|Home | About | Journals | Submit | Contact Us | Français|
Nodule CO2 fixation is of pivotal importance for N2 fixation. The process provides malate for bacteroids and oxaloacetate for nitrogen assimilation. The hypothesis of the present paper was that grain legume nodules would adapt to higher plant N demand and more restricted carbon availability at pod formation through increased nodule CO2 fixation and a more efficient N2 fixation. Growth, N2 fixation, and nodule composition during vegetative growth and at pod formation were studied in pea plants (Pisum sativum L.). In parallel experiments, 15N2 and 13CO2 uptake, as well as nodule hydrogen and CO2 release, was measured. Plants at pod formation showed higher growth rates and N2 fixation per plant when compared with vegetative growth. The specific activity of active nodules was about 25% higher at pod formation. The higher nodule activity was accompanied by higher amino acid concentration in nodules and xylem sap with a higher share of asparagine. Nodule 13CO2 fixation was increased at pod formation, both per plant and per 15N2 fixed unit. However, malate concentration in nodules was only 40% of that during vegetative growth and succinate was no longer detectable. The data indicate that increased N2 fixation at pod formation is connected with strongly increased nodule CO2 fixation. While the sugar concentration in nodules at pod formation was not altered, the concentration of organic acids, namely malate and succinate, was significantly lower. It is concluded that strategies to improve the capability of nodules to fix CO2 and form organic acids might prolong intensive N2 fixation into the later stages of pod formation and pod filling in grain legumes.
Legume nodules fix substantial amounts of CO2 largely through the combined activity of carbonic anhydrase (CA, EC 220.127.116.11) and phosphoenolpyruvate carboxylase (PEPC, EC 18.104.22.168) resulting in carboxylation of phosphoenolpyruvate (PEP) (Vance et al., 1983; King et al., 1986; Rosendahl et al., 1990). Nodule CO2 fixation rates are difficult to measure since any applied labelled CO2 is diluted by simultaneous intensive nodule respiration. Measurements thus tend to underestimate nodule CO2 fixation rates. This tendency is further enhanced through the fact, that a large proportion of any fixed CO2 is rapidly respired. Nevertheless, estimates reveal that nodule CO2 fixation rates per protein unit can reach those of young maize leaves (Miller et al., 1987) and contributes considerably to the overall nodule carbon turnover (Warembourg and Roumet, 1989). In fact, PEPC levels in nodules are in the range of those in leaves of C4 plants, and PEPC is expressed in alfalfa nodules about 10–15-fold greater than in other organs and can comprise up to 2% of the soluble protein in the nodules (Vance and Stade, 1984; Vance et al., 1994). Several nodule-enhanced forms of carbonic anhydrase, PEPC, and malate dehydrogenase (MDH, EC 22.214.171.124), have been identified from legume nodules (de la Pena et al., 1997; Miller et al., 1998; Atkins et al., 2001). The fixed carbon in nodules is channelled into malate formation. Malate is taken up by the symbiosome and drives nitrogen fixation. In addition, a substantial portion of the fixed carbon is used for nitrogen assimilation into aspartate and asparagine. Moreover, a possible involvement of malate in the functioning of a putative nodule oxygen diffusion barrier has been proposed (Vance and Heichel, 1991; Galvez et al., 2000), although the precise mechanism is not yet defined.
Nodule CO2 fixation is apparently tightly bound to nitrogenase activity. In developing nodules PEPC protein occurs alongside nitrogenase protein (Suganuma et al., 1997). A down-regulation of PEPC in nodules impairs nitrogen fixation (Schulze et al., 1998b; Nomura et al., 2006). Taking the whole plant into consideration, nodule CO2 fixation constitutes a carbon-saving mechanism when the fixed carbon is used for N assimilation and N transport to shoots. While any malate channelled into the symbiosome is rapidly respired, the use of oxaloacetate for N assimilation is connected with the carboxylation of PEP. In turn, any use of carbon skeletons from the tricarboxylic acid cycle (TCA cycle) for N assimilation is connected with a preceding decarboxylation of pyruvate. On a root/nodule basis the use of fixed carbon for N transport is connected to a certain energy gain from glycolysis together with CO2 fixation rather than CO2 loss.
Although it is commonly accepted that under most conditions nodules are sufficiently supplied with assimilates, various stress conditions affecting photosynthesis (drought, phosphorus deficiency) might alter the situation (Galvez et al., 2005; Schulze et al., 2006). Moreover, during ontogeny the onset of pod formation profoundly changes source–sink relations within plants, in particular, in grain legumes (Salon et al., 2001). Growing pods not only attract a considerably higher amount of nitrogen than emerging leaves, but significant amounts of carbon as well (Voisin et al., 2003a, b). Thus, nitrogen is in increasing demand from nodules during pod formation and pod filling, which at the same time most probably find themselves in a position of more unreliable assimilate supply. If nodules cannot meet the pods’ N requirements, additional nitrogen is attracted from older leaves, inducing progressive senescence (Schiltz et al., 2005). This, in turn, reduces the overall photosynthetic capacity of the plant. Eventually, these interconnections, at least in part, determine the rapidity of ripening and extent of nitrogen yield in pods (Schulze et al., 1998a). This might be critical, in particular in a grain legume like semi-leafless pea varieties. In such varieties, breeding efforts for agronomical benefits were connected with a considerable reduction in photosynthetic capacity.
The hypothesis of the present paper was that nitrogen fixation during pod formation versus vegetative growth would be supported through increased nodule CO2 fixation supplying organic acids and carbon skeletons for N assimilation, thereby achieving a more efficient nitrogen fixation. A semi-leafless variety of pea plants was chosen for this study. Growth, N assimilation, nodule amino acids (AA), organic acids (OA), and free sugar composition were compared between a 14 d growth interval before flowering (vegetative growth), and one during pod formation. Moreover, in additional experiments root/nodule fixation of 15N2 and 13CO2, as well as evolution of H2 and CO2 was measured at both stages of ontogeny.
The objective of the experiments was the comparison of nitrogen fixation efficiency in pea plants during a growth interval at vegetative growth and at early pod formation to test the hypothesis that increased nodule CO2 fixation and organic acid formation would support intensive nitrogen fixation at early pod formation. For that purpose data on growth, nodule activity (in vitro PEPC and AAT activity, nodule H2 and CO2 evolution), progressing nodule senescence, nodule composition (AA, OA, soluble sugars), nodule CO2 and N2 fixation (measured through 13CO2 and 15N2 uptake), and the composition of AA in the xylem sap at the two stages of ontogeny were obtained in four separate experiments. Experiment one was performed with nodulated pea plants in glass pots with quartz sand supplied with an N-free nutrient solution. Consecutive harvests were made at 10, 17, 24, 28, 35, and 42 d after emergence (DAE). The first pods became visible between 26–28 DAE. Since the substrate was N free, the nitrogen increment between consecutive harvests corresponded to nitrogen fixation. In this way it was possible to calculate plant dry matter increment and nitrogen fixation in 14-d growth intervals during vegetative growth (10–24 DAE) and at pod formation (28–42 DAE). Plants at the vegetative growth stage had formed 9–11 nodes at 10 DAE and 17–20 nodes at 24 DAE. Plants at pod formation had formed 21–24 nodes at 28 DAE and 27–28 nodes at 42 DAE. Seed water content was above 85% at 42 DAE indicating that the plants had not yet reached the state of seed abortion at the end of the growth interval (Dumoulin et al., 1994). Plants at the beginning of the respective growth intervals are shown in Fig. 1. The share of senescent nodules at the individual harvest was determined to relate nitrogen fixation to the amount of active nodules. In addition, at the harvests at 24 DAE and 42 DAE in vitro nodule PEPC and AAT activity was measured. To determine nodule activity and composition at the end of the respective growth intervals, plants grown exactly as in experiment one were transferred to a nutrient solution culture system at day 17 after emergence. The system allowed measurements of nodule H2 and CO2 evolution as well as of 13CO2, 15N2, and O2 uptake. Moreover, nodules could easily be harvested for AA, OA, and soluble sugar analyses, and xylem exudates could be obtained at days 23–24 after emergence (vegetative growth) and at days 41–42 (pod formation). In experiment two, nodule H2 and CO2 evolution along with O2 uptake were measured and nodules were harvested for analyses of AA, OA, and soluble sugar composition. In experiment three, xylem exudates was harvested at the respective time of growth and analysed for AA composition. Experiment four was performed to determine root and nodule 13CO2 and 15N2 uptake. For the experiments, Pisum sativum L. cv. ‘Davina’ plants were used and inoculated with Rhizobium leguminosarum (DSZM 30132).
Pisum sativum L. plants were grown in glass pots (diameter=100 mm, h=200 mm) on C- and N-free fine quartz sand amended with basic fertilization as previously described by Adgo and Schulze (2002). Plants were inoculated with 1 ml of Rhizobium leguminosarum grown on YEM to an approximate cell density of 109 ml−1. One ml of the cell suspension was applied to each plant on the sand at sowing and at days 3 and 6 after emergence. Inoculation led to effective nodulation and N2 fixation while an uninoculated control died due to N starvation. Plants were kept under controlled conditions in climate chamber with a 16/8 h day/night cycle at 24/16 °C and a photon flux density of about 360 μmol m−2 s−1 at stem height during light periods. The plants were watered daily with deionized water to 75% of the maximum water-holding capacity of the sand. The maximum water-holding capacity of the sand was 21% of its weight.
Ten days after emergence (DAE), four replicates were harvested and separated into shoots, roots, active (reddish), and inactive (greenish or brownish) nodules (Baudouin et al., 2004). A second and third harvest of a further four replicates was carried out at 17 DAE and 24 DAE, respectively (plants at 24 DAE corresponded to BBCH 36; Feller et al., 1995b). The 14 d between the first and third harvest are referred to as ‘the growth interval of vegetative growth’. At the third harvest, nodule in vitro PEPC and AAT activity was determined on active nodules of plants from six parallel pots.
A further experiment following the same experimental scheme was made with a growth interval from 28 DAE to 42 DAE (pod formation, plants at 42 DAE corresponded to BBCH 71) with harvest at 28, 35, and 42 DAE. This growth interval is subsequently referred to as ‘the growth interval during pod formation’. In vitro PEPC and AAT activity was determined on active nodules of plants from six parallel pots at 42 DAE.
To measure in vitro nodule PEPC and AAT activity, 100 mg of freshly detached nodules were ground in extraction buffer [100 mM MES-NaOH (pH 6.8), 100 mM sucrose, 2% v/v 2-mercaptoethanol, 15% v/v ethylene glycol, 2 mm PMSF] and centrifuged for 15 min at 10 000 g to obtain the soluble protein fraction. Protein content was measured using Bradford's reagent (Bradford, 1976). PEPC and AAT activity were measured in a coupled or direct assay, respectively, monitoring the disappearance of NADH at A340 and 20 °C using protocols described in Egli et al. (1989). All samples were measured in triplicate.
Plant material was dried at 60 °C to a constant weight. Dry shoot, root or nodule material was weighed and ground to a fine powder. N concentration was determined by means of an elementary analyser (Vario EL, Elementar Analysen GmbH, Hanau, Germany). All samples were measured in duplicate.
In a second experiment, root/nodule gas exchange and nodule AA, OA, and soluble sugar concentration was measured on plants in hydroponic culture (Schulze and Drevon, 2005). The measurements were performed on plants corresponding in age and developmental stage to the end of the growth intervals at vegetative growth (23–24 DAE) and during pod formation (41–42 DAE) in the first experiment. Root/nodule net CO2 release, O2 uptake, H2 evolution in air (apparent nitrogenase activity, ANA) and under argon (total nitrogenase activity, TNA) were measured. Inoculated plants were grown in N-free sand culture as described for experiment one and transferred to a hydroponic system at 17 DAE. The system was designed to allow more detailed root nodule gas exchange measurements. Plant roots were transferred into darkened glass cylinders (diameter=20 mm, h=600 mm) with rubber stoppers at both ends. The stem base of each plant was carefully placed through a hole (diameter=5 mm) in the upper rubber stopper. The cylinder contained 250 ml of the following nutrient solution (mM): KH2PO4 (0.06), K2SO4 (0.7); MgSO4 (0.5); CaCl2 (0.8); and micronutrients (μM): H3BO3 (4.0); Na2MoO4 (0.1); ZnSO4 (1.0); MnCl2 (2.0); Co(NO3)2 (0.2); CuCl2 (1.0), and FeNaEDTA (10). The pH was buffered with MES (0.25 mM) and adjusted to 6.5 in the nutrient solution by applying KOH. Plants were held at their stem bases with a sterilized sponge leaving roots in the nutrient solution. The solution was intensely aerated by a flow of normal air of about 1 vol min−1. The nutrient solution was renewed daily. Solution that had evaporated or passed through the plant was replaced by deionized water 1–3 times per day. After measurements, plants were separated into shoots, roots, and nodules. The plant fractions were dried at 60 °C to a constant weight and dry matter was determined. Roots and nodules of four parallel plants were submerged in liquid nitrogen. Active nodules were harvested and not allowed to melt before extraction for analyses on AA, OA, and soluble sugars.
For the H2 and CO2 evolution measurement, roots and nodules had to be enclosed under air-tight conditions in gas cylinders allowing the application of a regulated amount of air and to measure the composition of the outflowing gas. For that purpose, the hole in the upper rubber stopper was sealed with a plasticine material with a high beeswax content. The beeswax gave the material a soft, pliable consistency that ensured a tight adherence to the rubber stopper, tubing, and plant stem. The material is non-toxic to plants. Before sealing, stiff tubing to act as an inflow and outflow of the sealed root/nodule compartment was laid through the hole in the upper rubber stopper and also sealed with the plasticine material. The inflow tubing reached down to the lower end of the glass cylinder while the outflow was put above any nodules on the lower side of the upper rubber stopper.
The sealed root/nodule compartment was connected to an open flow gas exchange measurement system that allowed a mixture of N2/O2 (80/20, v/v) to be applied. For measurements, the nutrient solution level was lowered to about one-third of the glass cylinder, leaving the lower virtually nodule-free part of the root system in the solution. An airflow of 200 ml min−1 (about 1.2 vols min−1) was applied to the root compartment. A sub-sample (100 ml min−1) of the outflowing gas was taken, dried (ice trap and MgClO4) and passed through an H2 analyser and a CO2 analyser (S121 Hydrogen Analyser, S161 infrared CO2 analyser Quibit Systems, Canada). When a stable H2 and CO2 outflow from the root/nodule compartment was reached, the value was taken as ANA (apparent nitrogenase activity) and root nodule net CO2 release. In order to achieve information about the relative efficiency (electron allocation between N2 and H+), the inflow air composition was subsequently switched to Ar/O2 (80/20, v/v). Argon is inert to nitrogenase and thus the whole electron flow is diverted to H+. Consequently, H2 evolution under argon represents total nitrogenase activity (TNA). The peak value, taken approximately 5 min after switching to Ar/O2, was taken as the TNA value. The electron allocation coefficient (EAC) of nitrogenase activity was calculated as 1–(ANA/TNA). In pre-experiments the validity of the H2-evolution measurements were tested through parallel 15N2-application. O2 uptake was measured on parallel plants grown in glass containers with a volume of 150 ml. The open flow measurement of O2 uptake was done as described in Schulze and Drevon (2005). For measurements, the flow through the root/nodule compartment was lowered to 30 ml min−1. The oxygen concentration in the outflowing gas was measured with an Oxynos 100 (Rosamount, Germany).
For the analysis of free AAs and OAs in nodules, nodules were extracted as previously described (Fischinger et al., 2006). Amino acids were analysed by reverse phase HPLC in the ion suppression mode by precolumn derivatization through o-phthaldialdehyde (Chen et al., 1979). OAs were separated through HPLC and were detected by a photodiode array detector. For sugar analysis water extracts were used. Separation was achieved by reverse HPLC and detected by a refractometer. For analytical details see Keutgen and Pawelzik (2008).
In a third experiment plants were grown precisely as in experiment two. At 24 DAE and 42 DAE xylem sap from these plants was collected for analyses of AA. For xylem sap collection the shoot was cut directly under the cotyledons. To avoid any contamination, closing the phloem and removing the cell bleeding sap was achieved by rinsing the cut surface of the root part for about 15 s with 1 M CaCl2 solution. The root was then placed in a pressure chamber (Model 600 Pressure Chamber Instrument, PMS Instrument Co, Corvallis, Oregon, USA), where it was subjected to a 300 MPa overpressure. The xylem sap was collected for a period of 10 min. During the whole procedure the xylem sap was kept on ice and was then frozen immediately (–20 °C). AAs were analysed as described above.
In a fourth experiment plants were grown precisely as in experiment two. At 22–24 DAE and 40–42 DAE N2 and CO2 fixation of roots and nodules was measured through exposure of the root/nodule system to 15N2 and 13CO2. For 15N2 application, the whole tube (root compartment) was filled with nutrient solution and the nutrient solution was subsequently replaced by a 15N2 (99 atom%15Nexc)/O2 (80/20) mixture. Roots and nodules were exposed to the 15N2/O2 mixture for 30 min (Sulieman and Schulze, 2010). Subsequently, the root/nodule compartment was flushed with an N2/O2 mixture enriched with 2000 μl l−1 13CO2 (99 atom%) for 15 min in an open flow-through system to avoid a significant dilution of 13CO2 by root respiration. The O2 concentration during 15N2/O2 incubation did not decline below 18% (Fischinger et al., 2010).
Immediately after label application plants were fixed in liquid nitrogen and separated into roots, shoots, and nodules. The plant material was dried at 60 °C to a constant weight. The dried shoot and root fractions were weighed and milled using a pebble mill. Nodules were ground with a mortar and pestle.
For 15N and 13C analysis, subsamples of the dried plant material were measured with a combination of a C/N analyser (Porapak PQS) and a mass spectrometer (Finnigan MAT, model 252). The 13CO2 uptake was determined by multiplying the C content of a fraction with the 13C excess of this fraction over the 13C% of an unlabelled reference group:
The reference group consisted of plants grown under the same conditions but without any contact to excess 13CO2 in the atmosphere.
Pea plants during pod formation showed higher dry matter formation and N assimilation than pea plants before flowering in a 14 d growth interval (Table 1). N assimilation was measured as the total N increment of plants during the growth intervals. Since the plants received no combined nitrogen, N assimilation corresponded to N2 fixation. ΔN/Δdm ratios in plants at the end of both growth intervals were not different (Table 1). Total nodule dry matter per plant at the end of the growth intervals was about 180% at pod formation (42 DAE) when compared to the end of the growth interval at vegetative growth (24 DAE) (Fig. 2). However, about one-third of the nodules at 42 DAE showed clear signs of senescence, being either greenish or brownish and soft. Active nodules were selected by their reddish to pink colour indicating intact leghaemoglobin in the infected zone. Measurements (H2 evolution) on detached active nodules revealed a nitrogen fixation activity of 0.49 mg N g−1 fresh weight nodule−1 d−1, while nodules selected by their greenish or white colour as inactive displayed zero H2 evolution. Nodule senescence started between 24 DAE and 28 DAE progressing during pod formation (Fig. 2). The onset of nodule senescence coincided with the development of first pods (between 26–28 DAE). The higher plant productivity during pod formation coincided with about a 25% higher specific activity of nodules calculated on the basis of N-increment in the plants and the mean amount of active nodules during the growth interval (Fig. 3).
The higher specific activity of nodules during pod formation is reflected in a higher total concentration of AA in xylem sap and nodule tissue when compared with plants at the vegetative growth stages (Figs 4, ,5A).5A). In addition, in both nodules and xylem sap the increased total AA concentration was largely a result of more asparagine. Apart from changes in asparagine concentration in xylem sap and nodules, no significant shifts in the concentration of any of the other detected AAs were found.
Total sugar concentration in active nodules showed a tendency to be higher at pod formation, but at statistically non-significant levels (Fig. 5B). In addition to sucrose and galactose, significant concentrations of fructose in active nodules were detected during pod formation. Overall, the detected concentrations in free sugars varied strongly between replicates within both growth intervals.
By contrast, OA concentration was significantly higher in nodules at vegetative growth (Fig. 5C). This was a result of more malate and succinate, both known to be taken up by the symbiosome and to support N2 fixation (Vance and Heichel, 1991; White et al., 2007). In fact, succinate was not detectable in nodules of plants at the end of the growth interval during pod formation. The levels of tartrate in nodules were equal at both growth intervals. Fumarate was only detected in negligible concentrations.
Net CO2 release per unit root/nodule dry matter was much lower at pod formation (Table 2). CO2 release per unit of reduced nitrogen was equal between the growth intervals, while the O2 uptake per unit of reduced nitrogen was about 50% higher at pod formation, which resulted in a significantly lower respiratory coefficient of nodulated roots. Thus lower CO2 release per unit of nodulated root, in addition to higher oxygen uptake per unit of fixed nitrogen and a significantly lower apparent respiratory coefficient, coincides with nodules of higher specific activity and increased N2 fixation per plant.
Nodule phosphoenolpyruvate carboxylase (PEPC, EC 126.96.36.199) and aspartate aminotransferase (AAT, EC 188.8.131.52) activity was about one-third higher during pod formation when compared with nodules from plants at the vegetative growth stage (Fig. 6).
Nodule CO2 fixation was directly measured through 13CO2 application. Nodule CO2 fixation per plant was significantly increased at pod formation (Table 3). While nodule CO2 fixation per plant was increased 30-fold compared to vegetative growth, nodule CO2 fixation per nodule dry matter was about four times higher. Thus the higher CO2 fixation per plant during pod formation was a result of both increased nodule mass and a higher specific CO2 fixation activity. Alongside measurements of CO2 fixation, nodule N2 reduction was determined through 15N2 application. There was a slight non-significant tendency towards more specific N2 fixation during pod formation. However, no distinction between active and inactive nodules was made in this experiment. When nodule CO2 fixation was related to the amount of fixed nitrogen, the calculated value was increased about three times at pod formation when compared with nodules at the vegetative growth stage.
A measurement of nodule activity in terms of higher H2 evolution confirmed a higher N2 fixation at pod formation measured by N accumulation and 15N2 application (Table 4). However, the electron allocation coefficient (EAC) was significantly lower at pod formation, indicating an increased share of electron flow onto H+ versus N2, and thus a decreased relative efficiency of nitrogenase.
Our data clearly demonstrate a more intensive nitrogen fixation in pea plants at pod formation in comparison to vegetative growth. Nitrogen fixation peaks during that period, presumably due to newly developing nitrogen attraction throughout pod formation and pod filling (Peat et al., 1981). Pea pods grow quickly and have a high tissue N concentration (Salon et al., 2001). In our experiment the increasing N requirements of the growing pods are met by a higher N2 fixation per plant. The higher nitrogen fixation rate per plant was, in part, the result of more nodules; however, in addition the active share of nodules had a much higher specific activity. These two factors led to N2 fixation rates per plant that could support the altered growth rate of the plants (Δdm) at pod formation, resulting in an unchanged ΔN/Δdm ratio when compared to the vegetative growth stage. A significant amount of nodules already showed clear signs of senescence at pod formation. With regard to the physiological background of more intensive nitrogen fixation at pod formation, knowledge should be extended on the following.
There are several reports that nitrogen fixation in grain legumes peaks at early pod formation and shows a subsequent, occasionally steep, decline (Peat et al., 1981; Jensen, 1987; Imsande, 1989; Vikman and Vessey, 1993a). This decline is often connected to the briskly progressing senescence of leaves. Under optimal experimental conditions, i.e. sufficient water, nutrients, and light, intensive nitrogen fixation often remains stable into the later stages of pod formation and pod filling (Vikman and Vessey, 1993b). This was the case in our experiments, which lasted about 2 weeks into pod formation although per plant nitrogen fixation showed a brisk decline after that period. During the time of our measurements, plants kept green leaves and almost no leaf senescence was visible at the end of the experimental growth interval. Contrary to this, a significant share of nodules showed clear signs of senescence indicated by a greenish colour. The progressing nodule senescence might be connected to the measured increased oxygen uptake of the nodules and the presumably related oxidative stress (Puppo et al., 2005). Moreover, in senescing nodules of soybean, an increased activity of isocitrate lyase (EC 184.108.40.206) and malate synthase (EC 220.127.116.11) was detected (Fargeix et al., 2003). Increased activity of isocitrate lyase in the larger share of senescent and overall older nodules at pod formation circumvent a decarboxylation process in the TCA cycle and thus might have contributed to the observed low respiratory quotient and lower net carbon release of the root/nodule system.
In our experiment increased nitrogen fixation at pod formation was clearly connected with higher CO2 fixation and improved use of refixed carbon for nitrogen assimilation. This is supported by the measured increased nodule 13CO2 uptake and the larger share of asparagine in nodules and xylem sap at pod formation. These facts are supported by higher in vitro activity of PEPC and AAT, although a direct connection between in vitro and in vivo enzyme activity has to be considered with caution. In vitro activity represents potential activity. These data suggest that, at pod formation, more N assimilation is supported through the carbon skeletons from PEP carboxylation. Figure 7 demonstrates asparagine formation on the basis of carbon skeletons from that source versus drainage of keto-acids from the TCA cycle that constitutes a carbon-saving mechanism for roots and nodules. In the case of PEP carboxylation, one atom of carbon is gained for N transport while organic acids from the TCA cycle are eventually formed combined with a previous loss of carbon from pyruvate. In this way, increased CO2 fixation improves the overall root/nodule carbon balance for N turnover and N transport to shoots. In fact, various data achieved on different grain legumes yielded consistently lower specific root/nodule carbon loss for driving N2 fixation during pod formation and pod filling versus vegetative growth (Schulze et al., 2000; Adgo and Schulze, 2002). The data in this report suggest that nodule CO2 fixation and the use of the resulting carbon skeletons for N transport is a major mechanism for carbon-efficient N2 fixation.
Assimilate shortage could not be detected in active nodules during pod formation. However, it cannot be ruled out that a certain depletion of background reserves like starch or poly-hydroxybutyrate had already occurred and would result in unstable assimilate availability during pod filling. The high variability of our data with respect to nodule sugar concentration might be an indication of this. A higher concentration of fructose and a tendency of higher total sugar concentration at pod formation in conjunction with higher respiration and N assimilation might indicate an increased allocation of assimilates to nodules at that growth stage. Various reports indicate that although assimilate supply does not usually limit nodule activity (Vance and Heichel, 1991; Schulze, 2004), it might be critical during pod formation when growing pods not only induce high activity in nodules but, in addition, attract huge amounts of carbon. For example, the use of hup+ versus hup– Rhizobium strains show effects on nitrogen fixation and growth but not before the onset of pod growth (Bergersen et al., 1995). Provision of additional assimilates through sugar spraying on leaves had no effect on pea plants during vegetative growth while it significantly improved nitrogen fixation at pod formation and pod filling (Schulze et al., 1994). The interdependence between pea plant photosynthetic capacity and assimilate supply of nodules during pod formation and pod filling might be particularly critical in semi-leafless varieties like the one used in our experiments.
Stronger nodule CO2 fixation is also connected with emerging P deficiency in legumes (Schulze et al., 2006). P deficiency impairs photosynthetic activity and assimilate supply to nodules. Consequently, a carbon-saving mechanism like CO2 fixation might be of vital importance for adaptation to low P availability. Christeller et al. (1977) suggested that nodule N assimilation had to be almost completely supported by oxaloacetate from PEPC activity, a fact that would result in an up to 25% proportion of nodule fixed carbon in N transport. Otherwise, a significant drainage of keto-acids out of the TCA cycle pool would impair acetic acid influx into the cycle due to insufficient oxaloacetate availability (Walker, 1962). Overall our data support this hypothesis, in that the ability to convert sugar into organic acids was limited or at least on the edge of limitation at pod formation. In particular, the virtual ‘disappearance’ of succinate in nodules at pod formation suggests that the TCA cycle is significantly drained for N assimilation. This might constitute an additional factor in overstretching nodule activity and adding to emerging nodule senescence at pod formation, as the gap between increasing energy demand and impaired ability for energy provision widens.
In conclusion, the capacity of nodules to fix CO2 is of crucial importance for their efficiency. Nodule CO2 fixation provides a mechanism for saving carbon when the fixed carbon is used for N assimilation and N transport. The ability of nodules to fix CO2 can be influenced through the use of agronomic measures, breeding, and genetic techniques. High CO2 around nodules support nitrogen fixation (Yamakawa et al., 2004; Fischinger et al., 2010). There are reports that straw application positively affects nitrogen fixation in a subsequent crop (Shivashankar and Vlassak, 1978; Evans et al., 1997). This might be the result of higher CO2 concentration in the soil atmosphere. Selection for high PEPC activity in nodules of alfalfa plants improved their growth performance. Various nodule-enhanced forms of key enzymes of the biochemical pathways have been identified (Suganuma et al., 1997; Miller et al., 1998; Fedorova et al., 1999). While down-regulating of PEPC, for example, reduces nitrogen fixation (Nomura et al., 2006), over-expression of MDH increases nodule specific activity (Schulze et al., 2002). Our data suggest that an improvement of nodule capability to channel assimilates into oxaloacetate and malate formation through CO2 fixation might prolong intensive nitrogen fixation in grain legumes into the later stages of ontogeny.
We thank anonymous referees for helpful and constructive comments. We would like to express our gratitude for the outstanding technical assistant of Susanne Koch, Marlies Niebuhr, Reinhard Hilmer, and Ute Ronsöhr from the Department of Crop Science, University of Goettingen. For isotope analysis we want to thank Reinhard Langel from the Centre for Stable Isotope Research and Analysis, University of Goettingen. We would like to thank John Coats for correcting the language. The study was supported by the German Science Foundation (DFG, SCHU 1602/3-1).