Alanine Synthesis And Secretion By Rhizobium Leguminosarum

Division of Microbiology, School of AMS, University of Reading, Whiteknights, Reading, RG6 6AJ, UK

On the basis of 15N-labeling studies it had been generally accepted that ammonium is the sole secretion product of N2-fixation by the bacteroid and that the plant is responsible for assimilating it into amino acids. However, 15N-labeling studies indicated that soybean bacteroids only secrete alanine and not ammonia (Waters et al. 1998). We investigated this in pea bacteroids and found that both ammonium and alanine are secreted.

Alanine secretion could account for up to 20% of the combined secretion of alanine and ammonium (Allaway et al. 2000). The in vitro partitioning between them will depend on whether the system is open or closed, as well as the ammonium concentration and bacteroid density. This may be explained by the low affinity of alanine dehydrogenase for ammonium, which was determined as 5.1 mM. Similar to this whole cells excrete alanine with a Km for ammonium of 3.2 mM. Thus the ability to detect alanine synthesis in vitro depends on the accumulation of ammonium in the assay system.

The activity of alanine dehydrogenase is higher in soybean compared to pea bacteroids, potentially favoring alanine synthesis in soybean. However, since the assay conditions can alter the amount of alanine formed by isolated bacteroids we identified and mutated the gene for alanine dehydrogenase (aldA). This confirmed that AldA is the primary route for alanine synthesis in isolated bacteroids. Bacteroids of the aldA mutant fix nitrogen but only secrete ammonium at a significant rate, resulting in lower total nitrogen secretion. Peas inoculated with the aldA mutant are green and healthy, demonstrating that ammonium secretion by bacteroids can provide sufficient nitrogen for plant growth. Plants inoculated with the mutant were reduced in biomass compared to those inoculated with the wild type. The labeling and plant growth studies suggest that alanine synthesis and secretion contributes to the efficiency of N2-fixation and therefore biomass accumulation.

We are currently over-expressing AldA to test the hypothesis that the level of AldA will be crucial to the amount of alanine formed by bacteroids. Full expression of AldA appears to require that both aldR and aldA are present on a plasmid, presumably due to autoregulation of aldR (see below).

Divergently transcribed from aldA is a leucine regulator protein (LRP) homolog, aldR. Insertion of an interposon in aldR prevented transcription of aldA, as measured by gusA fusions and Northern blotting, as well as AldA enzyme activity. AldR is autoregulatory, since aldR insertions also prevent transcription of aldRr.gusA fusions. AldA activity is induced by growth on carboxylic acids, including pyruvate, malate and succinate as well as alanine. We are currently investigating whether alanine is excreted by free-living cultures and how factors such as O2 tension and growth phase may affect this. Alanine does not appear to be secreted by free-living cultures in exponential phase but a burst of alanine synthesis can be induced in stationary phase cultures by the addition of excess L-malate and ammonium.

Alanine synthesis can also be considered as a possible overflow pathway in the same sense as synthesis of polyhydroxybutyrate (PHB) and glycogen. We have therefore been investigating the effects of mutating each of these pathways individually as well as together. This is also being done for both indeterminate and determinate plants (pea and bean respectively) using isogenic parent strains that differ only in the sym plasmid. Glycogen synthase mutants (glgA) show a large increase in plant starch in the nodule but remain able to fix nitrogen. This suggests that glycogen synthesis in pea bacteroids may be a major carbon storage compound able to profoundly affect carbon metabolism in the plant. The construction of double and triple mutants should help us understand how plastic these pathways are. We are particularly interested to know whether blocking one or more pathways causes a diversion of carbon and reductant to other overflow pathways.


Allaway D et al. (2000) Mol. Microbiol. 36, 508-515

Waters JK et al. (1998) Proc. Natl. Acad. Sci. USA 95, 12038-12042


'Botany Dept, University of Western Australia, Crawley WA 6009, Australia

2Dept of Biology, MIT, Cambridge, MA 02139, USA

Recent data have challenged the long held view that the product of nitrogenase activity, ammonia, is transferred from the symbiosome to the host cell cytosol where it is assimilated (reviewed in Day et al. 2001). Waters et al. (1998) have presented evidence for the excretion of fixed N as alanine by bacteroids isolated from soybean (Glycine max [L.] Merr.) and purified by sucrose density gradient fractionation. However, excretion and accumulation of alanine by bacteroids was highest at p02 less than 0.01 while nitrogenase activity was maximal at p02 of 0.06. On the other hand, flow-chamber experiments under conditions where nitrogenase activity was optimized, also with isolated bacteroids from soybean, identified ammonia as the major excreted product of fixation (Li et al. 2001). A significant difference between the two types of experimental approach was the removal of the excreted products of fixation in the flow-through system while they accumulated in the closed system used by Waters et al. (1998) and Allaway et al. (2002) have suggested that this is the explanation for the conflicting results.

30 40 50 time (min)

Figure 1. Fixation of N2 assayed by 15N2 labeling of nodulated roots of cowpea. Data are means ± SE (n = 5) and the line is fitted by non-lmear regression. The atmosphere m each incubation vessel was assayed for isotopic enrichment just prior to collecting the root system and found to be 74.7 ± 0.5 at %xs 15N (n = 30).

The present study reports results of steady state I5N2 labeling of nodulated roots of intact cowpea plants (Vigna unguiculata [L.] Walp. cv Vita 3: Bradyrhizobium strain CB756) under conditions that maintained high rates of nitrogenase activity (250 p.mol N2 h"1 g"1 DW nodule; Figure 1). Root systems were frozen in liquid N2 after 5, 10, 20, 30, 45 or 60 min and nodules removed. The experiment was replicated five times and extracted solutes analyzed for isotopic enrichment by GC-MS of their TBDMS derivatives. Figure 2 indicates rapid labeling of the amide group of glutamine (single-labeled molecules) reaching equilibrium at ca. 50% of the isotopic enrichment of the 15N2 supplied after 5 min. Double-labeled molecules (amide and amino groups) accumulated more slowly, consistent with the slower transfer of label to the amino group of glutamine. In cowpea nodules, as in those of soybean, the amide group of glutamine is used by two amidotransferases of the de novo purine pathway to form IMP that is oxidized to xanthine and ureides (reviewed by Atkins, Smith 2000). The accumulation of double-labeled xanthine molecules (Figure 3) at a rate greater than either single-labeled molecules or those with 3 or 4 15N atoms is consistent with this pathway. Furthermore, the slightly slower labeling kinetics of xanthine would be expected for an intermediate of the pathway that is some 16 enzymic steps removed from the second of the glutamine-dependant steps (FGAR amidotransferase). The labeling of the 4 amino acids shown in Figure 4 is consistent with their acquisition of N from the amido group of glutamine through GOGAT and the activity of aminotransferases.

While this study is unable to take account of the influence that multiple pools of these compounds might have on the labeling kinetics shown there is no evidence that supports the idea that alanine is a precursor of the amide group of glutamine or of the purine ring.



















A m

glutamme-l 15N

1 glutamine-2 15N

/ I

Figure 2. 15N labeling of glutamine in nodules of cowpea during exposure to 15N2. Data are means ± SE (n = 5) and the lines are fitted by non-linear regression.

Figure 3. 15N labeling of xanthine in nodules of cowpea J during exposure to 15N2. Data E are means ± SE (n = 5) and the ^ lines are fitted by non-linear = £ regression. s

Figure 3. 15N labeling of xanthine in nodules of cowpea J during exposure to 15N2. Data E are means ± SE (n = 5) and the ^ lines are fitted by non-linear = £ regression. s

time (min)

glutamate aspartate glycire glutamate aspartate glycire

Figure 4. 15N labeling of amino acids in nodules of cowpea during exposure to 15N2. Data are means ± SE (n = 5) and the lines are fitted by non-linear regression.

time (rrin)

Figure 4. 15N labeling of amino acids in nodules of cowpea during exposure to 15N2. Data are means ± SE (n = 5) and the lines are fitted by non-linear regression.

time (rrin)

Was this article helpful?

0 -1
Growing Soilless

Growing Soilless

This is an easy-to-follow, step-by-step guide to growing organic, healthy vegetable, herbs and house plants without soil. Clearly illustrated with black and white line drawings, the book covers every aspect of home hydroponic gardening.

Get My Free Ebook

Post a comment