifically, MS/MS spectra and retention time (RT) data from the samples were compared to those of the standard, both acquired on a Dionex 3000 LC system coupled with Orbitrap Q Exactive HF mass spectrometer (ThermoFisher, Waltham, MA, US), using the same LC method and column. The MS/MS spectra of GHB in the samples, collected at multiple collision energies, matched those of the authentic standard, confirming the compound’s identity in the sample. DNA and RNA sequencing Genome resequencing of hybrid strains and nodule RNA sequencing was performed by the Joint Genome Institute using Illumina NovaSeq technology. For RNA sequencing, bacteroids were isolated from nodules collected directly onto dry ice and flesh frozen in liquid nitrogen from one pot per replicate. RNA was extracted using a QIAgen RNeasy Plus Mini Kit, and underwent on-column DNase treatment. Libraries were prepared using Illumina’s Ribo-Zero rRNA removal kit and TruSeq Standard Total RNA HT sample prep kit. Raw fastq reads were filtered and trimmed according to the JGI QC pipeline, and mapped to the reference genome. Raw read counts were normalized to determine transcripts per million for each gene in the reference. Further details of DNA and RNA extraction, sequencing and bioinformatics analysis are included in Supplemental Methods. Raw data was deposited in NCBI SRA using accessions in Supplemental Table S13. Routine whole plasmid and whole genome resequencing to verify constructs and genetic manipulations was performed by Plasmidsaurus using Oxford Nanopore technology. Comparative Genomics Analysis of homology between genes within clusters of interest was performed using clinker (Min. alignment sequence identity 0.3) and visualized with clustermap.js (> 0.5 alignment sequence identity shown) on the CAGECAT web server https://cagecat.bioinformatics.nl/ ( 71 ). Whole genome alignments and delineation of locally colinear gene blocks were performed using the MAUVE ( 72 ) plugin in the Geneious software with default parameters (automatically calculate seed weight and minimum LCB score with MUSCLE3.6 gapped aligner). Data Collection and Statistical Analysis Plant experiments included uninoculated controls that were screened for the absence of nodules to confirm lack of contamination before any data collection. Individual replicates that makeup datapoints or were used for RNAseq or flow cytometry consisted of the average of all plants in one pot (six for M. sativa and four for M. truncatula ). For statistical analysis we used a one-way ANOVA test with Dunnett’s multiple comparison test to establish significant differences between replicates. All data shown was significant as determined by one-way ANOVAs. In the figures, results from post hoc significance testing are indicated as * P < 0.0332, ** P < 0.0021, *** P < 0.0002, **** P < 0.0001. Declarations Acknowledgements and Funding We are grateful to Michael Sadowsky for the gift of S. meliloti strains. We thank Megan Ramsett for research and logistical support throughout the project. We thank Christa Pennachio for project coordination and Daniel Peterson (Joint Genome Institute) for RNAseq data analysis associated with JGI New Investigator proposal:10.46936/10.25585/60000503. We thank Scott Payne and Jayma Moore (NDSU Electron Microscopy Center core facility) for technical support with electron microscopy. We are grateful to the Geddes Lab and members of the Department of Microbiological Sciences, NDSU for critical review of the manuscript. This work was supported by startup funding provided to B. A. Geddes from the ND Agricultural Experiment Station, a New Innovator in Food & Agricultural Research (FFAR) grant to B. A. Geddes ID: FF-NIA21-0000000061, a Collaborative Research-PGR grant (NSF) to B. A. Geddes ID: Award number 2243818, an NSERC Discovery grant to T. M. Finan, and a NIFA Predoctoral Fellowship to G. J. Levin Grant Number: 2024-67011-42916. DNA and RNA sequencing work (proposal:10.46936/10.25585/60000503) conducted by the U.S. Department of Energy Joint Genome Institute, a DOE Office of Science User Facility, is supported by the Office of Science of the U.S. Department of Energy operated under Contract No. DE-AC02-05CH11231. Electron microscopy work reported in this publication was supported by the Office of Research Infrastructure Programs of the National Institutes of Health under award number S10OD030309. Competing Interests Barney Geddes is a cofounder of Lilac Agriculture, Inc. Author Contributions G. J. L., B. A. G., and T. M. F. contributed to experimental design. G. J. L., N. V., C. X. L., S. H., A. L., S. Z., and B. A. G. performed the experiments. G. J. L., B. A. G., S. H., S. Z., L. L., G. C. dC and T. M. F wrote the manuscript. T. M. F. and B. A. G. supervised the work. References A. J. Weisberg, et al. , Pangenome Evolution Reconciles Robustness and Instability of Rhizobial Symbiosis. mBio 13 , e00074-22 (2022). S. S. Porter, E. L. Simms, Selection for Cheating Across Disparate Environments in the Legume-Rhizobium Mutualism. Ecology Letters 17 , 1121–1129 (2014). K. D. Heath, J. R. Stinchcombe, Explaining Mutualism Variation: A New Evolutionary Paradox? Evolution 68 , 309–317 (2014). E. W. Triplett, M. J. Sadowsky, Genetics of Competition for Nodulation of Legumes. Annu. Rev. Microbiol. 46 , 399–428 (1992). L. T. Burghardt, G. C. diCenzo, The Evolutionary Ecology of Rhizobia: Multiple Facets of Competition Before, During, and After Symbiosis with Legumes. Current Opinion in Microbiology 72 (2023). M. Mendoza-Suárez, S. U. Andersen, P. S. Poole, C. Sánchez-Cañizares, Competition, Nodule Occupancy, and Persistence of Inoculant Strains: Key Factors in the Rhizobium-Legume Symbioses. Front. Plant Sci. 12 , 690567 (2021). R. F. Denison, Legume Sanctions and the Evolution of Symbiotic Cooperation by Rhizobia. The American Naturalist 156 , 567–576 (2000). A. Westhoek, et al. , Conditional Sanctioning in a Legume– Rhizobium Mutualism. Proc. Natl. Acad. Sci. U.S.A. 118 , e2025760118 (2021). E. T. Kiers, R. A. Rousseau, S. A. West, R. F. Denison, Host Sanctions and the Legume– Rhizobium Mutualism. Nature 425 , 78–81 (2003). A. P. Montoya, et al. , Hosts Winnow Symbionts with Multiple Layers of Absolute and Conditional Discrimination Mechanisms. Proc. R. Soc. B. 290 , 20222153 (2023). A. B. Riley, M. A. Grillo, B. Epstein, P. Tiffin, K. D. Heath, Discordant Population Structure Among Rhizobium Divided Genomes and Their Legume Hosts. Molecular Ecology 32 , 2646–2659 (2023). M. R. Riaz, et al. , Mobile Gene Clusters and Coexpressed Plant–Rhizobium Pathways Drive Partner Quality Variation in Symbiosis. Proc. Natl. Acad. Sci. U.S.A. 122 , e2411831122 (2025). G. E. D. Oldroyd, J. D. Murray, P. S. Poole, J. A. Downie, The Rules of Engagement in the Legume-Rhizobial Symbiosis. Annu. Rev. Genet. 45 , 119–144 (2011). B. A. Geddes, et al. , Minimal Gene Set from Sinorhizobium ( Ensifer ) meliloti pSymA Required for Efficient Symbiosis with Medicago . Proc. Natl. Acad. Sci. U.S.A. 118 , e2018015118 (2021). G. C. diCenzo, A. M. MacLean, B. Milunovic, G. B. Golding, T. M. Finan, Examination of Prokaryotic Multipartite Genome Evolution through Experimental Genome Reduction. PLoS Genet 10 , e1004742 (2014). T. M. Finan, B. Kunkel, G. F. De Vos, E. R. Signer, Second Symbiotic Megaplasmid in Rhizobium meliloti Carrying Exopolysaccharide and Thiamine Synthesis Genes. J Bacteriol 167 , 66–72 (1986). Y. Kawaharada, et al. , Receptor-Mediated Exopolysaccharide Perception Controls Bacterial Infection. Nature 523 , 308–312 (2015). J. L. Zimmerman, W. W. Szeto, F. M. Ausubel, Molecular Characterization of Tn5-Induced Symbiotic (Fix-) Mutants of Rhizobium meliloti . J Bacteriol 156 , 1025–1034 (1983). H. M. Meade, S. R. Long, G. B. Ruvkun, S. E. Brown, F. M. Ausubel, Physical and Genetic Characterization of Symbiotic and Auxotrophic Mutants of Rhizobium meliloti Induced by Transposon Tn5 Mutagenesis. J Bacteriol 149 , 114–122 (1982). S. Yang, F. Tang, M. Gao, H. B. Krishnan, H. Zhu, R Gene-Controlled Host Specificity in the Legume–Rhizobia Symbiosis. Proc. Natl. Acad. Sci. U.S.A. 107 , 18735–18740 (2010). J. Liu, et al. , Paired Medicago Receptors Mediate Broad-Spectrum Resistance to Nodulation by Sinorhizobium meliloti Carrying a Species-Specific Gene. Proc. Natl. Acad. Sci. U.S.A. 119 , e2214703119 (2022). M. Galardini, et al. , Exploring the Symbiotic Pangenome of the Nitrogen-Fixing Bacterium Sinorhizobium meliloti . BMC Genomics 12 , 235 (2011). L. T. Burghardt, et al. , Select and Resequence Reveals Relative Fitness of Bacteria in Symbiotic and Free-Living Environments. Proc. Natl. Acad. Sci. U.S.A. 115 , 2425–2430 (2018). M. A. Mendoza-Suárez, et al. , Optimizing Rhizobium- Legume Symbioses by Simultaneous Measurement of Rhizobial Competitiveness and N 2 Fixation in Nodules. Proc. Natl. Acad. Sci. U.S.A. 117 , 9822–9831 (2020). B. Epstein, et al. , Combining GWAS and Population Genomic Analyses to Characterize Coevolution in a Legume-Rhizobia Symbiosis. Molecular Ecology 32 , 3798–3811 (2023). B. Epstein, et al. , Genome-Wide Association Analyses in the Model Rhizobium Ensifer meliloti . mSphere 3 , e00386-18 (2018). R. T. Batstone, A. M. O’Brien, T. L. Harrison, M. E. Frederickson, Experimental Evolution Makes Microbes More Cooperative with Their Local Host Genotype. Science 370 , 476–478 (2020). I. J. Oresnik, S.-L. Liu, C. K. Yost, M. F. Hynes, Megaplasmid pRme2011a of Sinorhizobium meliloti Is Not Required for Viability. J Bacteriol 182 , 3582–3586 (2000). G. C. diCenzo, M. Zamani, B. Milunovic, T. M. Finan, Genomic Resources for Identification of the Minimal N 2 - Fixing Symbiotic Genome. Environmental Microbiology 18 , 2534–2547 (2016). A. Checcucci, et al. , Creation and Characterization of a Genomically Hybrid Strain in the Nitrogen-Fixing Symbiotic Bacterium Sinorhizobium meliloti . ACS Synth. Biol. 7 , 2365–2378 (2018). W. Reeve, et al. , Complete Genome Sequence of the Medicago Microsymbiont Ensifer (Sinorhizobium) medicae Strain WSM419. Standards in Genomic Sciences 2 , 77–86 (2012). M. Nelson, J. Guhlin, B. Epstein, P. Tiffin, M. J. Sadowsky, The Complete Replicons of 16 Ensifer meliloti Strains Offer Insights into Intra- and Inter-Replicon Gene Transfer, Transposon-Associated Loci, and Repeat Elements. Microbial Genomics 4 (2018). M. Sugawara, et al. , Comparative Genomics of the Core and Accessory Genomes of 48 Sinorhizobium Strains Comprising Five Genospecies. Genome Biol 14 , R17 (2013). Z.-C. Yuan, R. Zaheer, T. M. Finan, Regulation and Properties of PstSCAB, a High-Affinity, High-Velocity Phosphate Transport System of Sinorhizobium meliloti . J Bacteriol 188 , 1089–1102 (2006). E. M. Lodwig, et al. , Role of Polyhydroxybutyrate and Glycogen as Carbon Storage Compounds in Pea and Bean Bacteroids. MPMI 18 , 67–74 (2005). W. C. Ratcliff, S. V. Kadam, R. F. Denison, Poly-3-Hydroxybutyrate (PHB) Supports Survival and Reproduction in Starving Rhizobia: PHB Increases Rhizobium Fitness. FEMS Microbiology Ecology 65 , 391–399 (2008). A. Puppo, et al. , Legume Nodule Senescence: Roles for Redox and Hormone Signaling in the Orchestration of the Natural Aging Process. New Phytologist 165 , 683–701 (2005). Y. Chai, C. S. Tsai, H. Cho, S. C. Winans, Reconstitution of the Biochemical Activities of the AttJ Repressor and the AttK, AttL, and AttM Catabolic Enzymes of Agrobacterium tumefaciens . J Bacteriol 189 , 3674–3679 (2007). P. A. Price, et al. , Rhizobial Peptidase HrrP Cleaves Host-Encoded Signaling Peptides and Mediates Symbiotic Compatibility. Proc. Natl. Acad. Sci. U.S.A. 112 , 15244–15249 (2015). M. A. Trainer, T. C. Charles, The Role of PHB Metabolism in the Symbiosis of Rhizobia with Legumes. Appl Microbiol Biotechnol 71 , 377–386 (2006). C. Wang, et al. , Roles of Poly-3-Hydroxybutyrate (PHB) and Glycogen in Symbiosis of Sinorhizobium meliloti with Medicago sp. Microbiology 153 , 388–398 (2007). J. Prell, A. Bourdès, R. Karunakaran, M. Lopez-Gomez, P. Poole, Pathway of γ-Aminobutyrate Metabolism in Rhizobium leguminosarum 3841 and Its Role in Symbiosis. J Bacteriol 191 , 2177–2186 (2009). M. F. Dunn, Key Roles of Microsymbiont Amino Acid Metabolism in Rhizobia-Legume Interactions. Critical Reviews in Microbiology 41 , 411–451 (2015). C. E. Flores-Tinoco, et al. , Co-Catabolism of Arginine and Succinate Drives Symbiotic Nitrogen Fixation. Molecular Systems Biology 16 , e9419 (2020). G. C. diCenzo, M. Tesi, T. Pfau, A. Mengoni, M. Fondi, Genome-Scale Metabolic Reconstruction of the Symbiosis Between a Leguminous Plant and a Nitrogen-Fixing Bacterium. Nat Commun 11 , 2574 (2020). M. A. Cevallos, S. Encarnación, A. Leija, Y. Mora, J. Mora, Genetic and Physiological Characterization of a Rhizobium etli Mutant Strain Unable to Synthesize Poly-Beta-Hydroxybutyrate. J Bacteriol 178 , 1646–1654 (1996). J. I. Quelas, E. J. Mongiardini, J. Perez-Gimenez, G. Parisi, A. R. Lodeiro, Analysis of Two Polyhydroxyalkanoate Synthases in Bradyrhizobium japonicum USDA 110. Journal of Bacteriology 195 , 3145–3155 (2013). J. J. Terpolilli, et al. , Lipogenesis and Redox Balance in Nitrogen-Fixing Pea Bacteroids. J Bacteriol 198 , 2864–2875 (2016). F. J. Bergersen, G. L. Turner, Bacteroids from Soybean Root Nodules: Accumulation of Poly-β-Hydroxybutyrate During Supply of Malate and Succinate in Relation to N2 Fixation in Flow-Chamber Reactions. Proceedings of the Royal Society B: Biological Sciences 240 , 39–59 (1990). M. A. Trainer, T. C. Charles, The Role of PHB Metabolism in the Symbiosis of Rhizobia with Legumes. Appl Microbiol Biotechnol 71 , 377–386 (2006). C. C. M. Schulte, et al. , Metabolic Control of Nitrogen Fixation in Rhizobium-Legume Symbioses. Sci. Adv. 7 , eabh2433 (2021). K. A. Lesins, I. Lesins, Genus Medicago (Leguminosae): A Taxogenetic Study (Springer Science & Business Media, 2012). K. D. Heath, J. R. Stinchcombe, Explaining Mutualism Variation: A New Evolutionary Paradox? Evolution 68 , 309–317 (2014). S. A. West, E. T. Kiers, I. Pen, R. F. Denison, Sanctions and Mutualism Stability: When Should Less Beneficial Mutualists be Tolerated? J of Evolutionary Biology 15 , 830–837 (2002). L. Salvaudon, T. Giraud, J. A. Shykoff, Genetic Diversity in Natural Populations: A Fundamental Component of Plant–Microbe Interactions. Current Opinion in Plant Biology 11 , 135–143 (2008). P. A. Price, et al. , Rhizobial Peptidase HrrP Cleaves Host-Encoded Signaling Peptides and Mediates Symbiotic Compatibility. Proc. Natl. Acad. Sci. U.S.A. 112 , 15244–15249 (2015). X. Bailly, et al. , Population Genomics of Sinorhizobium medicae Based on Low-Coverage Sequencing of Sympatric Isolates. The ISME Journal 5 , 1722–1734 (2011). B. Epstein, M. J. Sadowsky, P. Tiffin, Selection on Horizontally Transferred and Duplicated Genes in Sinorhizobium (Ensifer), the Root-Nodule Symbionts of Medicago. Genome Biology and Evolution 6 , 1199–1209 (2014). A. J. Weisberg, et al. , Pangenome Evolution Reconciles Robustness and Instability of Rhizobial Symbiosis. mBio 13 , e00074-22 (2022). D. Wang, S. Yang, F. Tang, H. Zhu, Symbiosis Specificity in the Legume-Rhizobial Mutualism: Host Specificity in Root Nodule Symbiosis. Cellular Microbiology 14 , 334–342 (2012). S. S. Porter, S. E. Dupin, R. F. Denison, E. T. Kiers, J. L. Sachs, Host-Imposed Control Mechanisms in Legume–Rhizobia Symbiosis. Nat Microbiol 9 , 1929–1939 (2024). J. Liu, et al. , Paired Medicago Receptors Mediate Broad-Spectrum Resistance to Nodulation by Sinorhizobium meliloti Carrying a Species-Specific Gene. Proc. Natl. Acad. Sci. U.S.A. 119 , e2214703119 (2022). Z. Bozsoki, et al. , Ligand-Recognizing Motifs in Plant LysM Receptors are Major Determinants of Specificity. Science 369 , 663–670 (2020). B. Milunovic, G. C. diCenzo, R. A. Morton, T. M. Finan, Cell Growth Inhibition Upon Deletion of Four Toxin-Antitoxin Loci from the Megaplasmids of Sinorhizobium meliloti . J Bacteriol 196 , 811–824 (2014). T. M. Finan, et al. , General Transduction in Rhizobium meliloti . J Bacteriol 159 , 120–124 (1984). B. A. Geddes, M. A. Mendoza-Suárez, P. S. Poole, A Bacterial Expression Vector Archive (BEVA) for Flexible Modular Assembly of Golden Gate-Compatible Vectors. Front. Microbiol. 9 , 3345 (2019). R. D. Gietz, R. H. Schiestl, High-Efficiency Yeast Transformation Using the LiAc/SS Carrier DNA/PEG Method. Nat Protoc 2 , 31–34 (2007). J. M. Vincent, A Manual for the Practical Study of the Root-Nodule Bacteria (Blackwell Scientific Publications, 1970). M. Zamani, T. M. Finan, A. Cowie, G. C. diCenzo, Proline Auxotrophy in Sinorhizobium meliloti Results in a Plant-Specific Symbiotic Phenotype. Microbiology 161 , 2341–2351 (2015). S. Zhao, H. Li, W. Han, W. Chan, L. Li, Metabolomic Coverage of Chemical-Group-Submetabolome Analysis: Group Classification and Four-Channel Chemical Isotope Labeling LC-MS. Anal. Chem. 91 , 12108–12115 (2019). C. L. M. Gilchrist, Y.-H. Chooi, Clinker & clustermap.js: Automatic Generation of Gene Cluster Comparison Figures. Bioinformatics 37 , 2473–2475 (2021). A. C. E. Darling, B. Mau, F. R. Blattner, N. T. Perna, Mauve: Multiple Alignment of Conserved Genomic Sequence With Rearrangements. Genome Res. 14 , 1394–1403 (2004). Additional Declarations Yes there is potential Competing Interest. Barney Geddes is a cofounder of Lilac Agriculture, Inc. Supplementary Files SupplementaryInformation2.0Final9.25.25.pdf Supplementary Information Cite Share Download PDF Status: Under Review Version 1 posted You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-7715224","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Article","associatedPublications":[],"authors":[{"id":523913981,"identity":"bc1503be-ad0a-4a18-bbe1-9a1e4b201115","order_by":0,"name":"Barney Geddes","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAA+klEQVRIiWNgGAWjYBACAwQzAciuYGBsALF5iNdyhmQtjG1EaDFnP/vswcc2Brt+9uSND3/OuyM7f0YC44O3bbi1WPakmxvObGNIntnzrNiYd9sz4w03EpgN5+LRYnAgjU2aF6jF4EaOmTTjtsOJGyQSwCK4tZx/xib9F6jF/kaO+c+fcw4nAh3G/huvlhtAW4C+tjOQyDFj4G04nNhwI4GNGZ8WyxnP2CR7zkkkSJx5VizNc+yw8YYzD5sl55zDrcWcP41N4keZjT1/e/LGjz9qDsvOb08++OFNGW4tYMDIJpHYgMRtwKUQCfxhsCdC1SgYBaNgFIxUAACVZVT21kmGTAAAAABJRU5ErkJggg==","orcid":"https://orcid.org/0000-0001-8309-2083","institution":"North Dakota State University","correspondingAuthor":true,"prefix":"","firstName":"Barney","middleName":"","lastName":"Geddes","suffix":""},{"id":523913982,"identity":"c1c819a8-ac00-48ce-bb63-7210d747f04c","order_by":1,"name":"Garrett Levin","email":"","orcid":"https://orcid.org/0000-0002-9461-1747","institution":"North Dakota State University","correspondingAuthor":false,"prefix":"","firstName":"Garrett","middleName":"","lastName":"Levin","suffix":""},{"id":523913983,"identity":"d5342617-2a5e-4691-af5a-0c7e992121e4","order_by":2,"name":"Chinh Luu","email":"","orcid":"","institution":"North Dakota State University","correspondingAuthor":false,"prefix":"","firstName":"Chinh","middleName":"","lastName":"Luu","suffix":""},{"id":523913984,"identity":"47220ba1-201e-450f-b5c0-309b9ac5d2a4","order_by":3,"name":"Natalie Visich","email":"","orcid":"","institution":"North Dakota State University","correspondingAuthor":false,"prefix":"","firstName":"Natalie","middleName":"","lastName":"Visich","suffix":""},{"id":523913985,"identity":"f3421ba5-556d-42dc-a726-adecc8ac3c92","order_by":4,"name":"Scott Hoselton","email":"","orcid":"","institution":"North Dakota State University","correspondingAuthor":false,"prefix":"","firstName":"Scott","middleName":"","lastName":"Hoselton","suffix":""},{"id":523913986,"identity":"7a7ac85a-bcc8-431c-bc12-f73cb2c0ab9f","order_by":5,"name":"Anna Lipzen","email":"","orcid":"https://orcid.org/0000-0003-2293-9329","institution":"Lawrence Berkeley National Laboratory","correspondingAuthor":false,"prefix":"","firstName":"Anna","middleName":"","lastName":"Lipzen","suffix":""},{"id":523913987,"identity":"2e007d5a-878c-4c9b-a930-28290e7e3fa9","order_by":6,"name":"Shuang Zhao","email":"","orcid":"","institution":"University of Alberta","correspondingAuthor":false,"prefix":"","firstName":"Shuang","middleName":"","lastName":"Zhao","suffix":""},{"id":523913988,"identity":"2430be61-d7b6-4bec-acdd-23b7aba2faa9","order_by":7,"name":"Liang Li","email":"","orcid":"https://orcid.org/0000-0002-9347-2108","institution":"University of Alberta","correspondingAuthor":false,"prefix":"","firstName":"Liang","middleName":"","lastName":"Li","suffix":""},{"id":523913989,"identity":"de835a95-892f-403e-af53-6159ee71633a","order_by":8,"name":"George diCenzo","email":"","orcid":"https://orcid.org/0000-0003-3889-6570","institution":"","correspondingAuthor":false,"prefix":"","firstName":"George","middleName":"","lastName":"diCenzo","suffix":""},{"id":523913990,"identity":"a066d084-caa9-42f1-8315-b9492055599f","order_by":9,"name":"Turlough Finan","email":"","orcid":"https://orcid.org/0000-0003-4363-6001","institution":"McMaster University","correspondingAuthor":false,"prefix":"","firstName":"Turlough","middleName":"","lastName":"Finan","suffix":""}],"badges":[],"createdAt":"2025-09-25 17:20:09","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-7715224/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-7715224/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":94632727,"identity":"74242e70-4c37-41ed-b438-d4d084240a48","added_by":"auto","created_at":"2025-10-29 06:30:22","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":371914,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eDelineating a Genetic Region Leading to Ineffective Symbiosis with Alfalfa. \u003c/strong\u003e(\u003cstrong\u003eA\u003c/strong\u003e)\u003cstrong\u003e \u003c/strong\u003eSimplified scheme showing the transfer of pSymA from natural strains into a wildtype strain cured of its own pSymA. (\u003cstrong\u003eB\u003c/strong\u003e)\u003cstrong\u003e \u003c/strong\u003eComparison between the relative abilities of natural strains versus their corresponding pSymA hybrid to promote shoot dry weight production by \u003cem\u003eM. sativa\u003c/em\u003e. (\u003cstrong\u003eC\u003c/strong\u003e)\u003cstrong\u003e \u003c/strong\u003eComparison between the relative abilities of natural strains versus their corresponding pSymB hybrid to promote shoot dry weight production by \u003cem\u003eM. sativa\u003c/em\u003e. (\u003cstrong\u003eB and C\u003c/strong\u003e)\u003cstrong\u003e \u003c/strong\u003eData are expressed as relative shoot dry weight compared to the mean maximum shoot dry weight of the best performing strain. (\u003cstrong\u003eD\u003c/strong\u003e)\u003cstrong\u003e \u003c/strong\u003eDiagram highlighting unique regions present in the HM006 pSymA replicon and absent in the pSymA of the wildtype RmP110 (blue) and other strains (KH46C, T073, USDA1106, and WSM419) (orange). (\u003cstrong\u003eE\u003c/strong\u003e)\u003cstrong\u003e \u003c/strong\u003eRepresentative images of \u003cem\u003eM. sativa\u003c/em\u003eshoots and nodules inoculated with wildtype, pSymA hybrid, and Deletion 1. Scale bar represents 5 cm for the shoot panel and 0.5 cm for the nodule panel. (\u003cstrong\u003eF\u003c/strong\u003e) Quantitation of shoot dry weight accumulation by \u003cem\u003eM. sativa.\u003c/em\u003e\u003c/p\u003e","description":"","filename":"1.png","url":"https://assets-eu.researchsquare.com/files/rs-7715224/v1/c7c36dc037ee749b1106d1b3.png"},{"id":94640014,"identity":"3ff8e654-d466-4763-9647-65405acddcfb","added_by":"auto","created_at":"2025-10-29 07:47:51","extension":"png","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":332014,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eRelative locations and symbiotic effectiveness phenotypes of sub-deletions spanning the 223 kb region removed in D1. \u003c/strong\u003e(\u003cstrong\u003eA\u003c/strong\u003e)\u003cstrong\u003e \u003c/strong\u003eMap displaying the relative locations of genes and approximate lengths of deletions across the region. (\u003cstrong\u003eB\u003c/strong\u003e) Quantitation of\u003cstrong\u003e \u003c/strong\u003eshoot dry weight\u003cstrong\u003e \u003c/strong\u003eaccumulation by \u003cem\u003eM. sativa\u003c/em\u003e following inoculation with the wild type, the pSymA hybrid, D1, and the sub-deletions of D1: D3-D11.\u003cstrong\u003e \u003c/strong\u003e(\u003cstrong\u003eC\u003c/strong\u003e)\u003cstrong\u003e \u003c/strong\u003eRepresentative shoot and nodules images of the partial recovery phenotype resulting from Deletions 7 and 9. Scale bars represent 5 cm for shoot and 0.5 cm for nodule images. (\u003cstrong\u003eD\u003c/strong\u003e) Quantitation of shoot dry weight accumulation by \u003cem\u003eM. sativa \u003c/em\u003efollowing inoculation with the HM006 pSymA hybrid, and sub-deletions of D7 (D14-D16) and D9 (D17-D19).\u003cstrong\u003e \u003c/strong\u003e(\u003cstrong\u003eE\u003c/strong\u003e)\u003cstrong\u003e \u003c/strong\u003eGenetic diagram of the duplicated region found in both D15 and D18 and the annotations of predicted open reading frames within the putative operons.\u003c/p\u003e","description":"","filename":"2.png","url":"https://assets-eu.researchsquare.com/files/rs-7715224/v1/e85f10ef50371a1beb69f8d3.png"},{"id":94640831,"identity":"a276ed63-0d2e-45f1-bd94-d0abd8d37ae1","added_by":"auto","created_at":"2025-10-29 07:50:17","extension":"png","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":526934,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eSymbiotic PHA production caused by the newly discovered operons in Region 15 and Region 18. \u003c/strong\u003e(\u003cstrong\u003eA\u003c/strong\u003e)\u003cstrong\u003e \u003c/strong\u003eExamples of wildtype (RmP110) bacteroids in \u003cem\u003eM. sativa \u003c/em\u003enodules void of PHA in the top panel; examples of\u003cstrong\u003e \u003c/strong\u003eHM006\u003cstrong\u003e \u003c/strong\u003epSymA hybrid (RmP4566) bacteroids with an abundance of PHA granules in the middle panel; and examples of Deletion 1 (RmND063) bacteroids devoid of PHA granules in the bottom panel. Scale bar at 2 µm. (\u003cstrong\u003eB\u003c/strong\u003e) Quantification of PHA production by Nile Red staining of bacteroids collected from \u003cem\u003eM. truncatula \u003c/em\u003enodules. Data expressed as mean Nile Red relative fluorescence units (RFU). Images of Nile Red and DAPI fluorescence of representative isolated bacteroids are shown on left. (\u003cstrong\u003eC\u003c/strong\u003e)\u003cstrong\u003e \u003c/strong\u003eTranscription profile (RNAseq count) across the complete HM006 pSymA replicon (in HM006 pSymA hybrid strain) in bacteroids from nodules of \u003cem\u003eM. sativa\u003c/em\u003e. The \u003cem\u003efix\u003c/em\u003e and \u003cem\u003enif \u003c/em\u003egenes are marked to show relative expression levels with respect to genes in \u003cem\u003epha-1 \u003c/em\u003eand \u003cem\u003epha-2\u003c/em\u003e. Orange highlighted genes represent the \u003cem\u003epha-1 \u003c/em\u003eand \u003cem\u003epha-2 \u003c/em\u003ecluster. (\u003cstrong\u003eD\u003c/strong\u003e)\u003cstrong\u003e \u003c/strong\u003eGUS promoter fusion expression in \u003cem\u003eM. sativa \u003c/em\u003eand \u003cem\u003eM. truncatula \u003c/em\u003eroot nodules. Depicted are images of nodules of plants inoculated with the HM006 pSymA hybrid strain (RmP4566) carrying \u003cem\u003egus\u003c/em\u003e downstream of the pPha promoter region, the pNifH promoter region (bacteroid-on positive control), or the pNodA promoter region (bacteroid-off negative control). Scale bars depict 100 µm.\u003c/p\u003e","description":"","filename":"3.png","url":"https://assets-eu.researchsquare.com/files/rs-7715224/v1/f5c364043362c4f889785a9b.png"},{"id":94632725,"identity":"1296649a-977e-4b9d-bc7b-deaf41a7e940","added_by":"auto","created_at":"2025-10-29 06:30:22","extension":"png","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":382918,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eExpression of \u003c/strong\u003e\u003cem\u003e\u003cstrong\u003epha\u003c/strong\u003e\u003c/em\u003e\u003cstrong\u003e synthesis genes in a naïve background leads to a collapse in \u003c/strong\u003e\u003cem\u003e\u003cstrong\u003eM. sativa\u003c/strong\u003e\u003c/em\u003e\u003cstrong\u003e symbiosis that is alleviated by a local catabolism gene set.\u003c/strong\u003e (\u003cstrong\u003eA and D\u003c/strong\u003e)\u003cstrong\u003e \u003c/strong\u003eQuantitation of\u003cstrong\u003e \u003c/strong\u003eshoot dry weight\u003cstrong\u003e \u003c/strong\u003eaccumulation and (\u003cstrong\u003eE\u003c/strong\u003e) percent of pink (Fix +) nodules produced by \u003cem\u003eM. sativa\u003c/em\u003e following inoculation with the wild type, pSymA hybrid, or the wild type with various regions from HM006 pSymA integrated into the genome.\u003cstrong\u003e \u003c/strong\u003e\u003cem\u003ectrl-1\u003c/em\u003e\u003cstrong\u003e \u003c/strong\u003eand \u003cem\u003ectrl-2, \u003c/em\u003erespectively\u003cem\u003e, \u003c/em\u003erepresent the addition of Region 14 and Region 16 to the wild type.\u003cstrong\u003e \u003c/strong\u003e(\u003cstrong\u003eB and F\u003c/strong\u003e) Representative shoot and nodule images of shoot dry weight and nodule phenotypes from gain-of-function experiments in A, D, and E. Scale bars represents 5 cm in shoot pictures and 0.5 cm in nodule pictures. (\u003cstrong\u003eC\u003c/strong\u003e)\u003cstrong\u003e \u003c/strong\u003eAlignment of \u003cem\u003eS. meliloti \u003c/em\u003eHM006, USDA1963 and \u003cem\u003eRhizobium leguminosarum \u003c/em\u003e3841 \u003cem\u003epha \u003c/em\u003ehomolog-containing genomic regions (Min. alignment sequence identity 0.3), visualized with clustermap.js (\u0026gt;0.5 alignment sequence identity shown). The * indicates \u003cem\u003ehrrP \u003c/em\u003egene encoding a known NCR peptidase (\u003cstrong\u003e39\u003c/strong\u003e).\u003c/p\u003e","description":"","filename":"4.png","url":"https://assets-eu.researchsquare.com/files/rs-7715224/v1/540ca2f853267e53b876c8f8.png"},{"id":94632732,"identity":"23bc1b8e-9579-4d09-b068-373cdb5e8772","added_by":"auto","created_at":"2025-10-29 06:30:22","extension":"png","order_by":5,"title":"Figure 5","display":"","copyAsset":false,"role":"figure","size":277526,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eProposed symbiotic P4HB cycle encoded by \u003c/strong\u003e\u003cem\u003e\u003cstrong\u003eS. meliloti \u003c/strong\u003e\u003c/em\u003e\u003cstrong\u003eHM006 pSymA.\u003c/strong\u003e (\u003cstrong\u003eA\u003c/strong\u003e)\u003cstrong\u003e \u003c/strong\u003eDiagram and annotations of conserved gene clusters associated with \u003cem\u003epha \u003c/em\u003esynthesis genes.\u003cstrong\u003e \u003c/strong\u003e(\u003cstrong\u003eB\u003c/strong\u003e)\u003cstrong\u003e \u003c/strong\u003eProposed P4HB synthesis and catabolism pathways in the context of central metabolism and the well-characterized P3HB cycle in rhizobia. (\u003cstrong\u003eC\u003c/strong\u003e)\u003cstrong\u003e \u003c/strong\u003eLC-MS signal intensities for succinate-semialdehyde and gamma-hydroxybutyrate in \u003cem\u003eM. sativa \u003c/em\u003enodules occupied by wildtype or pSymA hybrid strains.\u003c/p\u003e","description":"","filename":"5.png","url":"https://assets-eu.researchsquare.com/files/rs-7715224/v1/23dd71d9cea2a56b26f9667e.png"},{"id":94632729,"identity":"102a890d-d796-4d63-9b6e-646d53cf137f","added_by":"auto","created_at":"2025-10-29 06:30:22","extension":"png","order_by":6,"title":"Figure 6","display":"","copyAsset":false,"role":"figure","size":283325,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eRemoving \u003c/strong\u003e\u003cem\u003e\u003cstrong\u003eghbD\u003c/strong\u003e\u003c/em\u003e\u003cstrong\u003e from the \u003c/strong\u003e\u003cem\u003e\u003cstrong\u003epha \u003c/strong\u003e\u003c/em\u003e\u003cstrong\u003eloci results in a recovery of symbiosis with \u003c/strong\u003e\u003cem\u003e\u003cstrong\u003eMedicago sativa.\u003c/strong\u003e\u003c/em\u003e\u003cstrong\u003e \u003c/strong\u003e(\u003cstrong\u003eA and C\u003c/strong\u003e)\u003cstrong\u003e \u003c/strong\u003eQuantitation of\u003cstrong\u003e \u003c/strong\u003eshoot dry weight\u003cstrong\u003e \u003c/strong\u003eaccumulation by \u003cem\u003eM. sativa\u003c/em\u003e following inoculation with the RmP110 wildtype, HM006 wildtype, pSymA hybrid, or knockout strains in the pSymA hybrid (\u003cstrong\u003eA\u003c/strong\u003e) or HM006 (\u003cstrong\u003eC\u003c/strong\u003e) backgrounds. (\u003cstrong\u003eB and D\u003c/strong\u003e) Representative images of shoot dry weight production from the experiments in A and D. (\u003cstrong\u003eE\u003c/strong\u003e)\u003cstrong\u003e \u003c/strong\u003eQuantification of PHA production by Nile Red staining of bacteroids collected from \u003cem\u003eM. sativa \u003c/em\u003enodules. Data expressed as mean Nile Red Height.\u003c/p\u003e","description":"","filename":"6.png","url":"https://assets-eu.researchsquare.com/files/rs-7715224/v1/c10773280628b62fc4d87960.png"},{"id":94640282,"identity":"54e33dd4-793a-4801-91e3-0f9620a5084d","added_by":"auto","created_at":"2025-10-29 07:49:00","extension":"png","order_by":7,"title":"Figure 7","display":"","copyAsset":false,"role":"figure","size":236769,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eNADH-dependent reduction of succinic-semialdehyde by GhbD causes a collapse in symbiosis with \u003c/strong\u003e\u003cem\u003e\u003cstrong\u003eMedicago sativa\u003c/strong\u003e\u003c/em\u003e\u003cstrong\u003e. \u003c/strong\u003e(\u003cstrong\u003eA and B\u003c/strong\u003e) Enzyme activity based on colorimetric reduction of NADH (A) or NADPH (B) at 340 nM with purified GhbD and succinate semialdehyde. (\u003cstrong\u003eC\u003c/strong\u003e)\u003cstrong\u003e \u003c/strong\u003eSchematic of the outcomes of flux from SSA via GhbD or AttK enzymes. (\u003cstrong\u003eD\u003c/strong\u003e)\u003cstrong\u003e \u003c/strong\u003eQuantitation of\u003cstrong\u003e \u003c/strong\u003eshoot dry weight\u003cstrong\u003e \u003c/strong\u003eaccumulation by \u003cem\u003eM. sativa\u003c/em\u003e following inoculation with wildtype RmP110, the pSymA hybrid, or RmP110 expressing either \u003cem\u003eghbD\u003c/em\u003e or \u003cem\u003eghbD\u003c/em\u003e with \u003cem\u003eattK\u003c/em\u003e from the native \u003cem\u003eghbD\u003c/em\u003epromoter. (\u003cstrong\u003eE\u003c/strong\u003e)\u003cstrong\u003e \u003c/strong\u003eRepresentative shoots and nodules of \u003cem\u003eM. sativa\u003c/em\u003e plants. Scale bar represents 5 cm in shoot panel and 0.5cm in nodule panel. (\u003cstrong\u003eF\u003c/strong\u003e) The proportion of pink nodules from the experiment in D.\u003c/p\u003e","description":"","filename":"7.png","url":"https://assets-eu.researchsquare.com/files/rs-7715224/v1/55233df3568c52e5b31fbf3b.png"},{"id":94641112,"identity":"a56f905d-21b0-4dbd-9130-efd37a0af935","added_by":"auto","created_at":"2025-10-29 07:51:16","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":4419507,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-7715224/v1/0210684f-a7b4-479c-85bc-f2fcb9418a2c.pdf"},{"id":94640291,"identity":"f1e4142b-0daa-44f1-a186-c50ba4c44e75","added_by":"auto","created_at":"2025-10-29 07:49:01","extension":"pdf","order_by":1,"title":"","display":"","copyAsset":false,"role":"supplement","size":2727795,"visible":true,"origin":"","legend":"Supplementary Information","description":"","filename":"SupplementaryInformation2.0Final9.25.25.pdf","url":"https://assets-eu.researchsquare.com/files/rs-7715224/v1/ce6b81b837c66773c5542abe.pdf"}],"financialInterests":"\u003cb\u003eYes\u003c/b\u003e there is potential Competing Interest.\nBarney Geddes is a cofounder of Lilac Agriculture, Inc.","formattedTitle":"Polyhydroxyalkanoate synthesis by Sinorhizobium meliloti drives a host-specific collapse in symbiosis with Medicago sativa","fulltext":[{"header":"Introduction","content":"\u003cp\u003eAs the primary sustainable source of nitrogen nutrition in agricultural and natural ecosystems, the nitrogen-fixing root nodule symbiosis between rhizobia and legumes is a central player in global biogeochemical cycling and sustainable agricultural productivity. Variation in the symbiotic capacity of rhizobia is widespread within symbionts that are able to nodulate and fix nitrogen with a given host legume (\u003cb\u003e1\u0026ndash;3\u003c/b\u003e). Locally adapted native strains that are less effective at fixing nitrogen for their hosts limit the contribution of the symbiosis to agricultural systems by outcompeting elite inoculant strains for nodulation, dubbed \u0026ldquo;the rhizobium competition problem\u0026rdquo; (\u003cb\u003e4\u0026ndash;6\u003c/b\u003e). Though mechanisms remain elusive, hosts partially overcome this variation by selectively rewarding or punishing (sanctioning) rhizobia based on their productivity (\u003cb\u003e7\u0026ndash;10\u003c/b\u003e). Uncovering the genetics responsible for variation in microbial services to their hosts and elucidating what forces (e.g., \u0026ldquo;cheating\u0026rdquo;) maintain such widespread variation in symbioses despite host sanctioning mechanisms have emerged as key questions in microbial ecology (\u003cb\u003e3\u003c/b\u003e). Recent studies point towards much of this variation being associated with mobile modular units such as plasmids and integrative conjugative elements (ICE) that are common features of rhizobium genomes (\u003cb\u003e1, 11, 12\u003c/b\u003e).\u003c/p\u003e\u003cp\u003eIn the tripartite genome of the model rhizobium \u003cem\u003eSinorhizobium meliloti\u003c/em\u003e, the \u0026ldquo;symbiotic megaplasmid\u0026rdquo; pSymA contains many essential genes for symbiosis with its \u003cem\u003eMedicago\u003c/em\u003e hosts (\u003cem\u003enod, nif\u003c/em\u003e and \u003cem\u003efix\u003c/em\u003e). These gene sets, which encode the ability to produce the \u0026ldquo;Nod Factor\u0026rdquo; that initiates nodulation of the host (the \u003cem\u003enod\u003c/em\u003e genes) and to fix nitrogen in root nodules (the \u003cem\u003enif\u003c/em\u003e and \u003cem\u003efix\u003c/em\u003e genes) (\u003cb\u003e13\u003c/b\u003e), account for ~\u0026thinsp;5% of its size, yet are sufficient for functional symbiosis in the absence of the remaining\u0026thinsp;~\u0026thinsp;95% of pSymA (\u003cb\u003e14\u003c/b\u003e). The chromid pSymB is more ancestral than pSymA and has been implicated in playing a primary role in saprophytic competence (\u003cb\u003e15\u003c/b\u003e), while also carrying the dicarboxylate transport genes (\u003cem\u003edct\u003c/em\u003e) essential for fueling nitrogen fixation in the nodule and exopolysaccharide synthesis (\u003cem\u003eexo\u003c/em\u003e) genes important for plant invasion (\u003cb\u003e16, 17\u003c/b\u003e). Despite decades of genetic study, the vast majority of genes on these replicons remain undefined with respect to their contributions to symbiosis. In part, this is due to the challenging nature of performing genetic screens for more subtle phenotypes (e.g., partner quality or competition for nodule occupancy) than the nodule-minus phenotypes of \u003cem\u003enod\u003c/em\u003e mutants or nitrogen-fixation-minus phenotypes of \u003cem\u003enif\u003c/em\u003e and \u003cem\u003efix\u003c/em\u003e mutants (\u003cb\u003e18, 19\u003c/b\u003e), or of genes associated with symbiotic incompatibility (\u003cb\u003e20, 21\u003c/b\u003e).\u003c/p\u003e\u003cp\u003eThe rapid increase in genome sequencing of rhizobium isolates in recent decades has revealed that rhizobia, such as \u003cem\u003eS. meliloti\u003c/em\u003e, contain large, open pangenomes (\u003cb\u003e11, 22\u003c/b\u003e), further demonstrating the unexplored genetics that could contribute to variation in symbiosis outcomes. This wealth of genomic data, coupled with next-generation sequencing approaches to screen rhizobium symbiosis traits \u003cem\u003een masse\u003c/em\u003e (\u003cb\u003e23, 24\u003c/b\u003e), enable the prediction of candidate genes that may impact the mutualism using quantitative genetics such as genome-wide association studies (GWAS) (\u003cb\u003e12, 25\u0026ndash;27\u003c/b\u003e). However, thus far, few candidates identified by these studies have been empirically validated for their role in symbiosis.\u003c/p\u003e\u003cp\u003eHere, we used strains wherein extrachromosomal replicons have been cured (\u003cb\u003e15, 28\u003c/b\u003e) as surrogate hosts to identify genetic determinants that define differences observed in the vast diversity of effectiveness of functional symbioses between \u003cem\u003eSinorhizobium\u003c/em\u003e strains and their \u003cem\u003eMedicago\u003c/em\u003e hosts. First, we investigated the co-transfer of partner quality with extrachromosomal replicons from five natural strains. The pSymA replicons were found to have the greatest impact on symbiotic performance and in the most extreme case, the transfer of pSymA from strain HM006 resulted in a substantial reduction of symbiotic performance with \u003cem\u003eMedicago sativa\u003c/em\u003e. Next, through a series of refined deletions and gain-of-function experiments, the locus responsible for limiting symbiotic nitrogen fixation was identified and shown to be dominant. Genes within the locus were found to be involved in carbon polymer synthesis during symbiosis, and their impact on symbiotic N\u003csub\u003e2\u003c/sub\u003e-fixation revealed an unanticipated aspect to carbon metabolism in root nodules and its impact on partner quality.\u003c/p\u003e"},{"header":"Results","content":"\u003cp\u003e\u003cb\u003eCo-transfer of host-specific symbiotic effectiveness phenotypes with the\u003c/b\u003e \u003cb\u003eSinorhizobium\u003c/b\u003e \u003cb\u003ereplicons.\u003c/b\u003e\u003c/p\u003e\u003cp\u003eWe hypothesized that most genes contributing to variance in partner quality would be present on the extrachromosomal replicons in sinorhizobia and reasoned we could identify key genes involved in partner quality by transferring the extrachromosomal replicons (~\u0026thinsp;40% of genome) to a surrogate \u003cem\u003eS. meliloti\u003c/em\u003e background in which pSymA, pSymB, or both pSymA and pSymB were removed (\u003cb\u003e14, 15, 29, 30\u003c/b\u003e). For replicon transfer, we used four genome-sequenced \u003cem\u003eS. meliloti\u003c/em\u003e strains: KH46, HM006, USDA1021, and T073 and one \u003cem\u003eSinorhizobium medicae\u003c/em\u003e strain, WSM419 (\u003cb\u003e31, 32\u003c/b\u003e). These genome-sequenced strains showed significant variation in partner quality (shoot-dry weight accumulation under nitrogen-limiting conditions) with \u003cem\u003eMedicago sativa\u003c/em\u003e and \u003cem\u003eMedicago truncatula\u003c/em\u003e (\u003cb\u003e33\u003c/b\u003e) (\u003cb\u003eSupplemental Figure \u003cspan refid=\"MOESM1\" class=\"InternalRef\"\u003eS1\u003c/span\u003e\u003c/b\u003e).\u003c/p\u003e\u003cp\u003eThe pSymA or pSymB replicons from the five strains were transferred independently or together to derivatives of the long-studied wildtype chromosomal background SU47 (Rm1021-derived background strain RmP110, or Rm5000) (\u003cb\u003e33, 34\u003c/b\u003e) (Fig.\u0026nbsp;1A). The symbiotic productivity of the resulting hybrid strains with \u003cem\u003eM. sativa\u003c/em\u003e revealed a clear positive correlation between the abilities of the natural strains to promote plant growth, and the relative abilities of hybrid strains containing their respective pSymA replicons (Fig.\u0026nbsp;1B, \u003cb\u003eSupplemental Figures S2A and S3A\u003c/b\u003e). In contrast, no significant difference in plant growth promotion was observed for pSymB transfer (Fig.\u0026nbsp;1C, \u003cb\u003eSupplemental Figures S2A and S3A\u003c/b\u003e). With \u003cem\u003eM. truncatula\u003c/em\u003e as a host, increased growth relative to the wildtype was also observed by transfer of pSymAs from more effective strains (2- to 3-fold increase in SDW) and was enhanced with co-transfer of pSymB (~\u0026thinsp;6-fold increase in SDW) (\u003cb\u003eSupplemental Figures S2B and S3B\u003c/b\u003e).\u003c/p\u003e\u003cp\u003e\u003cb\u003eFigure 1. Delineating a Genetic Region Leading to Ineffective Symbiosis with Alfalfa.\u003c/b\u003e (\u003cb\u003eA\u003c/b\u003e) Simplified scheme showing the transfer of pSymA from natural strains into a wildtype strain cured of its own pSymA. (\u003cb\u003eB\u003c/b\u003e) Comparison between the relative abilities of natural strains versus their corresponding pSymA hybrid to promote shoot dry weight production by \u003cem\u003eM. sativa\u003c/em\u003e. (\u003cb\u003eC\u003c/b\u003e) Comparison between the relative abilities of natural strains versus their corresponding pSymB hybrid to promote shoot dry weight production by \u003cem\u003eM. sativa\u003c/em\u003e. (\u003cb\u003eB and C\u003c/b\u003e) Data are expressed as relative shoot dry weight compared to the mean maximum shoot dry weight of the best performing strain. (\u003cb\u003eD\u003c/b\u003e) Diagram highlighting unique regions present in the HM006 pSymA replicon and absent in the pSymA of the wildtype RmP110 (blue) and other strains (KH46C, T073, USDA1106, and WSM419) (orange). (\u003cb\u003eE\u003c/b\u003e) Representative images of \u003cem\u003eM. sativa\u003c/em\u003e shoots and nodules inoculated with wildtype, pSymA hybrid, and Deletion 1. Scale bar represents 5 cm for the shoot panel and 0.5 cm for the nodule panel. (\u003cb\u003eF\u003c/b\u003e) Quantitation of shoot dry weight accumulation by \u003cem\u003eM. sativa.\u003c/em\u003e\u003c/p\u003e\u003cp\u003e\u003cb\u003eA genetic determinant on the\u003c/b\u003e \u003cb\u003eS. meliloti\u003c/b\u003e \u003cb\u003eHM006 pSymA leads to a collapse of symbiotic efficiency with\u003c/b\u003e \u003cb\u003eM. sativa.\u003c/b\u003e As a platform to investigate genetics that underly variation in pSymA partner quality, we chose the pSymA that conferred the most dramatic phenotype; a collapse in the productivity of the symbiosis with \u003cem\u003eM. sativa\u003c/em\u003e conferred by the pSymA from \u003cem\u003eS. meliloti\u003c/em\u003e HM006 when transferred to the RmP110 background, which forms an effective symbiosis with \u003cem\u003eM. sativa\u003c/em\u003e when bearing its native pSymA. We first compared the pSymA contents of HM006 and RmP110. While known symbiosis gene sets in the pSymAs were nearly identical (\u003cb\u003eSupplemental Fig.\u0026nbsp;4\u003c/b\u003e), two large regions from HM006 pSymA were found to be absent from the RmP110 pSymA (Fig.\u0026nbsp;1D, \u003cb\u003eSupplemental Figure S5\u003c/b\u003e). Deletion of these regions from the HM006 pSymA-hybrid strain revealed that the loss of 223 kb in one region (Deletion 1, D1) completely restored symbiotic effectiveness with \u003cem\u003eM. sativa\u003c/em\u003e, whereas the deletion of the second region (Deletion 2, D2) had no effect (Figs.\u0026nbsp;1E and 1F). Neither deletion significantly impacted symbiosis with \u003cem\u003eM. truncatula\u003c/em\u003e (\u003cb\u003eSupplemental Figure S6\u003c/b\u003e). We concluded that dominant genetic determinant(s) within the D1 region of the HM006 pSymA led to the collapse in effective symbiosis with \u003cem\u003eM. sativa\u003c/em\u003e.\u003c/p\u003e\u003cp\u003e\u003cb\u003eAdditive inhibition of symbiotic effectiveness by a duplicated gene cluster on HM006 pSymA.\u003c/b\u003e To identify genes within the 223 kb region removed in D1 responsible for reduced symbiotic N\u003csub\u003e2\u003c/sub\u003e-fixation with \u003cem\u003eM. sativa\u003c/em\u003e, we generated nine sub-deletions and screened them for their symbiotic productivity with \u003cem\u003eM. sativa\u003c/em\u003e (Figs.\u0026nbsp;2A and 2B). While none of the deletions fully restored symbiotic productivity, two sub-deletions, D7 and D9, each resulted in a\u0026thinsp;~\u0026thinsp;50% recovery in shoot dry weight production (Figs.\u0026nbsp;2A, 2B, and 2C). Further sub-deletions within the 60 kb region removed in D7 (D14-D16) and the 35 kb region removed in D9 (D17-D19) localized the genes responsible for the partial recovery of \u003cem\u003eM. sativa\u003c/em\u003e shoot dry weight production to two gene sets deleted in D15 and D18 (Figs.\u0026nbsp;2A and 2D). Both genetic loci, which we name \u003cem\u003epha-1\u003c/em\u003e (removed in D15, locus tags CDO22_RS21580-CDO22_RS21610) and \u003cem\u003epha-2\u003c/em\u003e (removed in D18, locus tags CDO22_RS21900-CDO22_RS21925), show an operon-like configuration with a 99.9% identity to one-another over a majority of their length (Fig.\u0026nbsp;2E). Both clusters included genes annotated as encoding an iron-containing alcohol dehydrogenase (TIGR02638 E-value 1.70e-60) and two subunits of a Class III polyhydroxyalkanoic acid synthase: PhaE (TIGR01834 E-value 1.33e-08) and PhaC (TIGR01836 E-value 2.94e-113/3.25e-85) in an operon-like configuration. Homology between the loci extends 2783 bp upstream of these gene sets and contains two annotated pseudogenes along with, presumably, the promoter. A putative operon in \u003cem\u003epha-1\u003c/em\u003e extends beyond the duplicated region and included genes predicted to encode a Coenzyme A transferase (TIGR02428 E-value 8.25e-11) and an acetyl-CoA synthetase (TIGR02188 E-value 0).\u003c/p\u003e\u003cp\u003e\u003cb\u003eFigure 2. Relative locations and symbiotic effectiveness phenotypes of sub-deletions spanning the 223 kb region removed in D1.\u003c/b\u003e (\u003cb\u003eA\u003c/b\u003e) Map displaying the relative locations of genes and approximate lengths of deletions across the region. (\u003cb\u003eB\u003c/b\u003e) Quantitation of shoot dry weight accumulation by \u003cem\u003eM. sativa\u003c/em\u003e following inoculation with the wild type, the pSymA hybrid, D1, and the sub-deletions of D1: D3-D11. (\u003cb\u003eC\u003c/b\u003e) Representative shoot and nodules images of the partial recovery phenotype resulting from Deletions 7 and 9. Scale bars represent 5 cm for shoot and 0.5 cm for nodule images. (\u003cb\u003eD\u003c/b\u003e) Quantitation of shoot dry weight accumulation by \u003cem\u003eM. sativa\u003c/em\u003e following inoculation with the HM006 pSymA hybrid, and sub-deletions of D7 (D14-D16) and D9 (D17-D19). (\u003cb\u003eE\u003c/b\u003e) Genetic diagram of the duplicated region found in both D15 and D18 and the annotations of predicted open reading frames within the putative operons.\u003c/p\u003e\u003cp\u003e\u003cb\u003eSymbiotic carbon polymer synthesis leads to a host-specific inhibition of productive symbiosis with\u003c/b\u003e \u003cb\u003eM. sativa.\u003c/b\u003e Based on their annotation, we hypothesized that the gene sets in \u003cem\u003epha-1\u003c/em\u003e and \u003cem\u003epha-2\u003c/em\u003e may be involved in the production of a polyhydroxyalkanoate (PHA) carbon polymer such as polyhydroxybutyrate (PHB) during symbiosis. PHB is visualized by transmission electron microscopy (TEM) as large granules within bacteroids of rhizobia that synthesize PHB during symbiosis (\u003cb\u003e35\u003c/b\u003e). TEM of \u003cem\u003eM. sativa\u003c/em\u003e and \u003cem\u003eM. truncatula\u003c/em\u003e nodules revealed the presence of large granules in the strains bearing pSymA-HM006. These granules were absent from bacteroids of the wild type or the D1 strain lacking both \u003cem\u003epha-1\u003c/em\u003e and \u003cem\u003epha-2\u003c/em\u003e (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e3\u003c/span\u003eA, \u003cb\u003eSupplemental Figure S7\u003c/b\u003e). While the small inefficient \u003cem\u003eM. sativa\u003c/em\u003e nodules restricted further characterization, we also measured PHA production in effective \u003cem\u003eM. truncatula\u003c/em\u003e nodules using flow cytometry, using PHA-bound Nile Red for quantitation (\u003cb\u003e36\u003c/b\u003e) (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e3\u003c/span\u003eB). A significant increase in Nile Red fluorescence of bacteroids was observed in wildtype HM006 and the HM006-pSymA hybrid strains relative to wildtype RmP110 (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e3\u003c/span\u003eB). The increased fluorescence was significantly reduced in D1, which lacked both \u003cem\u003epha-1\u003c/em\u003e and \u003cem\u003epha-2\u003c/em\u003e, as well as D15 which lacked the expanded gene set of \u003cem\u003epha-1\u003c/em\u003e (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e3\u003c/span\u003eB).\u003c/p\u003e\u003cp\u003eTo examine the expression of the PHA synthesis genes during symbiosis, we analyzed the transcriptome of the HM006 pSymA hybrid strain in \u003cem\u003eM. sativa\u003c/em\u003e and \u003cem\u003eM. truncatula\u003c/em\u003e root nodules. RNAseq showed a high level of transcription across the regions that include the PHA synthesis genes in both hosts (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e3\u003c/span\u003eC, \u003cb\u003eSupplemental Figure S8\u003c/b\u003e). Several genes within them were among the top 50 most highly expressed genes from HM006 pSymA during symbiosis (\u003cb\u003eSupplemental Table S10\u003c/b\u003e). We further observed robust \u003cem\u003epha\u003c/em\u003e cluster gene expression in nitrogen-fixing bacteroids of nodules using \u003cem\u003egusA\u003c/em\u003e promoter fusion constructs (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e3\u003c/span\u003eD, \u003cb\u003eSupplemental Figure S9\u003c/b\u003e). Altogether, these data indicate that the production of carbon polymers during symbiosis by bacteroids containing the newly discovered PHA genes drives a host-specific symbiotic incompatibility between \u003cem\u003eS. meliloti\u003c/em\u003e and \u003cem\u003eM. sativa.\u003c/em\u003e\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003cp\u003e\u003cb\u003eExpression of the\u003c/b\u003e \u003cb\u003epha\u003c/b\u003e \u003cb\u003egenes alone leads to a complete symbiotic collapse in a na\u0026iuml;ve background that is relieved by local catabolism genes.\u003c/b\u003e To independently verify that the genes present in the \u003cem\u003epha\u003c/em\u003e loci are the sole components of the HM006 pSymA replicon responsible for the reduced symbiotic N\u003csub\u003e2\u003c/sub\u003e-fixation in \u003cem\u003eM. sativa\u003c/em\u003e nodules, we applied a gain-of-function approach. We integrated the putative symbiotic PHA synthesis operon from \u003cem\u003epha-1\u003c/em\u003e (Figs.\u0026nbsp;2A and 2E) into the genome of RmP110 with its native pSymA. Strikingly, RmP110 with \u003cem\u003epha-1\u003c/em\u003e alone resulted in a more complete collapse of symbiosis than that associated with pSymA-HM006 (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e4\u003c/span\u003eA). Shoot dry weight production was similar to uninoculated control plants (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e4\u003c/span\u003eA), with few pink nodules (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e4\u003c/span\u003eE) and barely detectable acetylene reduction activity (\u003cb\u003eSupplemental Figure S10\u003c/b\u003e). TEM analysis showed that nodules were sparsely populated with bacteroids which showed signs of degradation by the host plant (\u003cb\u003e37\u003c/b\u003e) (\u003cb\u003eSupplemental Figure S11\u003c/b\u003e).\u003c/p\u003e\u003cp\u003eTo rationalize the discordance between the partially reduced symbiotic efficiency conferred by the complete HM006 pSymA, and the near non-functional symbiosis rendered by the \u003cem\u003epha\u003c/em\u003e gene cluster alone, we hypothesized that other genetic determinants on HM006 pSymA alleviate the inhibition of symbiosis caused by the \u003cem\u003epha\u003c/em\u003e cluster. In search of such determinants, we identified other strains bearing \u003cem\u003ephaC\u003c/em\u003e homologues by BLAST (\u003cb\u003eSupplemental Table S11\u003c/b\u003e) and compared the content of the local genetic region surrounding \u003cem\u003ephaC\u003c/em\u003e (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e4\u003c/span\u003eC). These included the well-studied pea symbiont \u003cem\u003eRhizobium johnstonii\u003c/em\u003e (formerly \u003cem\u003eRhizobium leguminosarum\u003c/em\u003e) Rlv3841 as well as \u003cem\u003eS. meliloti\u003c/em\u003e USDA1963. We found that a set of carbon metabolism genes previously characterized in \u003cem\u003eA. tumefaciens\u003c/em\u003e C58 as involved in N-acyl-homoserine lactone catabolism (\u003cem\u003eattKLM\u003c/em\u003e) were consistently collocated with the \u003cem\u003epha\u003c/em\u003e cluster in other strains (\u003cb\u003e38\u003c/b\u003e) (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e4\u003c/span\u003eC), and we considered they could be responsible for alleviating the inhibition of symbiosis caused by the \u003cem\u003epha\u003c/em\u003e cluster alone when they were transferred with HM006 pSymA. We cloned the catabolism gene set and introduced them to the wild type that contained the \u003cem\u003epha\u003c/em\u003e gene cluster. Introduction of the catabolism genes was sufficient to restore functional symbiosis (Fix\u0026thinsp;+\u0026thinsp;phenotype) to the wildtype background characterized by increased shoot dry weight and pink nodule number (Figs.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e4\u003c/span\u003eD, \u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e4\u003c/span\u003eE, and \u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e4\u003c/span\u003eF). Consistent with these data, Deletion 8 of HM006 pSymA which lacked these genes was more symbiotically impaired than other deletions (Fig.\u0026nbsp;2B).\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003cp\u003e\u003cb\u003eEvidence for a novel symbiotic poly(gamma)hydroxybutyrate cycle.\u003c/b\u003e We considered that the recovery of symbiosis mediated by the catabolism genes may be the result of a PHA cycle, wherein the \u003cem\u003epha\u003c/em\u003e synthesis genes produce a polymer, whereas the \u003cem\u003eatt\u003c/em\u003e catabolism genes are able to break it down. Based on the annotation of the function of homologues to the \u003cem\u003eatt\u003c/em\u003e genes in \u003cem\u003eA. tumefaciens\u003c/em\u003e (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e5\u003c/span\u003eA), we predicted a novel pathway for polymer synthesis and catabolism based on the intermediate gamma-hydroxybutyrate (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e5\u003c/span\u003eB). The proposed pathway places the synthesis genes in a rational scheme for polymer production based on their annotations; oxidation of succinate semialdehyde to gamma-hydroxybutyrate by the iron-containing dehydrogenase, the addition of an acetyl-CoA group by the CoA transferase generating GHB-CoA (with acetyl-CoA substrate generated by acetyl-CoA synthetase), and polymerization of GHB-CoA monomers via the PhaE/PhaC PHB synthase. The breakdown of the P4HB product of the proposed synthesis pathway then fits with the characterized activities of homologues to AttL and AttK, wherein they break down GHB to succinate, fueling the TCA cycle and bacteroid nitrogen-fixation. In this scheme, the lactonase AttM could conceivably resolve monomers from a GHB polymer with a similar hydration reaction to the cleaving of the lactone ring in gamma-hydroxybutyrolactone (\u003cb\u003e38\u003c/b\u003e). To provide evidence for the proposed pathway, we performed metabolomics of root nodules formed by rhizobia bearing or lacking these pathways, looking for the presence and abundance of pathway intermediates. Consistent with the pathway scheme above, we found significant increase of both the predicted pathway intermediates, SSA and GHB (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e5\u003c/span\u003eC, \u003cb\u003eSupplemental Figure S12\u003c/b\u003e).\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003cp\u003e\u003cb\u003eConversion of succinate semialdehyde to gamma-hydroxybutyrate by an iron-containing dehydrogenase in bacteroids is the mechanism for symbiotic collapse in\u003c/b\u003e \u003cb\u003eM. sativa\u003c/b\u003e. To gain further insight into the role of individual genes in the pathway, we constructed knockout mutations within genes from each step of the predicted pathway (\u003cem\u003eghbD\u003c/em\u003e, \u003cem\u003ephaC2\u003c/em\u003e, \u003cem\u003eghbT\u003c/em\u003e, and \u003cem\u003eacs\u003c/em\u003e), with double knockouts created for the duplicated genes \u003cem\u003eghbD\u003c/em\u003e and \u003cem\u003ephaC2\u003c/em\u003e. Strikingly, only a double knockout within the iron-containing dehydrogenase \u003cem\u003eghbD\u003c/em\u003e (the predicted first step of the synthesis pathway) was able to recover symbiosis (Figs.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e6\u003c/span\u003eA and \u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e6\u003c/span\u003eB). These data suggested that the first step of the predicted pathway, which we predicted to direct carbon from SSA away from succinate in a reductant consuming reaction, rather than polymer synthesis per se, was responsible for symbiotic collapse. To ensure these data were not an artifact of working in the HM006 pSymA hybrid background, we also constructed double-knockouts of the iron-containing dehydrogenase in HM006 and showed a similar recovery of symbiosis with \u003cem\u003eM. sativa\u003c/em\u003e (Figs.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e6\u003c/span\u003eC and \u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e6\u003c/span\u003eD), coincident with a reduction in carbon polymer synthesis observed by flow cytometry (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e6\u003c/span\u003eE, \u003cb\u003eSupplemental Figure S13\u003c/b\u003e).\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003cp\u003eSince GhbD, the iron-containing dehydrogenase, appeared the critical element of symbiotic collapse, we wished to verify its hypothesized activity in the predicted P4HB synthesis pathway (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e5\u003c/span\u003eB). We purified the iron-containing dehydrogenase and demonstrated its activity with SSA as a substrate (\u003cb\u003eSupplemental Figures S14 and S15\u003c/b\u003e). Purified GhbD showed a Km of 0.1129 or 0.7708 and a Vmax of 1.891 or 2.469 in an NADH/NADPH linked spectrophotometric assay when incubated with SSA and NADH or NADPH, respectively, indicating a higher affinity for NADH as a cofactor (Figs.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e7\u003c/span\u003eA and \u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e7\u003c/span\u003eB). The loss-of-function data combined with the verified activity of GhbD suggested that the primary mechanism for symbiotic collapse is the biochemical conversion of SSA to GHB in bacteroids rather than polymer synthesis, since \u003cem\u003ephaC\u003c/em\u003e double mutants retained symbiotic collapse phenotype (Figs.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e6\u003c/span\u003eA and \u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e6\u003c/span\u003eB). To further support this presumption, we expressed the dehydrogenase in the na\u0026iuml;ve wildtype background, without the downstream portions of the operon predicted to be involved in GHB polymerization (Figs.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e7\u003c/span\u003eD to \u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e7\u003c/span\u003eF). We found this construct was still sufficient for the collapse of \u003cem\u003eM. sativa\u003c/em\u003e symbiosis previously observed by transferring the whole operon (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e4\u003c/span\u003e). Based on the confirmed biochemical activity of GhbD (Figs.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e7\u003c/span\u003eA and \u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e7\u003c/span\u003eB), we further predicted we could rescue the symbiotic phenotype caused by \u003cem\u003eghbD\u003c/em\u003e expression by overexpressing \u003cem\u003eattK\u003c/em\u003e from the catabolism gene set which in \u003cem\u003eAgrobacterium\u003c/em\u003e is involved in the conversion of SSA into succinate in a reductant generating reaction (\u003cb\u003e38\u003c/b\u003e) (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e7\u003c/span\u003eC). Such an activity would compete with the conversion of SSA to GHB in a reductant-consuming reaction by GhbD (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e7\u003c/span\u003eC). Indeed, we found that including the \u003cem\u003eattK\u003c/em\u003e gene downstream of \u003cem\u003eghbD\u003c/em\u003e in the gain-of-function construct significantly restored symbiosis between \u003cem\u003eS. meliloti\u003c/em\u003e and \u003cem\u003eM. sativa\u003c/em\u003e (Figs.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e7\u003c/span\u003eD to \u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e7\u003c/span\u003eF).\u003c/p\u003e\u003cp\u003e\u003c/p\u003e"},{"header":"Discussion","content":"\u003cp\u003eReplicon transplantation experiments led to the finding that the synthesis of GHB-based hydrocarbon polymers is linked to the collapse of symbiotic effectiveness in \u003cem\u003eM. sativa\u003c/em\u003e nodules leading to a reduction in partner quality. Several differences are notable between the carbon polymer production genes discovered here (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e4\u003c/span\u003eH) and the well-characterized polyhydroxybutyrate synthesis genes (\u003cem\u003ephbABC\u003c/em\u003e). First, annotations of genes in the operon (e.g., the inclusion of an iron-dependent dehydrogenase) are inconsistent with those in the well-characterized PHB synthesis pathway in rhizobia (\u003cb\u003e40\u003c/b\u003e). Second, the high level of expression of these genes during symbiosis contrasts with the typically described pattern of PHB synthesis in \u003cem\u003eS. meliloti\u003c/em\u003e, where PHB is produced in the free-living state and during infection thread invasion but not during symbiotic nitrogen fixation (\u003cb\u003e36, 41\u003c/b\u003e). The inclusions in bacteroids observed in electron micrographs did not show the typical bright granules due to a lack of staining by lead citrate as is often observed for PHB (\u003cb\u003e35\u003c/b\u003e). Finally, the observation of GHB accumulation in \u003cem\u003eM. sativa\u003c/em\u003e nodules occupied by rhizobia containing the new \u003cem\u003epha\u003c/em\u003e synthesis genes (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e5\u003c/span\u003eC), and the verification of GhbD as a dehydrogenase that reduces SSA to GHB (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e7\u003c/span\u003e) further highlight the unique nature of this pathway. Altogether, these data support its existence as a novel symbiotic PHA synthesis pathway in rhizobia. We further supply evidence that the \u003cem\u003eatt\u003c/em\u003e catabolism genes impact the symbiotic phenotype associated with symbiotic PHA synthesis and propose that the \u003cem\u003epha\u003c/em\u003e and \u003cem\u003eatt\u003c/em\u003e clusters act together in a symbiotic P4HB cycle (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e5\u003c/span\u003eB). A likely source of SSA, the substrate for the first step in the proposed synthesis pathway, is gamma-aminobutyrate (GABA) catabolism, which is provided to bacteroids from plant cells and catabolised to SSA via highly redundant transamination reactions (\u003cb\u003e42\u0026ndash;44\u003c/b\u003e).\u003c/p\u003e\u003cp\u003eIt has been suggested that the production of reduced carbon storage polymers such as PHB may negatively impact the symbiosis by diverting resources away from nitrogen fixation (\u003cb\u003e7\u003c/b\u003e). If the \u003cem\u003eM. sativa\u003c/em\u003e phenotypes we document in this study was the result of diversion of resources from symbiosis via PHA synthesis, they would be by far the most dramatic observation of this phenomenon to date. To investigate the plausibility of this, we utilized an available host/symbiont metabolic model (VINE (\u003cb\u003e45\u003c/b\u003e)) to investigate the degree to which the reduction in \u003cem\u003eM. sativa\u003c/em\u003e partner quality could be explained by resource consumption (either reductant or carbon) by symbiotic PHA synthesis (\u003cb\u003eSupplemental Fig.\u0026nbsp;16\u003c/b\u003e). While the modelling was consistent with resource consumption negatively impacting nitrogen fixation, the magnitude of reduction in shoot dry weight accumulation predicted by the model was significantly less than observed in \u003cem\u003eM. sativa\u003c/em\u003e experimentally. We thus conclude that the observed symbiotic collapse cannot be explain simply due to diversion of reductant or carbon away from nitrogen fixation.\u003c/p\u003e\u003cp\u003eInstead, we hypothesize that the accumulation of GHB intermediates during PHA polymer synthesis in \u003cem\u003eM. sativa\u003c/em\u003e nodules is toxic to \u003cem\u003eSinorhizobium\u003c/em\u003e bacteroids, resulting in a loss of nitrogen fixation capacity. This is supported by multiple observations. First, only mutations in the \u003cem\u003epha\u003c/em\u003e loci that abolish \u003cem\u003eghbD\u003c/em\u003e function enable the recovery of symbiosis in \u003cem\u003eM. sativa\u003c/em\u003e (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e6\u003c/span\u003eA). Similarly, expressing GhbD alone (converting SSA to GHB) in a na\u0026iuml;ve background resulted in a substantial reduction in symbiosis in \u003cem\u003eM. truncatula\u003c/em\u003e, which was partially rescued by co-expressing AttL which diverts SSA to succinate (\u003cb\u003eSupplemental Figure S18\u003c/b\u003e). Both of these observations suggest that synthesis of the pathway intermediates (rather than the polymer itself) was sufficient to collapse the symbiosis. Lastly, even though large PHA granules accumulate in bacteroids within both \u003cem\u003eM. sativa\u003c/em\u003e and \u003cem\u003eM. truncatula\u003c/em\u003e nodules, a substantial increase in GHB and SSA intermediates was only observed in \u003cem\u003eM. sativa\u003c/em\u003e nodules (\u003cb\u003eSupplemental Figure S17\u003c/b\u003e), consistent with our observation of a strong symbiotic phenotype only in this host.\u003c/p\u003e\u003cp\u003eThe production of reduced carbon polymers such as PHB have been implicated in rhizobium \u0026ldquo;cheating\u0026rdquo;, in which the rhizobium benefits (storage of carbon for use once released from the nodule) at the expense of the host (reduced rate of nitrogen fixation) (\u003cb\u003e7\u003c/b\u003e). However, it is not clear whether carbon polymer synthesis is a cheating behavior. Although PHB mutants showed increased nitrogen fixation in some rhizobia (\u003cb\u003e46\u003c/b\u003e), limited effects were observed in others (\u003cb\u003e35, 41, 46, 47\u003c/b\u003e). In fact, PHB production and other lipogenesis represent important reductant stores during symbiotic nitrogen fixation and therefore may facilitate rather than limit nitrogen-fixation capacity (\u003cb\u003e48\u003c/b\u003e). In this study, removing the \u003cem\u003epha\u003c/em\u003e gene sets from HM006 pSymA did not significantly impact partner quality with \u003cem\u003eM. truncatula\u003c/em\u003e (\u003cb\u003eSupplemental Figure S6\u003c/b\u003e), even though the bacteroids within \u003cem\u003eM. truncatula\u003c/em\u003e nodules did accumulate large PHA granules (Figs.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e3\u003c/span\u003eB and \u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e3\u003c/span\u003eD, \u003cb\u003eSupplemental Figures S7, S8 and S9\u003c/b\u003e), suggesting that the polymer synthesis was not at the expense of the host. While P4HB synthesis in bacteriods came at a large cost to \u003cem\u003eM. sativa\u003c/em\u003e, the proposed toxicity of the intermediates to the bacteriods and the complete collapse of the symbiosis likely means that P4HB synthesis hurt, rather than benefited, \u003cem\u003eS. meliloti\u003c/em\u003e. Collectively, these results suggest that at least on the tested plants, P4HB synthesis is unlikely to represent a cheating behavior.\u003c/p\u003e\u003cp\u003ePresumably, the \u003cem\u003epha\u003c/em\u003e clusters confer some benefit to the rhizobia and/or the host to maintain them in rhizobium populations despite the potential for limiting partner quality on some hosts. PHA clusters have been theorized to benefit the symbiosis by providing a source of carbon to fuel SNF during carbon-limited scenarios (\u003cb\u003e49\u003c/b\u003e), and the cycle we propose here would be particularly well suited to this purpose by engaging with central metabolism through dicarboxylates (unlike the canonical PHB catabolism pathway (\u003cb\u003e50\u003c/b\u003e)), which must allow SNF in the nodule to drive ammonium excretion to the plant (\u003cb\u003e51\u003c/b\u003e). PHAs accumulated in nodules could also benefit rhizobia outside of the nodule where they often encounter carbon-limited soil environments (\u003cb\u003e36, 51\u003c/b\u003e). However, as bacteriods are terminally differentiated in \u003cem\u003eMedicago\u003c/em\u003e nodules and thus cannot regrow in the soil, it may be that PHAs accumulated in bacteriods do not benefit free-living rhizobia. Alternatively, selection for maintenance of the \u003cem\u003epha\u003c/em\u003e clusters may be independent of the symbiosis, while their expression in bacteriods is non-beneficial and the result of inappropriate activation following detection of plant-produced signals.\u003c/p\u003e\u003cp\u003eThough the benefits conferred by the \u003cem\u003epha\u003c/em\u003e clusters to the microsymbiont and/or host remain to be elucidated, we expect that coevolution of these clusters in rhizobia with specific hosts can tune their activity to limit their negative impact on symbiosis, for example, through ensuring rapid flux through toxic pathway intermediates. Indeed, HM006 was isolated from \u003cem\u003eM. truncatula\u003c/em\u003e nodules grown with soil from its native range (\u003cb\u003e33, 52\u003c/b\u003e). When HM006 is present in \u003cem\u003eM. truncatula\u003c/em\u003e, although the genes are highly expressed, we observes different gene expression patterns across the region relative to HM006 bacteroids in \u003cem\u003eM. sativa\u003c/em\u003e nodules (\u003cb\u003eSupplemental Figure S8\u003c/b\u003e), which may reflect the ability to fine tune the pathways through adaptation of transcriptional control. In addition, other genetic elements in the loci may participate in limiting the impact to specific hosts (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e4\u003c/span\u003eC).\u003c/p\u003e\u003cp\u003ePartner quality differences in nature are widespread in many if not all rhizobium/legume systems (\u003cb\u003e53\u0026ndash;55\u003c/b\u003e). The replicon transplantation experiments described in this report revealed a correlation of partner quality phenotypes between \u003cem\u003eM. sativa\u003c/em\u003e and the pSymA replicons of the tested strains. Thus pSymA (and perhaps the mobile symbiotic element in many rhizobia) likely represents the primary reservoir for accessary genes influencing partner quality in \u003cem\u003eSinorhizobium\u003c/em\u003e, although accessory genes on pSymB or additional plasmids can also influence symbiotic outcomes (\u003cb\u003e21, 56\u003c/b\u003e). Given the much higher degree of genetic variation on extrachromosomal replicons relative to the chromosome (\u003cb\u003e11, 22, 32, 57, 58\u003c/b\u003e), it is unsurprising these elements contain important determinants of the partner quality differences observed across different strains. Certainly, many mechanisms exist for the diversity in partner quality phenotypes observed between rhizobia and their different hosts, distributed through a reshuffling of genetic modules that engender metabolic flexibility but also can render individual partnerships inefficient (\u003cb\u003e12, 59\u003c/b\u003e). This study suggests that along with diversity in symbiotic and immune signaling (\u003cb\u003e60\u0026ndash;63\u003c/b\u003e), metabolic incompatibility may be a key theme governing the diversity of these interactions we observe in nature.\u003c/p\u003e"},{"header":"Materials and Methods","content":"\u003cdiv id=\"Sec5\" class=\"Section2\"\u003e\u003ch2\u003eGenetic Manipulations\u003c/h2\u003e\u003cp\u003eBacterial strains and plasmids used in this work are listed in \u003cb\u003eSupplemental Table \u003cspan refid=\"MOESM1\" class=\"InternalRef\"\u003eS1\u003c/span\u003e\u003c/b\u003e. Media used for routine growth and selection of microbes included LB for \u003cem\u003eEscherichia coli\u003c/em\u003e, LBmc for \u003cem\u003eS. meliloti\u003c/em\u003e, and YPD or SC defined media for \u003cem\u003eSaccharomyces cereviciae\u003c/em\u003e. Antibiotics or supplements were added at routine concentrations when appropriate and details are summarized in the \u003cb\u003eSupplemental Methods\u003c/b\u003e.\u003c/p\u003e\u003cp\u003eWe routinely used a triparental mating protocol to transfer plasmids between donor and recipient strains by conjugation using the \u003cem\u003eE. coli\u003c/em\u003e helper strain MT616 to drive mobilization (\u003cb\u003e16\u003c/b\u003e). Transfer of extrachromosomal replicons between \u003cem\u003eS. meliloti\u003c/em\u003e strains also utilized conjugation; either via an integrated \u003cem\u003eoriT\u003c/em\u003e with triparental mating, or through the use of \u003cem\u003erctB\u003c/em\u003e in the donor strain to drive biparental conjugation (\u003cb\u003e1\u003c/b\u003e). Deletions were made by FLP-\u003cem\u003eFRT\u003c/em\u003e recombination (\u003cb\u003e64\u003c/b\u003e), wherein \u003cem\u003eFRT\u003c/em\u003e sites were integrated flanking a target region by homologous recombination, and subsequent excision of the region was catalyzed by introduction of FLP recombinase. Alternatively, region capture from \u003cem\u003eS. meliloti\u003c/em\u003e involved a quadriparental mating with \u003cem\u003eS. meliloti\u003c/em\u003e donor, \u003cem\u003eE. coli\u003c/em\u003e helper, \u003cem\u003eE. coli\u003c/em\u003e with a FLP-recombinase plasmid, and an \u003cem\u003eE. coli\u003c/em\u003e recipient. FLP-\u003cem\u003eFRT\u003c/em\u003e recombination was also used to introduce gene clusters into the genome by catalyzing recombination between an \u003cem\u003eFRT\u003c/em\u003e site in a plasmid containing the cluster and a single \u003cem\u003eFRT\u003c/em\u003e site in a landing pad in the genome (\u003cb\u003e14\u003c/b\u003e). We used Φ M12 bacteriophage to transfer insertions between \u003cem\u003eS. meliloti\u003c/em\u003e backgrounds using transduction (\u003cb\u003e49\u003c/b\u003e). Selection of desired events involved the use of antibiotics as well as prototrophic selection. Further details of genetic manipulations are included in \u003cb\u003eSupplemental Methods.\u003c/b\u003e Verification of genetic modifications was performed by PCR using primers in \u003cb\u003eSupplemental Tables S4 to S7\u003c/b\u003e or by whole genome sequencing.\u003c/p\u003e\u003c/div\u003e\n\u003ch3\u003ePlasmid Construction\u003c/h3\u003e\n\u003cp\u003ePlasmids for homologous recombination were routinely assembled using Level 1 (BsaI) Golden Gate cloning into suicide destination vectors that follow the BEVA architecture (\u003cb\u003e66\u003c/b\u003e). Inserts were PCR amplified with primers found in \u003cb\u003eSupplemental Table S3.\u003c/b\u003e Plasmids for gene cluster integration were assembled by yeast recombineering as previously described (\u003cb\u003e66, 67\u003c/b\u003e). Primers for amplification of regions that were combined in the assembly are in \u003cb\u003eSupplemental Table S8.\u003c/b\u003e Further details of plasmid construction are available in \u003cb\u003eSupplemental Methods.\u003c/b\u003e Plasmids were verified by restriction digest or whole plasmid sequencing.\u003c/p\u003e\n\u003ch3\u003eSymbiotic Assays\u003c/h3\u003e\n\u003cp\u003ePlant materials used in this work included \u003cem\u003eMedicago sativa\u003c/em\u003e cv. Iroquois (alfalfa) and \u003cem\u003eMedicago truncatula\u003c/em\u003e A17 (barrel medic). Plant assays were conducted in Leonard jar assembles with a 1:1 (wt:wt) mixture of sand and vermiculite with 250 mL of Jensen\u0026rsquo;s medium (\u003cb\u003e68\u003c/b\u003e). Four to six seedlings were added to each Leonard jar and inoculated two days after with 10 mL of sterile water containing 100 \u0026micro;L of dense overnight culture. \u003cem\u003eM. sativa\u003c/em\u003e seeds were surface sterilized for five minutes with 95% ethanol and for 20 minutes with 2.5% sodium hypochlorite. Seeds were then rinsed with sterile water for one hour, with water replaced every 15 minutes. \u003cem\u003eM. sativa\u003c/em\u003e seeds were placed on 1.5% water agar, placed in the dark at room temperature, and allowed two days to germinate. \u003cem\u003eM. truncatula\u003c/em\u003e seeds were first scarified with sulfuric acid for 10 minutes. After scarifying, seeds were rinsed 10 times with sterile water and then proceeded to be surface sterilized with 2.5% sodium hypochlorite for two minutes. Seeds were rinsed another 10 times with sterile water. \u003cem\u003eM. truncatula\u003c/em\u003e seeds were placed on 1.5% water agar and placed in the dark at 4\u0026deg;C for three days. The seeds were then moved to room temperature and allowed two more days to germinate. Four to six seedlings were added to each Leonard jar and inoculated two days after with 10 mL of sterile water containing 100 \u0026micro;L of dense overnight culture.\u003c/p\u003e\u003cp\u003ePlants were grown in Conviron Gen1000/2000 growth chambers programmed with a day cycle of 18 hours at 21\u0026deg;C with maximum light followed by a night cycle of six hours at 17\u0026deg;C with no light. Plants were allowed 42 days post-inoculation to grow. After the 42-day period, shoots were cut from roots and placed in a drying oven for one week, after which they were weighed for shoot dry weight. Fix\u003csup\u003e+\u003c/sup\u003e (pink) nodules were picked from roots, counted, and weighed immediately for average nodule fresh weight. Collected nodules were placed in an air-tight container with desiccant packets and let dry for a week, after which they were weighed for nodule dry weight. In some experiments HP6890 gas chromatograph was used to perform gas chromatography for acetylene reduction assays as previously described (\u003cb\u003e69\u003c/b\u003e). This assay was performed immediately after shoots were detached from the root system. A summary of all plant data collected is included in a \u003cb\u003eSupplemental Data\u003c/b\u003e file.\u003c/p\u003e\u003cdiv id=\"Sec8\" class=\"Section2\"\u003e\u003ch2\u003eFlow Cytometry\u003c/h2\u003e\u003cp\u003eNodules pooled from four plants from one pot were macerated in 750 \u0026micro;L of phosphate-ascorbate buffer (PAB) using a mortar and pestle. An additional 750 \u0026micro;L of PAB was used to rinse and collect any additional bacteroids from the mortar. After vortexing briefly, samples were spun at 200g for 5 minutes to spin down residual plant debris. The resulting supernatant was then pelleted and resuspended in 100 \u0026micro;L of fixative (30% molecular-grade ethanol). After 15 minutes of incubation, samples were pelleted to remove the fixative and resuspended in 500 \u0026micro;L of PBS.\u003c/p\u003e\u003cp\u003eA portion of each sample was pooled to create the three control samples: unstained, DAPI-only, and Nile Red only. Prior to staining, samples were diluted 1:10 in PBS. 1 \u0026micro;L of 10% Nile Red in DMSO and 1 \u0026micro;L of DAPI (1 mg/mL in PBS) were added to samples and allowed 15 minutes to incubate in darkness. Following staining, samples were again diluted 1:10 in PBS, giving a final 1:100 sample dilution. Cells were analyzed on a Beckman Cytoflex S flow cytometer equipped with violet, yellow, red, and blue lasers. The 561 nm yellow laser was used to excite Nile Red, and fluorescence was detected using the 610/20 bandpass filter. DAPI was excited using the 405nm violet laser and detected with the 450/45 bandpass filter. DAPI-positive bacteroids were gated and analyzed for Nile Red fluorescence intensity. Details on the gating schemes are described in \u003cb\u003eSupplemental Figures S19 and S20\u003c/b\u003e.\u003c/p\u003e\u003c/div\u003e\n\u003ch3\u003eElectron Microscopy\u003c/h3\u003e\n\u003cp\u003eSpecimens for transmission electron microscopy were fixed in 2.5% glutaraldehyde in sodium phosphate buffer, postfixed in 2% osmium tetroxide, dehydrated through a graded acetone series including saturated uranyl acetate in 70% acetone, and embedded in Spurr\u0026rsquo;s resin polymerized at 60\u003csup\u003eO\u003c/sup\u003eC for 24 hours. Embedded samples were sectioned at 70\u0026ndash;90 nm thickness on an RMC MT XL ultramicrotome. Ultrathin sections were collected on copper grids with a formvar-carbon supporting film, then stained with Reynolds\u0026rsquo; lead citrate. Observation and imaging were performed on a JEOL JEM-1400 electron microscope operating at 120 kV and equipped with an AMT NanoSprint 15L bottom-mount camera.\u003c/p\u003e\n\u003ch3\u003eGUS Staining and Microscopic Observations\u003c/h3\u003e\n\u003cp\u003eThe promoter regions of the \u003cem\u003epha\u003c/em\u003e clusters, \u003cem\u003enifH\u003c/em\u003e, and \u003cem\u003enodA\u003c/em\u003e were cloned via Golden Gate Cloning. Plasmids and strains used are listed in \u003cb\u003eSupplemental Table \u003cspan refid=\"MOESM1\" class=\"InternalRef\"\u003eS1\u003c/span\u003e\u003c/b\u003e. Primers used for promoter region amplification are listed in \u003cb\u003eSupplemental Table S3\u003c/b\u003e. The nodules were fixed in 100 mM sodium-phosphate buffer with 0.1% Triton X-100 and then in 1.5% glutaraldehyde in 100 mM phosphate buffer under vacuum for 30 minutes and 90 minutes. After that, the nodules were embedded in 3% (w/v) low-melting agarose and sectioned into 100-\u0026micro;m slices using vibrating microtome VF-210-0Z (Precisionary, USA). The slices were then vacuum immersed in GUS Staining Solution for 30 min. The solution contains 2 mM 5-bromo-4-chloro-3-indoxyl-β-d-glucuronide cyclohexyl ammonium salt (X-Gluc), 0.2% Triton X-100, and 2 mM each of K3[Fe(CN)6] and K4[Fe(CN)6] in 50 mM sodium phosphate buffer (pH 7.2). After incubating at 37\u0026deg;C for 2 hours, the stained nodule slices were observed using a BioTek Cytation 5 Cell Imaging Multimode Reader (Agilent, USA).\u003c/p\u003e\u003cdiv id=\"Sec11\" class=\"Section2\"\u003e\u003ch2\u003eEnzyme Assays\u003c/h2\u003e\u003cp\u003eA NADPH/NADH colorimetric assay was used to determine the Km and Vmax values of GhbD reacting with succinate semialdehyde (SSA). Individual 100 \u0026micro;L reactions comprised of purified GhbD (1 \u0026micro;g), NADPH (1 mM) or NADH (1 mM), SSA (0, 0.125, 0.25, 0.5, 1, 2.5, 5, or 10 mM), and reaction buffer (50 mM Na\u003csub\u003e3\u003c/sub\u003ePO\u003csub\u003e4\u003c/sub\u003e pH 7.0, 200 mM NaCl, 10 mM MgCl\u003csub\u003e2\u003c/sub\u003e). The 340 nm absorbance value of every reaction occurring at 28\u0026deg;C was measured in 1-minute intervals for 11 minutes starting immediately after the addition of GhbD. Absorbance values were read using a BioTek Cytation 5 Cell Imaging Multimode Reader by Agilent. Simple linear regression tests were completed with absorbance values over time for specific substrate concentrations. The resulting slopes, with a R\u003csup\u003e2\u003c/sup\u003e value\u0026thinsp;\u0026gt;\u0026thinsp;0.95, were used to determine the rate of reaction per mg of GhbD present. These rates, at different substrate concentrations, were used in the Michaelis-Menten equation to determine the Km and Vmax values for GhbD with SSA and NADPH/NADH. GraphPad Prism Version 10.4.1 was used to perform simple linear regression and non-linear regression tests.\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec12\" class=\"Section2\"\u003e\u003ch2\u003eMetabolomics\u003c/h2\u003e\u003cp\u003eMetabolomics analysis of plant tissue samples was performed using a 4-channel chemical isotope labeling (CIL) LC-MS method (\u003cb\u003e70\u003c/b\u003e). Normalization and labeling kits were obtained from Nova Medical Testing Inc. (Edmonton, AB, Canada). Each tissue sample was homogenized using a bead beater with steel beads, followed by the addition of ice-cold methanol/water (4:1, v/v) for metabolite extraction. The resulting extracts were normalized and labeled according to the protocols provided with the CIL kits.\u003c/p\u003e\u003cp\u003eLC-MS analysis was carried out using a Thermo Vanquish UHPLC system (ThermoFisher, Waltham, MA, US) coupled to a Bruker Impact II QTOF mass spectrometer (Bremen, Germany). Chromatographic separation was performed on an Agilent Eclipse Plus C18 reversed-phase column (150 \u0026times; 2.1 mm, 1.8 \u0026micro;m particle size) maintained at 40\u0026deg;C. The mobile phases consisted of 0.1% (v/v) formic acid in water (A) and 0.1% (v/v) formic acid in acetonitrile (B). The gradient program was as follows: 0 min, 25% B; 10 min, 99% B; 15 min, 99% B; 15.1 min, 25% B; 18 min, 25% B. The flow rate was 400 \u0026micro;L/min. Mass spectra were acquired at 1 Hz over a mass range of m/z 220\u0026ndash;1000. LC-MS data were processed using DataAnalysis (Bruker Daltonics) and IsoMS Pro (Nova Medical Testing Inc.). Manual verification of the key compound gamma-hydroxybutyrate (GHB) was performed using an authentic standard on an alternative instrumental platform. Specifically, MS/MS spectra and retention time (RT) data from the samples were compared to those of the standard, both acquired on a Dionex 3000 LC system coupled with Orbitrap Q Exactive HF mass spectrometer (ThermoFisher, Waltham, MA, US), using the same LC method and column. The MS/MS spectra of GHB in the samples, collected at multiple collision energies, matched those of the authentic standard, confirming the compound\u0026rsquo;s identity in the sample.\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec13\" class=\"Section2\"\u003e\u003ch2\u003eDNA and RNA sequencing\u003c/h2\u003e\u003cp\u003eGenome resequencing of hybrid strains and nodule RNA sequencing was performed by the Joint Genome Institute using Illumina NovaSeq technology. For RNA sequencing, bacteroids were isolated from nodules collected directly onto dry ice and flesh frozen in liquid nitrogen from one pot per replicate. RNA was extracted using a QIAgen RNeasy Plus Mini Kit, and underwent on-column DNase treatment. Libraries were prepared using Illumina\u0026rsquo;s Ribo-Zero rRNA removal kit and TruSeq Standard Total RNA HT sample prep kit. Raw fastq reads were filtered and trimmed according to the JGI QC pipeline, and mapped to the reference genome. Raw read counts were normalized to determine transcripts per million for each gene in the reference. Further details of DNA and RNA extraction, sequencing and bioinformatics analysis are included in \u003cb\u003eSupplemental Methods.\u003c/b\u003e Raw data was deposited in NCBI SRA using accessions in \u003cb\u003eSupplemental Table S13.\u003c/b\u003e Routine whole plasmid and whole genome resequencing to verify constructs and genetic manipulations was performed by Plasmidsaurus using Oxford Nanopore technology.\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec14\" class=\"Section2\"\u003e\u003ch2\u003eComparative Genomics\u003c/h2\u003e\u003cp\u003eAnalysis of homology between genes within clusters of interest was performed using clinker (Min. alignment sequence identity 0.3) and visualized with clustermap.js (\u0026gt;\u0026thinsp;0.5 alignment sequence identity shown) on the CAGECAT web server \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://cagecat.bioinformatics.nl/\u003c/span\u003e\u003cspan address=\"https://cagecat.bioinformatics.nl/\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (\u003cb\u003e71\u003c/b\u003e). Whole genome alignments and delineation of locally colinear gene blocks were performed using the MAUVE (\u003cb\u003e72\u003c/b\u003e) plugin in the Geneious software with default parameters (automatically calculate seed weight and minimum LCB score with MUSCLE3.6 gapped aligner).\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec15\" class=\"Section2\"\u003e\u003ch2\u003eData Collection and Statistical Analysis\u003c/h2\u003e\u003cp\u003ePlant experiments included uninoculated controls that were screened for the absence of nodules to confirm lack of contamination before any data collection. Individual replicates that makeup datapoints or were used for RNAseq or flow cytometry consisted of the average of all plants in one pot (six for \u003cem\u003eM. sativa\u003c/em\u003e and four for \u003cem\u003eM. truncatula\u003c/em\u003e). For statistical analysis we used a one-way ANOVA test with Dunnett\u0026rsquo;s multiple comparison test to establish significant differences between replicates. All data shown was significant as determined by one-way ANOVAs. In the figures, results from post hoc significance testing are indicated as *\u003cem\u003eP\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;0.0332, **\u003cem\u003eP\u0026thinsp;\u0026lt;\u003c/em\u003e\u0026thinsp;0.0021, ***\u003cem\u003eP\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;0.0002, ****\u003cem\u003eP\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;0.0001.\u003c/p\u003e"},{"header":"Declarations","content":"\u003cp\u003e\u003cstrong\u003eAcknowledgements and Funding\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eWe are grateful to Michael Sadowsky for the gift of \u003cem\u003eS. meliloti\u0026nbsp;\u003c/em\u003estrains. We thank Megan Ramsett for research and logistical support throughout the project. We thank Christa Pennachio for project coordination and Daniel Peterson (Joint Genome Institute) for RNAseq data analysis associated with JGI New Investigator proposal:10.46936/10.25585/60000503. We thank Scott Payne and Jayma Moore (NDSU Electron Microscopy Center core facility) for technical support with electron microscopy. We are grateful to the Geddes Lab and members of the Department of Microbiological Sciences, NDSU for critical review of the manuscript. This work was supported by startup funding provided to B. A. Geddes from the ND Agricultural Experiment Station, a New Innovator in Food \u0026amp; Agricultural Research (FFAR) grant to B. A. Geddes ID: FF-NIA21-0000000061, a Collaborative Research-PGR grant (NSF) to B. A. Geddes ID: Award number 2243818, an NSERC Discovery grant to T. M. Finan, and a NIFA Predoctoral Fellowship to G. J. Levin Grant Number: 2024-67011-42916. DNA and RNA sequencing work (proposal:10.46936/10.25585/60000503) conducted by the U.S. Department of Energy Joint Genome Institute, a DOE Office of Science User Facility, is supported by the Office of Science of the U.S. Department of Energy operated under Contract No. DE-AC02-05CH11231. Electron microscopy work reported in this publication was supported by the Office of Research Infrastructure Programs of the National Institutes of Health under award number S10OD030309. \u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eCompeting Interests\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eBarney Geddes is a cofounder of Lilac Agriculture, Inc.\u003c/p\u003e\u003ch2\u003eAuthor Contributions\u003c/h2\u003e\u003cp\u003eG. J. L., B. A. G., and T. M. F. contributed to experimental design. G. J. L., N. V., C. X. L., S. H., A. L., S. Z., and B. A. G. performed the experiments. G. J. L., B. A. G., S. H., S. Z., L. L., G. C. dC and T. M. F wrote the manuscript. T. M. F. and B. A. G. supervised the work.\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\n\u003cli\u003eA. J. Weisberg, \u003cem\u003eet al.\u003c/em\u003e, Pangenome Evolution Reconciles Robustness and Instability of Rhizobial Symbiosis. \u003cem\u003emBio\u003c/em\u003e \u003cstrong\u003e13\u003c/strong\u003e, e00074-22 (2022).\u003c/li\u003e\n\u003cli\u003eS. S. Porter, E. L. Simms, Selection for Cheating Across Disparate Environments in the Legume-Rhizobium Mutualism. \u003cem\u003eEcology Letters\u003c/em\u003e \u003cstrong\u003e17\u003c/strong\u003e, 1121\u0026ndash;1129 (2014).\u003c/li\u003e\n\u003cli\u003eK. D. Heath, J. R. Stinchcombe, Explaining Mutualism Variation: A New Evolutionary Paradox? \u003cem\u003eEvolution\u003c/em\u003e \u003cstrong\u003e68\u003c/strong\u003e, 309\u0026ndash;317 (2014).\u003c/li\u003e\n\u003cli\u003eE. W. Triplett, M. J. Sadowsky, Genetics of Competition for Nodulation of Legumes. \u003cem\u003eAnnu. Rev. Microbiol.\u003c/em\u003e \u003cstrong\u003e46\u003c/strong\u003e, 399\u0026ndash;428 (1992).\u003c/li\u003e\n\u003cli\u003eL. T. Burghardt, G. C. diCenzo, The Evolutionary Ecology of Rhizobia: Multiple Facets of Competition Before, During, and After Symbiosis with Legumes. \u003cem\u003eCurrent Opinion in Microbiology\u003c/em\u003e \u003cstrong\u003e72\u003c/strong\u003e (2023).\u003c/li\u003e\n\u003cli\u003eM. Mendoza-Su\u0026aacute;rez, S. U. Andersen, P. S. Poole, C. S\u0026aacute;nchez-Ca\u0026ntilde;izares, Competition, Nodule Occupancy, and Persistence of Inoculant Strains: Key Factors in the Rhizobium-Legume Symbioses. \u003cem\u003eFront. Plant Sci.\u003c/em\u003e \u003cstrong\u003e12\u003c/strong\u003e, 690567 (2021).\u003c/li\u003e\n\u003cli\u003eR. F. Denison, Legume Sanctions and the Evolution of Symbiotic Cooperation by Rhizobia. \u003cem\u003eThe American Naturalist\u003c/em\u003e \u003cstrong\u003e156\u003c/strong\u003e, 567\u0026ndash;576 (2000).\u003c/li\u003e\n\u003cli\u003eA. Westhoek, \u003cem\u003eet al.\u003c/em\u003e, Conditional Sanctioning in a Legume\u0026ndash;\u003cem\u003eRhizobium\u003c/em\u003e Mutualism. \u003cem\u003eProc. Natl. Acad. Sci. U.S.A.\u003c/em\u003e \u003cstrong\u003e118\u003c/strong\u003e, e2025760118 (2021).\u003c/li\u003e\n\u003cli\u003eE. T. Kiers, R. A. Rousseau, S. A. West, R. F. Denison, Host Sanctions and the Legume\u0026ndash;\u003cem\u003eRhizobium\u003c/em\u003e Mutualism. \u003cem\u003eNature\u003c/em\u003e \u003cstrong\u003e425\u003c/strong\u003e, 78\u0026ndash;81 (2003).\u003c/li\u003e\n\u003cli\u003eA. P. Montoya, \u003cem\u003eet al.\u003c/em\u003e, Hosts Winnow Symbionts with Multiple Layers of Absolute and Conditional Discrimination Mechanisms. \u003cem\u003eProc. R. Soc. B.\u003c/em\u003e \u003cstrong\u003e290\u003c/strong\u003e, 20222153 (2023).\u003c/li\u003e\n\u003cli\u003eA. B. Riley, M. A. Grillo, B. Epstein, P. Tiffin, K. D. Heath, Discordant Population Structure Among Rhizobium Divided Genomes and Their Legume Hosts. \u003cem\u003eMolecular Ecology\u003c/em\u003e \u003cstrong\u003e32\u003c/strong\u003e, 2646\u0026ndash;2659 (2023).\u003c/li\u003e\n\u003cli\u003eM. R. Riaz, \u003cem\u003eet al.\u003c/em\u003e, Mobile Gene Clusters and Coexpressed Plant\u0026ndash;Rhizobium Pathways Drive Partner Quality Variation in Symbiosis. \u003cem\u003eProc. Natl. Acad. Sci. U.S.A.\u003c/em\u003e \u003cstrong\u003e122\u003c/strong\u003e, e2411831122 (2025).\u003c/li\u003e\n\u003cli\u003eG. E. D. Oldroyd, J. D. Murray, P. S. Poole, J. A. Downie, The Rules of Engagement in the Legume-Rhizobial Symbiosis. \u003cem\u003eAnnu. Rev. Genet.\u003c/em\u003e \u003cstrong\u003e45\u003c/strong\u003e, 119\u0026ndash;144 (2011).\u003c/li\u003e\n\u003cli\u003eB. A. Geddes, \u003cem\u003eet al.\u003c/em\u003e, Minimal Gene Set from \u003cem\u003eSinorhizobium\u003c/em\u003e ( \u003cem\u003eEnsifer\u003c/em\u003e ) \u003cem\u003emeliloti\u003c/em\u003e pSymA Required for Efficient Symbiosis with \u003cem\u003eMedicago\u003c/em\u003e. \u003cem\u003eProc. Natl. Acad. Sci. U.S.A.\u003c/em\u003e \u003cstrong\u003e118\u003c/strong\u003e, e2018015118 (2021).\u003c/li\u003e\n\u003cli\u003eG. C. diCenzo, A. M. MacLean, B. Milunovic, G. B. Golding, T. M. Finan, Examination of Prokaryotic Multipartite Genome Evolution through Experimental Genome Reduction. \u003cem\u003ePLoS Genet\u003c/em\u003e \u003cstrong\u003e10\u003c/strong\u003e, e1004742 (2014).\u003c/li\u003e\n\u003cli\u003eT. M. Finan, B. Kunkel, G. F. De Vos, E. R. Signer, Second Symbiotic Megaplasmid in \u003cem\u003e Rhizobium\u003c/em\u003e \u003cem\u003emeliloti\u003c/em\u003e Carrying Exopolysaccharide and Thiamine Synthesis Genes. \u003cem\u003eJ Bacteriol\u003c/em\u003e \u003cstrong\u003e167\u003c/strong\u003e, 66\u0026ndash;72 (1986).\u003c/li\u003e\n\u003cli\u003eY. Kawaharada, \u003cem\u003eet al.\u003c/em\u003e, Receptor-Mediated Exopolysaccharide Perception Controls Bacterial Infection. \u003cem\u003eNature\u003c/em\u003e \u003cstrong\u003e523\u003c/strong\u003e, 308\u0026ndash;312 (2015).\u003c/li\u003e\n\u003cli\u003eJ. L. Zimmerman, W. W. Szeto, F. M. Ausubel, Molecular Characterization of Tn5-Induced Symbiotic (Fix-) Mutants of \u003cem\u003eRhizobium\u003c/em\u003e \u003cem\u003emeliloti\u003c/em\u003e. \u003cem\u003eJ Bacteriol\u003c/em\u003e \u003cstrong\u003e156\u003c/strong\u003e, 1025\u0026ndash;1034 (1983).\u003c/li\u003e\n\u003cli\u003eH. M. Meade, S. R. Long, G. B. Ruvkun, S. E. Brown, F. M. Ausubel, Physical and Genetic Characterization of Symbiotic and Auxotrophic Mutants of \u003cem\u003eRhizobium\u003c/em\u003e \u003cem\u003emeliloti\u003c/em\u003e Induced by Transposon Tn5 Mutagenesis. \u003cem\u003eJ Bacteriol\u003c/em\u003e \u003cstrong\u003e149\u003c/strong\u003e, 114\u0026ndash;122 (1982).\u003c/li\u003e\n\u003cli\u003eS. Yang, F. Tang, M. Gao, H. B. Krishnan, H. Zhu, \u003cem\u003eR\u003c/em\u003e Gene-Controlled Host Specificity in the Legume\u0026ndash;Rhizobia Symbiosis. \u003cem\u003eProc. Natl. Acad. Sci. U.S.A.\u003c/em\u003e \u003cstrong\u003e107\u003c/strong\u003e, 18735\u0026ndash;18740 (2010).\u003c/li\u003e\n\u003cli\u003eJ. Liu, \u003cem\u003eet al.\u003c/em\u003e, Paired \u003cem\u003eMedicago\u003c/em\u003e Receptors Mediate Broad-Spectrum Resistance to Nodulation by \u003cem\u003eSinorhizobium meliloti\u003c/em\u003e Carrying a Species-Specific Gene. \u003cem\u003eProc. Natl. Acad. Sci. U.S.A.\u003c/em\u003e \u003cstrong\u003e119\u003c/strong\u003e, e2214703119 (2022).\u003c/li\u003e\n\u003cli\u003eM. Galardini, \u003cem\u003eet al.\u003c/em\u003e, Exploring the Symbiotic Pangenome of the Nitrogen-Fixing Bacterium \u003cem\u003eSinorhizobium meliloti\u003c/em\u003e. \u003cem\u003eBMC Genomics\u003c/em\u003e \u003cstrong\u003e12\u003c/strong\u003e, 235 (2011).\u003c/li\u003e\n\u003cli\u003eL. T. Burghardt, \u003cem\u003eet al.\u003c/em\u003e, Select and Resequence Reveals Relative Fitness of Bacteria in Symbiotic and Free-Living Environments. \u003cem\u003eProc. Natl. Acad. Sci. U.S.A.\u003c/em\u003e \u003cstrong\u003e115\u003c/strong\u003e, 2425\u0026ndash;2430 (2018).\u003c/li\u003e\n\u003cli\u003eM. A. Mendoza-Su\u0026aacute;rez, \u003cem\u003eet al.\u003c/em\u003e, Optimizing \u003cem\u003eRhizobium-\u003c/em\u003eLegume Symbioses by Simultaneous Measurement of Rhizobial Competitiveness and N\u003csub\u003e2\u003c/sub\u003e Fixation in Nodules. \u003cem\u003eProc. Natl. Acad. Sci. U.S.A.\u003c/em\u003e \u003cstrong\u003e117\u003c/strong\u003e, 9822\u0026ndash;9831 (2020).\u003c/li\u003e\n\u003cli\u003eB. Epstein, \u003cem\u003eet al.\u003c/em\u003e, Combining GWAS and Population Genomic Analyses to Characterize Coevolution in a Legume-Rhizobia Symbiosis. \u003cem\u003eMolecular Ecology\u003c/em\u003e \u003cstrong\u003e32\u003c/strong\u003e, 3798\u0026ndash;3811 (2023).\u003c/li\u003e\n\u003cli\u003eB. Epstein, \u003cem\u003eet al.\u003c/em\u003e, Genome-Wide Association Analyses in the Model Rhizobium \u003cem\u003eEnsifer meliloti\u003c/em\u003e. \u003cem\u003emSphere\u003c/em\u003e \u003cstrong\u003e3\u003c/strong\u003e, e00386-18 (2018).\u003c/li\u003e\n\u003cli\u003eR. T. Batstone, A. M. O\u0026rsquo;Brien, T. L. Harrison, M. E. Frederickson, Experimental Evolution Makes Microbes More Cooperative with Their Local Host Genotype. \u003cem\u003eScience\u003c/em\u003e \u003cstrong\u003e370\u003c/strong\u003e, 476\u0026ndash;478 (2020).\u003c/li\u003e\n\u003cli\u003eI. J. Oresnik, S.-L. Liu, C. K. Yost, M. F. Hynes, Megaplasmid pRme2011a of \u003cem\u003eSinorhizobium meliloti\u003c/em\u003e Is Not Required for Viability. \u003cem\u003eJ Bacteriol\u003c/em\u003e \u003cstrong\u003e182\u003c/strong\u003e, 3582\u0026ndash;3586 (2000).\u003c/li\u003e\n\u003cli\u003eG. C. diCenzo, M. Zamani, B. Milunovic, T. M. Finan, Genomic Resources for Identification of the Minimal N\u003csub\u003e2\u003c/sub\u003e - Fixing Symbiotic Genome. \u003cem\u003eEnvironmental Microbiology\u003c/em\u003e \u003cstrong\u003e18\u003c/strong\u003e, 2534\u0026ndash;2547 (2016).\u003c/li\u003e\n\u003cli\u003eA. Checcucci, \u003cem\u003eet al.\u003c/em\u003e, Creation and Characterization of a Genomically Hybrid Strain in the Nitrogen-Fixing Symbiotic Bacterium \u003cem\u003eSinorhizobium meliloti\u003c/em\u003e. \u003cem\u003eACS Synth. Biol.\u003c/em\u003e \u003cstrong\u003e7\u003c/strong\u003e, 2365\u0026ndash;2378 (2018).\u003c/li\u003e\n\u003cli\u003eW. Reeve, \u003cem\u003eet al.\u003c/em\u003e, Complete Genome Sequence of the \u003cem\u003eMedicago\u003c/em\u003e Microsymbiont \u003cem\u003e Ensifer (Sinorhizobium) medicae\u003c/em\u003e Strain WSM419. \u003cem\u003eStandards in Genomic Sciences\u003c/em\u003e \u003cstrong\u003e2\u003c/strong\u003e, 77\u0026ndash;86 (2012).\u003c/li\u003e\n\u003cli\u003eM. Nelson, J. Guhlin, B. Epstein, P. Tiffin, M. J. Sadowsky, The Complete Replicons of 16 \u003cem\u003eEnsifer meliloti\u003c/em\u003e Strains Offer Insights into Intra- and Inter-Replicon Gene Transfer, Transposon-Associated Loci, and Repeat Elements. \u003cem\u003eMicrobial Genomics\u003c/em\u003e \u003cstrong\u003e4\u003c/strong\u003e (2018).\u003c/li\u003e\n\u003cli\u003eM. Sugawara, \u003cem\u003eet al.\u003c/em\u003e, Comparative Genomics of the Core and Accessory Genomes of 48 Sinorhizobium Strains Comprising Five Genospecies. \u003cem\u003eGenome Biol\u003c/em\u003e \u003cstrong\u003e14\u003c/strong\u003e, R17 (2013).\u003c/li\u003e\n\u003cli\u003eZ.-C. Yuan, R. Zaheer, T. M. Finan, Regulation and Properties of PstSCAB, a High-Affinity, High-Velocity Phosphate Transport System of \u003cem\u003eSinorhizobium meliloti\u003c/em\u003e. \u003cem\u003eJ Bacteriol\u003c/em\u003e \u003cstrong\u003e188\u003c/strong\u003e, 1089\u0026ndash;1102 (2006).\u003c/li\u003e\n\u003cli\u003eE. M. Lodwig, \u003cem\u003eet al.\u003c/em\u003e, Role of Polyhydroxybutyrate and Glycogen as Carbon Storage Compounds in Pea and Bean Bacteroids. \u003cem\u003eMPMI\u003c/em\u003e \u003cstrong\u003e18\u003c/strong\u003e, 67\u0026ndash;74 (2005).\u003c/li\u003e\n\u003cli\u003eW. C. Ratcliff, S. V. Kadam, R. F. Denison, Poly-3-Hydroxybutyrate (PHB) Supports Survival and Reproduction in Starving Rhizobia: PHB Increases Rhizobium Fitness. \u003cem\u003eFEMS Microbiology Ecology\u003c/em\u003e \u003cstrong\u003e65\u003c/strong\u003e, 391\u0026ndash;399 (2008).\u003c/li\u003e\n\u003cli\u003eA. Puppo, \u003cem\u003eet al.\u003c/em\u003e, Legume Nodule Senescence: Roles for Redox and Hormone Signaling in the Orchestration of the Natural Aging Process. \u003cem\u003eNew Phytologist\u003c/em\u003e \u003cstrong\u003e165\u003c/strong\u003e, 683\u0026ndash;701 (2005).\u003c/li\u003e\n\u003cli\u003eY. Chai, C. S. Tsai, H. Cho, S. C. Winans, Reconstitution of the Biochemical Activities of the AttJ Repressor and the AttK, AttL, and AttM Catabolic Enzymes of \u003cem\u003eAgrobacterium tumefaciens\u003c/em\u003e. \u003cem\u003eJ Bacteriol\u003c/em\u003e \u003cstrong\u003e189\u003c/strong\u003e, 3674\u0026ndash;3679 (2007).\u003c/li\u003e\n\u003cli\u003eP. A. Price, \u003cem\u003eet al.\u003c/em\u003e, Rhizobial Peptidase HrrP Cleaves Host-Encoded Signaling Peptides and Mediates Symbiotic Compatibility. \u003cem\u003eProc. Natl. Acad. Sci. U.S.A.\u003c/em\u003e \u003cstrong\u003e112\u003c/strong\u003e, 15244\u0026ndash;15249 (2015).\u003c/li\u003e\n\u003cli\u003eM. A. Trainer, T. C. Charles, The Role of PHB Metabolism in the Symbiosis of Rhizobia with Legumes. \u003cem\u003eAppl Microbiol Biotechnol\u003c/em\u003e \u003cstrong\u003e71\u003c/strong\u003e, 377\u0026ndash;386 (2006).\u003c/li\u003e\n\u003cli\u003eC. Wang, \u003cem\u003eet al.\u003c/em\u003e, Roles of Poly-3-Hydroxybutyrate (PHB) and Glycogen in Symbiosis of \u003cem\u003eSinorhizobium meliloti\u003c/em\u003e with \u003cem\u003eMedicago\u003c/em\u003e sp. \u003cem\u003eMicrobiology\u003c/em\u003e \u003cstrong\u003e153\u003c/strong\u003e, 388\u0026ndash;398 (2007).\u003c/li\u003e\n\u003cli\u003eJ. Prell, A. Bourd\u0026egrave;s, R. Karunakaran, M. Lopez-Gomez, P. Poole, Pathway of \u0026gamma;-Aminobutyrate Metabolism in \u003cem\u003eRhizobium leguminosarum\u003c/em\u003e 3841 and Its Role in Symbiosis. \u003cem\u003eJ Bacteriol\u003c/em\u003e \u003cstrong\u003e191\u003c/strong\u003e, 2177\u0026ndash;2186 (2009).\u003c/li\u003e\n\u003cli\u003eM. F. Dunn, Key Roles of Microsymbiont Amino Acid Metabolism in Rhizobia-Legume Interactions. \u003cem\u003eCritical Reviews in Microbiology\u003c/em\u003e \u003cstrong\u003e41\u003c/strong\u003e, 411\u0026ndash;451 (2015).\u003c/li\u003e\n\u003cli\u003eC. E. Flores-Tinoco, \u003cem\u003eet al.\u003c/em\u003e, Co-Catabolism of Arginine and Succinate Drives Symbiotic Nitrogen Fixation. \u003cem\u003eMolecular Systems Biology\u003c/em\u003e \u003cstrong\u003e16\u003c/strong\u003e, e9419 (2020).\u003c/li\u003e\n\u003cli\u003eG. C. diCenzo, M. Tesi, T. Pfau, A. Mengoni, M. Fondi, Genome-Scale Metabolic Reconstruction of the Symbiosis Between a Leguminous Plant and a Nitrogen-Fixing Bacterium. \u003cem\u003eNat Commun\u003c/em\u003e \u003cstrong\u003e11\u003c/strong\u003e, 2574 (2020).\u003c/li\u003e\n\u003cli\u003eM. A. Cevallos, S. Encarnaci\u0026oacute;n, A. Leija, Y. Mora, J. Mora, Genetic and Physiological Characterization of a \u003cem\u003eRhizobium etli\u003c/em\u003e Mutant Strain Unable to Synthesize Poly-Beta-Hydroxybutyrate. \u003cem\u003eJ Bacteriol\u003c/em\u003e \u003cstrong\u003e178\u003c/strong\u003e, 1646\u0026ndash;1654 (1996).\u003c/li\u003e\n\u003cli\u003eJ. I. Quelas, E. J. Mongiardini, J. Perez-Gimenez, G. Parisi, A. R. Lodeiro, Analysis of Two Polyhydroxyalkanoate Synthases in \u003cem\u003eBradyrhizobium japonicum\u003c/em\u003e USDA 110. \u003cem\u003eJournal of Bacteriology\u003c/em\u003e \u003cstrong\u003e195\u003c/strong\u003e, 3145\u0026ndash;3155 (2013).\u003c/li\u003e\n\u003cli\u003eJ. J. Terpolilli, \u003cem\u003eet al.\u003c/em\u003e, Lipogenesis and Redox Balance in Nitrogen-Fixing Pea Bacteroids. \u003cem\u003eJ Bacteriol\u003c/em\u003e \u003cstrong\u003e198\u003c/strong\u003e, 2864\u0026ndash;2875 (2016).\u003c/li\u003e\n\u003cli\u003eF. J. Bergersen, G. L. Turner, Bacteroids from Soybean Root Nodules: Accumulation of Poly-\u0026beta;-Hydroxybutyrate During Supply of Malate and Succinate in Relation to N2 Fixation in Flow-Chamber Reactions. \u003cem\u003eProceedings of the Royal Society B: Biological Sciences\u003c/em\u003e \u003cstrong\u003e240\u003c/strong\u003e, 39\u0026ndash;59 (1990).\u003c/li\u003e\n\u003cli\u003eM. A. Trainer, T. C. Charles, The Role of PHB Metabolism in the Symbiosis of Rhizobia with Legumes. \u003cem\u003eAppl Microbiol Biotechnol\u003c/em\u003e \u003cstrong\u003e71\u003c/strong\u003e, 377\u0026ndash;386 (2006).\u003c/li\u003e\n\u003cli\u003eC. C. M. Schulte, \u003cem\u003eet al.\u003c/em\u003e, Metabolic Control of Nitrogen Fixation in Rhizobium-Legume Symbioses. \u003cem\u003eSci. Adv.\u003c/em\u003e \u003cstrong\u003e7\u003c/strong\u003e, eabh2433 (2021).\u003c/li\u003e\n\u003cli\u003eK. A. Lesins, I. Lesins, \u003cem\u003eGenus Medicago (Leguminosae): A Taxogenetic Study\u003c/em\u003e (Springer Science \u0026amp; Business Media, 2012).\u003c/li\u003e\n\u003cli\u003eK. D. Heath, J. R. Stinchcombe, Explaining Mutualism Variation: A New Evolutionary Paradox? \u003cem\u003eEvolution\u003c/em\u003e \u003cstrong\u003e68\u003c/strong\u003e, 309\u0026ndash;317 (2014).\u003c/li\u003e\n\u003cli\u003eS. A. West, E. T. Kiers, I. Pen, R. F. Denison, Sanctions and Mutualism Stability: When Should Less Beneficial Mutualists be Tolerated? \u003cem\u003eJ of Evolutionary Biology\u003c/em\u003e \u003cstrong\u003e15\u003c/strong\u003e, 830\u0026ndash;837 (2002).\u003c/li\u003e\n\u003cli\u003eL. Salvaudon, T. Giraud, J. A. Shykoff, Genetic Diversity in Natural Populations: A Fundamental Component of Plant\u0026ndash;Microbe Interactions. \u003cem\u003eCurrent Opinion in Plant Biology\u003c/em\u003e \u003cstrong\u003e11\u003c/strong\u003e, 135\u0026ndash;143 (2008).\u003c/li\u003e\n\u003cli\u003eP. A. Price, \u003cem\u003eet al.\u003c/em\u003e, Rhizobial Peptidase HrrP Cleaves Host-Encoded Signaling Peptides and Mediates Symbiotic Compatibility. \u003cem\u003eProc. Natl. Acad. Sci. U.S.A.\u003c/em\u003e \u003cstrong\u003e112\u003c/strong\u003e, 15244\u0026ndash;15249 (2015).\u003c/li\u003e\n\u003cli\u003eX. Bailly, \u003cem\u003eet al.\u003c/em\u003e, Population Genomics of \u003cem\u003eSinorhizobium medicae\u003c/em\u003e Based on Low-Coverage Sequencing of Sympatric Isolates. \u003cem\u003eThe ISME Journal\u003c/em\u003e \u003cstrong\u003e5\u003c/strong\u003e, 1722\u0026ndash;1734 (2011).\u003c/li\u003e\n\u003cli\u003eB. Epstein, M. J. Sadowsky, P. Tiffin, Selection on Horizontally Transferred and Duplicated Genes in Sinorhizobium (Ensifer), the Root-Nodule Symbionts of Medicago. \u003cem\u003eGenome Biology and Evolution\u003c/em\u003e \u003cstrong\u003e6\u003c/strong\u003e, 1199\u0026ndash;1209 (2014).\u003c/li\u003e\n\u003cli\u003eA. J. Weisberg, \u003cem\u003eet al.\u003c/em\u003e, Pangenome Evolution Reconciles Robustness and Instability of Rhizobial Symbiosis. \u003cem\u003emBio\u003c/em\u003e \u003cstrong\u003e13\u003c/strong\u003e, e00074-22 (2022).\u003c/li\u003e\n\u003cli\u003eD. Wang, S. Yang, F. Tang, H. Zhu, Symbiosis Specificity in the Legume-Rhizobial Mutualism: Host Specificity in Root Nodule Symbiosis. \u003cem\u003eCellular Microbiology\u003c/em\u003e \u003cstrong\u003e14\u003c/strong\u003e, 334\u0026ndash;342 (2012).\u003c/li\u003e\n\u003cli\u003eS. S. Porter, S. E. Dupin, R. F. Denison, E. T. Kiers, J. L. Sachs, Host-Imposed Control Mechanisms in Legume\u0026ndash;Rhizobia Symbiosis. \u003cem\u003eNat Microbiol\u003c/em\u003e \u003cstrong\u003e9\u003c/strong\u003e, 1929\u0026ndash;1939 (2024).\u003c/li\u003e\n\u003cli\u003eJ. Liu, \u003cem\u003eet al.\u003c/em\u003e, Paired \u003cem\u003eMedicago\u003c/em\u003e Receptors Mediate Broad-Spectrum Resistance to Nodulation by \u003cem\u003eSinorhizobium meliloti\u003c/em\u003e Carrying a Species-Specific Gene. \u003cem\u003eProc. Natl. Acad. Sci. U.S.A.\u003c/em\u003e \u003cstrong\u003e119\u003c/strong\u003e, e2214703119 (2022).\u003c/li\u003e\n\u003cli\u003eZ. Bozsoki, \u003cem\u003eet al.\u003c/em\u003e, Ligand-Recognizing Motifs in Plant LysM Receptors are Major Determinants of Specificity. \u003cem\u003eScience\u003c/em\u003e \u003cstrong\u003e369\u003c/strong\u003e, 663\u0026ndash;670 (2020).\u003c/li\u003e\n\u003cli\u003eB. Milunovic, G. C. diCenzo, R. A. Morton, T. M. Finan, Cell Growth Inhibition Upon Deletion of Four Toxin-Antitoxin Loci from the Megaplasmids of \u003cem\u003eSinorhizobium meliloti\u003c/em\u003e. \u003cem\u003eJ Bacteriol\u003c/em\u003e \u003cstrong\u003e196\u003c/strong\u003e, 811\u0026ndash;824 (2014).\u003c/li\u003e\n\u003cli\u003eT. M. Finan, \u003cem\u003eet al.\u003c/em\u003e, General Transduction in \u003cem\u003eRhizobium meliloti\u003c/em\u003e. \u003cem\u003eJ Bacteriol\u003c/em\u003e \u003cstrong\u003e159\u003c/strong\u003e, 120\u0026ndash;124 (1984).\u003c/li\u003e\n\u003cli\u003eB. A. Geddes, M. A. Mendoza-Su\u0026aacute;rez, P. S. Poole, A Bacterial Expression Vector Archive (BEVA) for Flexible Modular Assembly of Golden Gate-Compatible Vectors. \u003cem\u003eFront. Microbiol.\u003c/em\u003e \u003cstrong\u003e9\u003c/strong\u003e, 3345 (2019).\u003c/li\u003e\n\u003cli\u003eR. D. Gietz, R. H. Schiestl, High-Efficiency Yeast Transformation Using the LiAc/SS Carrier DNA/PEG Method. \u003cem\u003eNat Protoc\u003c/em\u003e \u003cstrong\u003e2\u003c/strong\u003e, 31\u0026ndash;34 (2007).\u003c/li\u003e\n\u003cli\u003eJ. M. Vincent, \u003cem\u003eA Manual for the Practical Study of the Root-Nodule Bacteria\u003c/em\u003e (Blackwell Scientific Publications, 1970).\u003c/li\u003e\n\u003cli\u003eM. Zamani, T. M. Finan, A. Cowie, G. C. diCenzo, Proline Auxotrophy in \u003cem\u003eSinorhizobium meliloti\u003c/em\u003e Results in a Plant-Specific Symbiotic Phenotype. \u003cem\u003eMicrobiology\u003c/em\u003e \u003cstrong\u003e161\u003c/strong\u003e, 2341\u0026ndash;2351 (2015).\u003c/li\u003e\n\u003cli\u003eS. Zhao, H. Li, W. Han, W. Chan, L. Li, Metabolomic Coverage of Chemical-Group-Submetabolome Analysis: Group Classification and Four-Channel Chemical Isotope Labeling LC-MS. \u003cem\u003eAnal. Chem.\u003c/em\u003e \u003cstrong\u003e91\u003c/strong\u003e, 12108\u0026ndash;12115 (2019).\u003c/li\u003e\n\u003cli\u003eC. L. M. Gilchrist, Y.-H. Chooi, Clinker \u0026amp; clustermap.js: Automatic Generation of Gene Cluster Comparison Figures. \u003cem\u003eBioinformatics\u003c/em\u003e \u003cstrong\u003e37\u003c/strong\u003e, 2473\u0026ndash;2475 (2021).\u003c/li\u003e\n\u003cli\u003eA. C. E. Darling, B. Mau, F. R. Blattner, N. T. Perna, Mauve: Multiple Alignment of Conserved Genomic Sequence With Rearrangements. \u003cem\u003eGenome Res.\u003c/em\u003e \u003cstrong\u003e14\u003c/strong\u003e, 1394\u0026ndash;1403 (2004).\u003c/li\u003e\n\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":true,"hideJournal":false,"highlight":"","institution":"","isAcceptedByJournal":false,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"nature-portfolio","isNatureJournal":true,"hasQc":false,"allowDirectSubmit":false,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"","title":"Nature Portfolio","twitterHandle":"","acdcEnabled":false,"dfaEnabled":false,"editorialSystem":"ejp","reportingPortfolio":"","inReviewEnabled":true,"inReviewRevisionsEnabled":false},"keywords":"Symbiosis, Nitrogen Fixation, Genetics","lastPublishedDoi":"10.21203/rs.3.rs-7715224/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-7715224/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eNaturally occurring root-nodule bacteria (rhizobia) vary substantially in their effectiveness at promoting growth of different plant hosts via symbiotic nitrogen fixation. These variations in rhizobial partner quality have important implications for the productivity of nitrogen-fixing symbioses in natural and agricultural ecosystems, yet we have a limited understanding of the genetic basis for this variation. In a case of host-specific reduction in symbiotic effectiveness (N\u003csub\u003e2\u003c/sub\u003e-fixation) with \u003cem\u003eMedicago sativa\u003c/em\u003e, we identified the causative genetic elements from the pSymA replicon of \u003cem\u003eSinorhizobum meliloti\u003c/em\u003e HM006 and show them to be involved in polyhydroxyalkanoate (PHA) production in nitrogen-fixing bacteroids. Transfer of this gene region to a strain that forms an effective symbiosis with \u003cem\u003eMedicago sativa\u003c/em\u003e resulted in a complete loss of symbiotic N\u003csub\u003e2\u003c/sub\u003e-fixation. We showed the mechanism for symbiotic collapse is the diversion of succinate semialdehyde pools in the bacteroid to gamma-hydroxybutyrate (GHB) by an iron-containing dehydrogenase, GhbD. These findings reveal unexpected impacts of carbon metabolism changes in nodules on symbiont performance and provide a rare example of mechanism for variation in rhizobium partner quality, suggesting that host-specific metabolic incompatibility may be a key player in the variations in partner quality observed in nature.\u003c/p\u003e","manuscriptTitle":"Polyhydroxyalkanoate synthesis by Sinorhizobium meliloti drives a host-specific collapse in symbiosis with Medicago sativa","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-10-29 06:30:18","doi":"10.21203/rs.3.rs-7715224/v1","editorialEvents":[],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"nature-communications","isNatureJournal":true,"hasQc":false,"allowDirectSubmit":false,"externalIdentity":"NCOMMS","sideBox":"Learn more about [Nature Communications](http://www.nature.com/ncomms/)","snPcode":"","submissionUrl":"https://mts-ncomms.nature.com/","title":"Nature Communications","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"ejp","reportingPortfolio":"Nature Communications","inReviewEnabled":true,"inReviewRevisionsEnabled":false}}],"origin":"","ownerIdentity":"1024d8d0-51b9-4ff9-9fb2-2980436c5965","owner":[],"postedDate":"October 29th, 2025","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"under-review","subjectAreas":[{"id":55684506,"name":"Biological sciences/Microbiology/Microbial genetics/Bacterial genetics"},{"id":55684507,"name":"Biological sciences/Plant sciences/Plant symbiosis/Rhizobial symbiosis"}],"tags":[],"updatedAt":"2025-10-29T06:30:18+00:00","versionOfRecord":[],"versionCreatedAt":"2025-10-29 06:30:18","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-7715224","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-7715224","identity":"rs-7715224","version":["v1"]},"buildId":"8U1c8b4HqxoKbykW_rLl7","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}