Sinorhizobium medicae WSM419 Genes That Improve Symbiosis between Sinorhizobium meliloti Rm1021 and Medicago truncatula Jemalong A17 and in Other Symbiosis Systems

Abstract

Some soil bacteria, called rhizobia, can interact symbiotically with legumes, in which they form nodules on the plant roots, where they can reduce atmospheric dinitrogen to ammonia, a form of nitrogen that can be used by growing plants. Rhizobium-plant combinations can differ in how successful this symbiosis is: for example, Sinorhizobium meliloti Rm1021 forms a relatively ineffective symbiosis with Medicago truncatula Jemalong A17, but Sinorhizobium medicae WSM419 is able to support more vigorous plant growth. Using proteomic data from free-living and symbiotic S. medicae WSM419, we previously identified a subset of proteins that were not closely related to any S. meliloti Rm1021 proteins and speculated that adding one or more of these proteins to S. meliloti Rm1021 would increase its effectiveness on M. truncatula A17. Three genes, Smed_3503, Smed_5985, and Smed_6456, were cloned into S. meliloti Rm1021 downstream of the E. coli lacZ promoter. Strains with these genes increased nodulation and improved plant growth, individually and in combination with one another. Smed_3503, renamed iseA (increased symbiotic effectiveness), had the largest impact, increasing M. truncatula biomass by 61%. iseA homologs were present in all currently sequenced S. medicae strains but were infrequent in other Sinorhizobium isolates. Rhizobium leguminosarum bv. viciae 3841 containing iseA led to more nodules on pea and lentil. Split-root experiments with M. truncatula A17 indicated that S. meliloti Rm1021 carrying the S. medicae iseA is less sensitive to plant-induced resistance to rhizobial infection, suggesting an interaction with the plant's regulation of nodule formation. IMPORTANCE Legume symbiosis with rhizobia is highly specific. Rhizobia that can nodulate and fix nitrogen on one legume species are often unable to associate with a different species. The interaction can be more subtle. Symbiotically enhanced growth of the host plant can differ substantially when nodules are formed by different rhizobial isolates of a species, much like disease severity can differ when conspecific isolates of pathogenic bacteria infect different cultivars. Much is known about bacterial genes essential for a productive symbiosis, but less is understood about genes that marginally improve performance. We used a proteomic strategy to identify Sinorhizobium genes that contribute to plant growth differences that are seen when two different strains nodulate M. truncatula A17. These genes could also alter the symbiosis between R. leguminosarum bv. viciae 3841 and pea or lentil, suggesting that this approach identifies new genes that may more generally contribute to symbiotic productivity.

Keywords: Medicago truncatula; Sinorhizobium medicae WSM419; Sinorhizobium meliloti Rm1021; alfalfa; autoregulation of nodulation (AON); model legume; nitrogen fixation; root nodules; symbiosis.

FIG 1

Genes from S. medicae WSM419 increase the symbiotic productivity of S. meliloti Rm1021 with M. truncatula A17. Shown is the shoot dry weight of M. truncatula A17 inoculated with S. meliloti Rm1021 expressing the candidate genes (iseA, Smed_5985, and Smed_6456) independently and in different combinations. All data were collected at 28 dpi, the time of harvest (n = 18). Error bars indicate standard errors of the means (SEM). Significant differences (*, P < 0.05; **, P < 0.01; ***, P < 0.001) for Student's t test versus the vector control (pCPP30 and pSRKGm) The plasmids pCPP30 (Tc^r) and pSRKGm (Gm^r) are represented as pC and pS, respectively.

FIG 2

Expression pattern of iseA, Smed_5985, and Smed_6456 promoters. (A) Histochemical β-glucuronidase (GUS) staining of M. truncatula A17 root nodules from 10 dpi. The GUS staining of nodules induced by S. meliloti Rm1021 carrying no promoter-GUS, Pfla (flagellin)-GUS, PhrrP-GUS, PiseA-GUS, PSmed_5985-GUS, and PSmed_6456-GUS, respectively. Scale bar, 100 μm. (B) GUS expression levels in free-living S. meliloti Rm1021 cells after growth to log phase in MMNH₄ and LB. The constitutive Salmonella-derived trp promoter (Ptrp) was used as a positive control. Error bars represent standard errors of the means from three biological replicates.

FIG 3

These S. medicae WSM419 genes do not significantly improve the S. meliloti Rm1021 symbiosis with M. sativa. (A and B) Nodule number (A) and plant shoot dry weight data (B) 28 dpi for M. sativa inoculated with S. meliloti Rm1021 containing the pCPP30 vector control, pCPP30-iseA, pCPP30-Smed_5985, and pCPP30-Smed_6456, respectively (n = 16). Error bars indicate SEM. No significant differences were observed between transgenic strains and vector control (pCPP30 [Tc^r]) using Student's t test. Similar results were obtained with the pSRKGm (Gm^r) plasmid carrying the genes (data not shown).

FIG 4

Deleting the iseA gene from S. medicae WSM419 lowers symbiotic productivity with M. truncatula A17. (A and B) Nodule number (A) and plant shoot dry weight data (B) at 28 dpi for M. truncatula A17 plants infected with S. medicae WSM419, WSM419ΔiseA, WSM419ΔiseA(pCPP30), and WSM419ΔiseA(pCPP30-iseA) strains (n = 16). Error bars indicate SEM. Significant differences (*, P < 0.05; **, P < 0.01; ***, P < 0.001) for Student's t test are shown.

FIG 5

S. medicae WSM419 genes increased pea and lentil performance when expressed in R. leguminosarum bv. viciae 3841. (A and B) Number of nodules (A) and plant shoot dry weight (B) at 5 weeks postinoculation for pea (P. sativum cv. Green Arrow) inoculated with R. leguminosarum bv. viciae 3841 containing the pCPP30 vector control, pCPP30-iseA, pCPP30-Smed_5985, and pCPP30-Smed_6456, respectively (n = 16). (C and D) Number of nodules (C) and plant shoot dry weight (D) at 5 weeks postinoculation for lentil (Lens culinaris cv. Avondale) inoculated with R. leguminosarum bv. viciae 3841 containing the pCPP30 vector control, pCPP30-iseA, pCPP30-Smed_5985, and pCPP30-Smed_6456, respectively (n = 16). Error bars indicate SEM. Shown are significant differences (*, P < 0.05; **, P < 0.01; ***, P < 0.001) for Student's t test versus the vector control (pCPP30 [Tc^r]). Similar results were obtained with the pSRKGm (Gm^r) plasmid carrying the genes (data not shown).

FIG 6

iseA altered nodulation in an M. truncatula A17 split-root system. M. truncatula A17 roots were manipulated to produce equal lateral roots partitioned in two sections, represented by two paired bars. The early infected roots (represented by red) were inoculated 3 days earlier than the delayed infected roots (represented by blue). The split roots were inoculated with S. meliloti Rm1021 expressing pCPP30 and pCPP30-iseA in different combinations (n = 16). The numbers of nodules per root from both the early and delayed inoculated roots were counted 4 weeks after the delayed inoculation. Error bars indicate SEM. Significant differences (*, P < 0.05; **, P < 0.01; ***, P < 0.001) for Student's t test are indicated.