Host recognition in the Rhizobium-soybean symbiosis

Polar binding of Rhizobium japonicum to roots and root hairs of Glycine soja (L.) Sieb. and Zucc. is specifically inhibited by d-galactose and N-acetyl-d-galactosamine, haptens of Glycine max seed lectin. A protein, immunologically cross-reactive with the G. max seed lectin, is present in G. soja seed extracts. Peptide mapping of the purified G. max and G. soja lectins indicates that the two are similar in structure. Soybean lectin can be localized on the surface of both G. max and G. soja roots by indirect immunolatex techniques. These observations indicate that the Rhizobium-binding lectin, previously isolated from seeds, also is present on the root surface-the site of the initial steps in the infection. This lectin is capable of binding Rhizobium japonicum to the root.     

Host recognition in the Rhizobium-soybean symbiosis : evidence for the involvement of lectin in nodulation

Rhizobium japonicum mutant strain HS111 was previously shown to be defective in the rate of initiation of infection leading to subsequent nodule formation (1984 Plant Physiol 74: 84-89). Mutant strain HS111's defect in nodulation can be phenotypically reversed to wild type levels by pretreatment with root exudates from all soybean varieties that have been tested. The data indicate that lectin-Rhizobium interaction is necessary for the phenotypic reversal of the nodulation characteristics of mutant strain HS111. Pretreatment of strain HS111 with soybean seed lectin mimics the effect of root exudate pretreatment. In addition, the presence of 30 millimolar d-galactose, a hapten of soybean seed lectin, in the root exudate or soybean seed lectin pretreatment solution prevents enhancement of nodulation of strain HS111. Pretreatment of mutant strain HS111 in soybean root exudate which has had galactose-specific lectin(s) removed by affinity chromatography (affinity eluate) results in no enhancement of nodulation by strain HS111. Lectin(s) subsequently removed from the affinity column possesses 100% of the stimulatory activity originally found in the root exudate. Pretreatment of strain HS111 in root exudate from a soybean seed line (T102) known to lack seed lectin due to an insertion in the structural gene results in the reversal of the defective nodulation phenotype. This latter result indicates that the lectin found in soybean root exudate is genetically distinct from the seed lectin. It is apparently this root lectin that is involved in nodulation.     

Early recognition in the Rhizobium meliloti-alfalfa symbiosis: root exudate factor stimulates root adsorption of homologous rhizobia.

Adsorption of Rhizobium meliloti to alfalfa roots before their infection and nodule formation shows the specificity of the symbiotic association (G. Caetano-Anollés and G. Favelukes, Appl. Environ. Microbiol. 52:377-382, 1986). The time course of specific adsorption of R. meliloti (10(3) to 10(4) cells per ml) to roots shows an initial lag period of 3 h, suggesting that either or both symbionts must become conditioned for the adsorption process. Preincubation of R. meliloti L5-30 for 3 h with dialyzed alfalfa root exudate (RE) markedly increased early adsorption of rhizobia to alfalfa roots. The activity in RE was linked to a nondialyzable, thermolabile, trypsin-sensitive factor(s), very different from the root-exuded flavonoid compounds also involved in early Rhizobium-legume interactions. The lack of activity in the RE from plants grown in 5 mM NO3- suggested its negative regulation by the nitrogen nutritional status of the plant. Preincubation of R. meliloti with heterologous clover RE did not stimulate adsorption of rhizobial cells to roots. A short pretreatment of RE with homologous (but not heterologous) strains eliminated the stimulatory activity from solution. The stimulation of adsorption of R. meliloti to alfalfa roots was strongly dependent on the growth phase of the rhizobia, being greater at the late exponential stage. Nevertheless, the capacity of R. meliloti L5-30 to eliminate from solution the stimulatory activity in RE appeared to be constitutive in the rhizobia. The low concentration of rhizobial cells used in these experiments was critical to detect the stimulation of adsorption. The early interaction of spontaneously released alfalfa root macromolecular factor(s) and free-living R. meliloti, which shows the specificity and regulatory properties characteristic of infection and nodulation, would be an initial recognition event in the rhizosphere which triggers the process of symbiotic association.     

The Rhizobium leguminosarum nodulation gene nodF encodes a polypeptide similar to acyl-carrier protein and is regulated by nodD plus a factor in pea root exudate

The DNA sequence of ~3.5 kb of the nodulation (nod) region of the Rhizobium leguminosarum symbiotic plasmid pRL1JI was determined. Three open reading frames were identified; genes corresponding to these have been called nodD, nodE and nodF.nodD is adjacent to nodA and is transcribed in the opposite direction. The nodF and nodE genes are downstream of, and transcribed in the same direction as, nodD with 667 nucleotides between nodD and nodF and three nucleotides separating nodF and nodE. The induction of the nodFE operon requires the nodD gene product and a component present in plant root exudate. Regions of DNA sequence preceding nodF are similar to those preceding nodA; these sequences may be involved in the regulation of the expression of nodA and nodF. Analysis of nodD revealed an amino acid sequence similar to the predicted DNA-binding domain of known DNA-binding proteins. A protein comparison of the nodF protein showed it to be similar to the acyl-carrier protein from Escherichia coli and barley, especially around the pantothenate-binding region and on this basis it is thought that this protein may be involved in an acyl transfer reaction.     

A dual-targeted soybean protein is involved in Bradyrhizobium japonicum infection of soybean root hair and cortical cells

The symbiotic interaction between legumes and soil bacteria (e.g., soybean [Glycine max L.] and Bradyrhizobium japonicum]) leads to the development of a new root organ, the nodule, where bacteria differentiate into bacteroids that fix atmospheric nitrogen for assimilation by the plant host. In exchange, the host plant provides a steady carbon supply to the bacteroids. This carbon can be stored within the bacteroids in the form of poly-3-hydroxybutyrate granules. The formation of this symbiosis requires communication between both partners to regulate the balance between nitrogen fixation and carbon utilization. In the present study, we describe the soybean gene GmNMNa that is specifically expressed during the infection of soybean cells by B. japonicum. GmNMNa encodes a protein of unknown function. The GmNMNa protein was localized to the nucleolus and also to the mitochondria. Silencing of GmNMNa expression resulted in reduced nodulation, a reduction in the number of bacteroids per infected cell in the nodule, and a clear reduction in the accumulation of poly-3-hydroxybutyrate in the bacteroids. Our results highlight the role of the soybean GmNMNa gene in regulating symbiotic bacterial infection, potentially through the regulation of the accumulation of carbon reserves.     

Structural determination of bacterial nodulation factors involved in the Rhizobium meliloti-alfalfa symbiosis

Extracellular signals produced by Rhizobium meliloti are able to induce root hair deformations and nodule organogenesis on alfalfa. The production of these signals is controlled by bacterial nod genes. To enable their isolation in significant amounts, an overproducting strain was constructed. These Nod factors were first extracted by butanol from the culture medium and further purified by reverse-phase high performance liquid chromatography, ion-exchange, and Sephadex LH-20 chromatographies. The structure of the major signal, called NodRm-1, was determined by mass spectrometry, nuclear magnetic resonance, 35S labeling, chemical analysis, and enzymatic degradation, and was shown to be a sulfated and acylated tetramer of glucosamine namely, beta-D-GlcpN(2,9-hexadecadie-noyl) - (1----4) - beta - D - Glc p Nac - (1----4) - beta - D - Glc p NAc - (1----4) - D - GlcpNAc-6-SO3H. Another Nod factor (called Ac-NodRm-1) was co-purified and identified as NodRm-1 acetylated on the C-6 of the nonreducing end sugar. NodRm-1 elicits root hair deformation specifically on alfalfa at a concentration less than 10(-10) M but has no effect on vetch (a heterologous host plant).     

Role of plant root exudate and Sym plasmid-localized nodulation genes in the synthesis by Rhizobium leguminosarum of Tsr factor, which causes thick and short roots on common vetch.

In a previous paper it was shown that cocultivation of Rhizobium leguminosarum with the plant Vicia sativa subsp. nigra on solid medium causes a changed mode of growth of the plant roots, resulting in thick and short roots (Tsr). The Sym plasmid present in the bacterium appeared to be essential for causing Tsr (A. A. N. van Brussel, T. Tak, A. Wetselaar, E. Pees, and C. A. Wijffelman, Plant Sci. Lett. 27:317-325, 1982). In the present paper, we show that a role in causing Tsr is general for Sym plasmids of R. leguminosarum and Rhizobium trifolii. Moreover, mutants with transposon insertions in the Sym plasmid-localized nodulation genes nodA, B, C, and D are unable to cause Tsr, in contrast to nodulation mutants localized in other parts of the Sym plasmid. The observation that Tsr could also be brought about in liquid medium enabled us to show that Tsr is caused by a soluble factor. Experiments in which plants and bacteria were grown separately in the sterile supernatant fluids of each other resulted in establishing the following sequence of events. (i) The plant produces a factor, designated as factor A. (ii) Factor A causes the Sym plasmid-harboring bacteria to produce Tsr factor. (iii) Growth of young plants in the presence of Tsr factor results in the Tsr phenotype. Models explaining this example of molecular signalling between bacteria and plants are discussed.     

Host genetic control of symbiosis in soybean (Glycine max L.)

Genes controlling nitrogen-fixing symbioses of legumes with specialized bacteria known as rhizobia are presumably the products of many millions of years of evolution. Different adaptative solutions evolved in response to the challenge of survival in highly divergent complexes of symbionts. Whereas efficiency of nitrogen fixation appears to be controlled by quantitative inheritance, genes controlling nodulation are qualitatively inherited. Genes controlling nodulation include those for non-nodulation, those that restrict certain microsymbionts, and those conditioning hypernodulation, or supernodulation. Some genes are naturally occurring polymorphisms, while others were induced or were the result of spontaneous mutations. The geographic patterns of particular alleles indicate the role of coevolution in determining symbiont specificites and compatibilities. For example, the Rj4 allele occurs with higher frequency (over 50%) among the soybean (G. max) from Southeast Asia. DNA homology studies of strains of Bradyrhizobium that nodulate soybean indicated two groups so distinct as to warrant classification as two species. Strains producing rhizobitoxine-induced chlorosis occur only in Group II, now classified as B. elkanii. Unlike B. japonicum, B. elkanii strains are characterized by (1) the ability to nodulate the rj1 genotype, (2) the formation of nodule-like structures on peanut, (3) a relatively high degree of ex planta nitrogenase activity, (4) distinct extracellular polysaccharide composition, (5) distinct fatty acid composition, (6) distinct antibiotic resistance profiles, and (7) low DNA homology with B. japonicum. Analysis with soybean lines near isogenic for the Rj4 versus rj4 alleles indicated that the Rj4 allele excludes a high proportion of B. elkanii strains and certain strains of B. japonicum such as strain USDA62 and three serogroup 123 strains. These groups, relatively inefficient in nitrogen fixation with soybean, tend to predominate in soybean nodules from many US soils. The Rj4 allele, the most common allelic form in the wild species, has a positive value for the host plants in protecting them from nodulation by rhizobia poorly adapted for symbiosis.     

Outer membrane of Escherichia coli: properties of the F sex factor traT protein which is involved in surface exclusion.

The traT protein (TraTp) of the F sex factor is the product of one of the two genes involved in surface exclusion. Several detergents were examined under different conditions in order to determine their ability to solubilize TraTp from membrane vesicles. These experiments showed that TraTp behaved similar to a number of peptidoglycan-associated outer membrane proteins and that it existed in multimeric aggregates within the membrane. However, unlike other major outer membrane proteins, the amount of TraTp incorporated into the membrane was not affected by lipopolysaccharide-deficient mutants, even when mutants totally lacking the neutral sugars in their lipopolysaccharide backbone were used. TraTp wqs also examined by two-dimensional gel electrophoresis, where it ran as a discrete spot with a very basic isoelectric point. By coupling cyanogen bromide-activated dextran onto whole cells and by labeling whole cells with 125I (via lactoperoxidase), it was shown that TraTp was exposed on the cell surface. TraTp in a membrane environment was also insensitive to proteolytic attack by trypsin.     

Production of root hair deformation factors by Rhizobium meliloti nodulation genes in Escherichia coli: HsnD (NodH) is involved in the plant host-specific modification of the NodABC factor

The role of the hsnD (nodH) gene in the determination of the host-specific nodulation ability of Rhizobium meliloti was studied by expressing the common nodulation genes (nodABC) with or without the hsnD gene in Escherichia coli and testing for biological activity on various leguminous plants. In this way, four categories of plants were established. Upon infection with E. coli carrying the nodABC construct, root hair deformation (Had) was detected on clovers while the hsnD gene was additionally needed for the elicitation of the same response on alfalfa and sweet clover. A weak root hair deformation was seen on siratro by inoculation with E. coli harbouring the nodABC genes and was highly increased when hsnD was also introduced. Cowpea and Desmodium did not respond to any of the E. coli strains constructed. Exudates or cytosolicfractions of the respective E. coli derivatives elicited the same root hair deformation as the intact bacteria. These data indicate that not only the nodABC gene products but also the hsnD product are involved in the synthesis of Had factors. Subclones expressing only the nodA, nodB, or nodC genes or the same genes in pairs (nodAB, nodBC, nodAC) did not provide a compound with activity comparable to the NodABC factor, suggesting that all three genes are required for the production of the Had factor which is active on clover. Coinoculation of alfalfa plants with two strains of E. coli, one carrying the nodABC genes and the other expressing only hsnD, or combining exudates or cytosolic fractions from these strains did not result in root hair deformation on alfalfa. These data indicate that the HsnD protein itself or its product is not an additional alfalfa-specific extracellular signal but more likely is enzymatically involved in the modification of the basic compound determined by the nodABC genes.     

Rhizobium nodulation protein NodA is a host-specific determinant of the transfer of fatty acids in Nod factor biosynthesis

In the biosynthesis of lipochitin oligosaccharides (LCOs) theRhizobium nodulation protein NodA plays an essential role in the transfer of an acyl chain to the chitin oligosaccharide acceptor molecule. The presence ofnodA in thenodABCIJ operon makes genetic studies difficult to interpret. In order to be able to investigate the biological and biochemical functions of NodA, we have constructed a test system in which thenodA, nodB andnodC genes are separately present on different plasmids. Efficient nodulation was only obtained ifnodC was present on a low-copy-number vector. Our results confirm the notion thatnodA ofRhizobium leguminosarum biovarviciae is essential for nodulation onVicia. Surprisingly, replacement ofR. l. bv.viciae nodA by that ofBradyrhizobium sp. ANU289 results in a nodulation-minus phenotype onVicia. Further analysis revealed that theBradyrhizobium sp. ANU289 NodA is active in the biosynthesis of LCOs, but is unable to direct the transfer of theR. l. bv.viciae nodF E-dependent multi-unsaturated fatty acid to the chitin oligosaccharide acceptor. These results lead to the conclusion that the original notion thatnodA is a commonnod gene should be revised.     

High-resolution structure of NodZ fucosyltransferase involved in the biosynthesis of the nodulation factor

The fucosyltransferase NodZ is involved in the biosynthesis of the nodulation factor in nitrogen-fixing symbiotic bacteria. It catalyzes alpha1,6 transfer of l-fucose from GDP-fucose to the reducing residue of the synthesized Nod oligosaccharide. We present the structure of the NodZ protein from Bradyrhizobium expressed in Escherichia coli and crystallized in the presence of phosphate ions in two crystal forms. The enzyme is arranged into two domains of nearly equal size. Although NodZ falls in one broad class (GT-B) with other two-domain glycosyltransferases, the topology of its domains deviates from the canonical Rossmann fold, with particularly high distortions in the N-terminal domain. Mutational data combined with structural and sequence alignments indicate residues of potential importance in GDP-fucose binding or in the catalytic mechanism. They are all clustered in three conserved sequence motifs located in the C-terminal domain.     

Enzyme immunoassay in which a myeloma protein is used for detection of salmonellae

An enzyme immunoassay (EIA) in which an immunoglobulin A monoclonal antibody from a myeloma (MOPC 467) is used was developed to detect the presence of Salmonella organisms. This myeloma protein binds to a flagellar determinant of the organisms but is not directed toward the H antigens. Of 100 strains tested, 94% were detectable with this antibody. The EIA, used with MOPC 467, is quick, sensitive, and specific, showing virtually no cross-reactivity to other enteric organisms. Initial screening of antibody reactivity was performed by Ouchterlony gel diffusion with the supernatants of heat-treated Salmonella cultures. After this, an EIA was performed on the heat extracts with the myeloma protein, which had been directly coupled to alkaline phosphatase. A positive reaction was indicated by the production of a yellow color after the addition of a substrate (p-nitrophenylphosphate), and this was quantitated by determining the absorbance at 405 nm. The EIA proved to be slightly more sensitive than the Ouchterlony analysis. The sensitivity of the EIA is such that as few as 10(6) Salmonella organisms per ml were detected. This concentration was easily obtained after a 24-h preenrichment incubation of the sample. Mixtures of Salmonella strains with a 10 x concentration of Escherichia coli did not prevent detection of the Salmonella strains. This EIA can be successfully used to detect contamination of foods, as it was used to detect the intentional contamination of infant formula in these studies. Indications are that the EIA is sensitive enough to detect Salmonella strains in M broth subcultures taken directly from a preenrichment culture. Testing of samples could thus be completed 36 h after culture initiation, rather than after 96 h, the time currently needed.     

Lotus corniculatus nodulation specificity is changed by the presence of a soybean lectin gene

Plant lectins have been implicated as playing an important role in mediating recognition and specificity in the Rhizobium-legume nitrogen-fixing symbiosis. To test this hypothesis, we introduced the soybean lectin gene Le1 either behind its own promoter or behind the cauliflower mosaic virus 35S promoter into Lotus corniculatus, which is nodulated by R. loti. We found that nodulelike outgrowths developed on transgenic L. corniculatus plant roots in response to Bradyrhizobium japonicum, which nodulates soybean and not Lotus spp. Soybean lectin was properly targeted to L. corniculatus root hairs, and although infection threads formed, they aborted in epidermal or hypodermal cells. Mutation of the lectin sugar binding site abolished infection thread formation and nodulation. Incubation of bradyrhizobia in the nodulation (nod) gene-inducing flavonoid genistein increased the number of nodulelike outgrowths on transgenic L. corniculatus roots. Studies of bacterial mutants, however, suggest that a component of the exopolysaccharide surface of B. japonicum, rather than Nod factor, is required for extension of host range to the transgenic L. corniculatus plants.     

Regulation of nodulation in the soybean-Rhizobium symbiosis : strain and cultivar variability

Double inoculation (15 h apart) of the soybean cultivar Williams with Bradyrhizobium japonicum I-110ARS reveals a rapid regulatory plant response that inhibits nodulation of distal portions of the primary root (M Pierce, WD Bauer 1984 Plant Physiol 73: 286-290). Only living, homologous rhizobia elicit the response. We conducted similar double inoculation experiments to test the hypothesis that this is a universal phenomenon in soybean symbioses. We investigated interactions of the cultivar McCall with the slow-growing strain Bradyrhizobium sp. 3185 (=3G4b16) and strains of the fast-growing soybean symbiont, Rhizobium fredii (USDA191 [Nod⁺ on McCall] and USDA257 [Nod⁻ on McCall]). Nodulation was not detectably inhibited when USDA257 was included in various combinations with an inoculum of USDA191. Strain USDA257 cohabited nodules with strain USDA191 when plants were inoculated sequentially with both strains, but USDA257 did not nodulate McCall when a sterile culture filtrate of USDA191 was added to USDA257 inoculum. There was only a slight inhibition of nodulation of distal portions of the primary root in double inoculation experiments with McCall and strain 3185. Because these results were unexpected, we repeated the experiments with Williams and strain I-110ARS. The response was similar to that observed in the McCall × 3185 interaction. Regulation of nodulation on the primary root thus appears to be variable and depend on strain X cultivar interactions.     

Resistance to nodulation of cv. Afghanistan peas is overcome by nodX, which mediates an O-acetylation of the Rhizobium leguminosarum lipo-oligosaccharide nodulation factor

Only some strains of Rhizobium leguminosarum biovar viciae can efficiently nodulate varieties of peas such as cv. Afghanistan, which carry a recessive allele that blocks efficient nodulation by most western isolates of R. I. viciae. One strain (TOM) which can nodulate cv. Afghanistan peas has a gene (nodX) that is required to overcome the nodulation resistance. Strain TOM makes significantly lower amounts of lipo-oligosaccharide nodulation factors than other strains of R. I. viciae. and this effect appears to be due to lower levels of nod gene induction. These nodulation factors are similar to those from other R. I. viciae. strains in that they consist of an oligomer of four or five β1-4-linked N-acetylglucosamine residues in which the terminal non-reducing glucosamine carries an O-acetyl group and a C_18:4 or C_18:1N-acyl group. However, one of the nodulation factors made by strain TOM differs from the factors made by other strains of R. I. viciae. in that it carries an O-acetyl group on the C-6 of the reducing N-acetylglucosamine residue. This acetylation is NodX-dependent and the pentameric nodulation factor is acetylated on the reducing N-acetylglucosamine residue whereas the tetrameric nodulation factor is not. Although the nodL gene product is also an O-acetyl transferase (it O-acetylates the C-6 of the terminal non-reducing glucosamine), there is very little similarity between the amino acid sequences of these two acetyl transferases.     

nodZ, a unique host-specific nodulation gene, is involved in the fucosylation of the lipooligosaccharide nodulation signal of Bradyrhizobium japonicum.

The nodulation genes of rhizobia are regulated by the nodD gene product in response to host-produced flavonoids and appear to encode enzymes involved in the production of a lipo-chitose signal molecule required for infection and nodule formation. We have identified the nodZ gene of Bradyrhizobium japonicum, whose product is required for the addition of a 2-O-methylfucose residue to the terminal reducing N-acetylglucosamine of the nodulation signal. This substitution is essential for the biological activity of this molecule. Mutations in nodZ result in defective nodulation of siratro. Surprisingly, although nodZ clearly codes for nodulation function, it is not regulated by NodD and, indeed, shows elevated expression in planta. Therefore, nodZ represents a unique nodulation gene that is not under the control of NodD and yet is essential for the synthesis of an active nodulation signal.     

Enzyme immunoassay in which a myeloma protein is used for detection of salmonellae.

An enzyme immunoassay (EIA) in which an immunoglobulin A monoclonal antibody from a myeloma (MOPC 467) is used was developed to detect the presence of Salmonella organisms. This myeloma protein binds to a flagellar determinant of the organisms but is not directed toward the H antigens. Of 100 strains tested, 94% were detectable with this antibody. The EIA, used with MOPC 467, is quick, sensitive, and specific, showing virtually no cross-reactivity to other enteric organisms. Initial screening of antibody reactivity was performed by Ouchterlony gel diffusion with the supernatants of heat-treated Salmonella cultures. After this, an EIA was performed on the heat extracts with the myeloma protein, which had been directly coupled to alkaline phosphatase. A positive reaction was indicated by the production of a yellow color after the addition of a substrate (p-nitrophenylphosphate), and this was quantitated by determining the absorbance at 405 nm. The EIA proved to be slightly more sensitive than the Ouchterlony analysis. The sensitivity of the EIA is such that as few as 10(6) Salmonella organisms per ml were detected. This concentration was easily obtained after a 24-h preenrichment incubation of the sample. Mixtures of Salmonella strains with a 10 x concentration of Escherichia coli did not prevent detection of the Salmonella strains. This EIA can be successfully used to detect contamination of foods, as it was used to detect the intentional contamination of infant formula in these studies. Indications are that the EIA is sensitive enough to detect Salmonella strains in M broth subcultures taken directly from a preenrichment culture. Testing of samples could thus be completed 36 h after culture initiation, rather than after 96 h, the time currently needed.