INTRODUCTION

Top Abstract Introduction Materials and Methods Results Discussion References

Alders form root nodules in symbiosis with actinomycetes of the genus Frankia that have the ability to fix N₂. Since between 70 and 90% of the total nitrogen assimilated by the host plant can be provided by Frankia in root nodules, the plant is to a large extent independent of soil nitrogen (11, 23, 28). Alders therefore represent successful pioneer plants frequently coming up after flooding, fires, landslides, glacial activity, and volcanic eruptions (8). They grow on soils with a wide range of properties (8, 10). They physically enhance the stability of these soils with their well-developed root system (26) and increase nitrogen mineralization rates in soil, thereby enhancing nitrogen availability and thus improving the quality of impoverished soils. Economically, alders are therefore useful for reforestation and reclamation of nitrogen-depleted, nitrogen-limiting soils. They are also used as nurse trees in mixed plantations with valuable tree species, i.e., by interplanting them with suitable tree crops such as walnut, for production of fuel wood and as a source of timber in monocultures (6, 7, 13, 14).

Mixed plantations of Alnus or Elaeagnus spp. and valuable tree species are a proven silvicultural practice that exploits the ability of actinorhizal plants to increase soil nitrogen contents for subsequent use as nitrogen resource by the tree crops (6, 7). The efficiency of the symbiosis between Frankia and woody plants of the genus Alnus is largely determined by environmental factors such as the soil pH (5, 15), the soil matric potential (9, 41), and the availability of elements such as nitrogen (27, 49) or phosphorus (40, 51). Other factors, however, also help to determine the genotypes of both partners of this symbiosis (19, 37). An improvement in the symbiosis for economic purposes therefore requires the selection of optimal growth sites but also an optimal combination of plants of interest, e.g., forest ecotypes of Alnus glutinosa and superior genotypes of Frankia as inocula (19, 22, 50).

Recent studies have shown that inoculation of Frankia strains is an appropriate strategy to improve the Frankia-Alnus symbiosis resulting in increased plant growth performance and nitrogen availability (36, 47, 48). Through inoculation, Frankia populations can be established in root nodules under conditions that do not favor vesicle formation in nodules formed by the indigenous Frankia population (36). Since nodules are perennial, the positive effect of such inoculations can continue over several years. However, for a long-term effect, the introduced strain not only should compete with the indigenous Frankia populations for nodule formation but also should remain active in the nodules and survive in soil. The introduced Frankia strains must be able to persist in soil in a physiologically active state since only the physiologically active fraction of the Frankia soil population is thought to form root nodules (32).

The physiological status of a specific Frankia population in soil might be triggered by environmental factors such as the presence of vegetation that favors saprophytic growth of this population and increases its competitive abilities with respect to root nodule formation (32). Plant bioassays have demonstrated that members of the genus Frankia survive and remain infective in soils that are devoid of host plants (2, 24, 32, 42-46). This suggests that Frankia strains have the ability to grow in soil. Nutrient resources might be obtained from root exudates since it has been shown that Frankia strains are able to colonize and grow on the root surface of different host and nonhost plants without addition of exogenous carbon sources (39, 43). Alternative carbon resources might be obtained from the decomposition of organic material such as leaf litter. Leaves of Casuarina, for example, have been found to contain compounds that promote growth of Casuarina-infective Frankia strains (54). In addition, compounds detected in seeds of A. rubra are found to enhance nodulation by frankiae (4).

The aim of our study was to determine whether Frankia strains inoculated and incubated in soil amended with leaf litter of Alnus glutinosa but without host plants for several weeks remain infective and competitive for nodulation on A. glutinosa with the indigenous Frankia population. Therefore, after the initial incubation period of soils without plants, soils were planted with seedlings of A. glutinosa. After 4 months of growth, plants were analyzed for growth performance, for specific Frankia populations in root nodules and for N₂ fixation rates. Frankia populations in root nodules were analyzed by in situ hybridization with fluorescent probes targeting specific groups of Frankia strains (52, 53) and their N₂ fixation rates were quantified using the ¹⁵N dilution method (12).

	MATERIALS AND METHODS

Top Abstract Introduction Materials and Methods Results Discussion References

Experimental setup. Surface soil samples (down to a depth of 20 cm) were collected from a sandy loam supporting a natural stand of A. glutinosa (located in Ettiswil, Switzerland) (52). This soil is characterized by high NO₃ concentrations (10 to 20 mM), a low content of organic material (0.02%), and the presence of Frankia subgroups IIIa, IIIb, and IV of the Alnus host infection group. At the natural site, however, nodules were formed by only subgroup IIIa (52). Freshly sampled soil was cleared of larger particles, e.g., roots and stones, and then sieved (mesh size, 5 mm).

1

Analysis of soil parameters. Anion concentrations (NO₃, NO₂, SO₄², PO₄³, and Cl) were determined in pore water of soil samples collected (i) at the beginning of the study, (ii) after 2 weeks of incubation without plants, (iii) at planting time 6 weeks after initiation of the experiment, and (iv) at the end of the plant growth experiment, by ion chromatography (Dionex DX-100 ion chromatograph equipped with an IonPac AS4A-SC column; Dionex, Sunnyvale, Calif.) (21). Pore water was obtained as described by Nickel et al. (36). Micromolar anion concentrations in pore water were correlated to water contents and expressed in micromoles per gram of soil (dry weight).

Analysis of plant parameters. Plant height was measured monthly. At the end of the plant growth experiment, shoots, roots, and nodules of the plants were harvested separately (36). Shoots and roots were dried at 105°C for 24 h for dry-weight determination and stable-isotope analysis. For the stable-isotope analysis, the dried plant material was ground to a fine powder with a steel ball mill (Mixer Mill, Retsch MM2000). A DELTA-S isotope ratio mass spectrometer (Finnigan MAT, Bremen, Germany), which was coupled via an interface to an EA-1110 elemental analyzer (Carlo Erba, Rodano, Italy), was used to determine total nitrogen and carbon concentrations and ¹⁵N/¹⁴N and ¹³C/¹²C ratios. The isotopic values of the samples were expressed in the delta notation relative to the international standard for carbon (PeeDee Belemnite limestone):

(1)

¹⁵N values were expressed as atom% ¹⁵N excess and were calculated from measured  ¹⁵N values:

(2)

The ¹⁵N/¹⁴N ratio of soil-derived nitrogen was obtained from atom% ¹⁵N excess values of a "non-N₂-fixing" reference Alnus plant. This plant showed nitrogen deficiency symptoms (that is, chlorotic leaves) and had only two small nodules (total nodule weight, 10 mg) with Frankia populations that belonged to group IIIa. This plant displayed the highest atom% ¹⁵N excess value in leaves of all plants (3.8 atom% ¹⁵N) evidencing a predominant nitrogen accumulation from soil. In our N₂ fixation model, all fixation abilities were minimum estimates, which were internally consistent in terms of treatment comparisons.

(3)

Analysis of Frankia populations in root nodules. For the analysis of Frankia populations in root nodules, all nodules were harvested. Nodule lobes were counted, and fresh weights of nodules were determined (36). Nodules were split into lobes, and Frankia populations in the lobes were identified by in situ hybridization using Cy-3-labeled oligonucleotide probes targeting the 16S rRNA of members of the domain Bacteria (EUB338) (1) or specific sequences on the 23S rRNA insertion of Frankia strains AgB1.9 (probe B1.9), ArI3 (probe 23ArI3), and Ag45/Mut15 [probe 23Mut(II)] (52), representing Alnus host infection groups I, IIIa, and IV, respectively. Conditions for hybridization, washing, and analysis were as described by Maunuksela et al. (32) and Nickel et al. (36).

Statistical analysis. All data were expressed as means ± standard deviations and assessed by multiple pairwise comparisons with Tukey's honestly significant difference test (SYSTAT) or two-way analyses of variance (ANOVA) (Frankia inoculation × leaf litter amendment). Before the two-way ANOVA, normality and homocedasticity of the data sets were checked, and the data, which were expressed as percentages, e.g., Frankia infection groups in lobes, were arcsinus transformed. The significance level was set at P < 0.05.

	RESULTS

Top Abstract Introduction Materials and Methods Results Discussion References

Soil parameters. During the 6-week-incubation period of soils without plants, NO₃ was the only soil parameter changing significantly. Within 2 weeks, the initially high NO₃ concentration in pore water [13 ± 3 µmol g of soil¹ (dry weight)] had decreased to a concentration below the detection limit at 0.001 mM in soils amended with leaves. After 6 weeks, the NO₃ concentration in these soils had increased again to the initial concentration. In nonamended soils, the NO₃ concentrations remained nearly unchanged, close to the original high value during the whole incubation period. Six weeks after initiation of the experiment when planting began, NO₃ concentrations in all treatments (i.e., leaf litter amended and nonamended, and also inoculated and noninoculated) were comparably high. NO₂ was not detectable (detection limit 0.001 mM) in all treatments during initial incubation without plants and at the beginning of the plant growth experiment.

1

2

1

2

1

3

1

Plant parameters. Leaves of plants growing on noninoculated, nonamended soil were slightly chlorotic in contrast to the dark green leaves of plants from the remaining treatments. Monthly plant height measurements indicated a faster growth of inoculated plants compared to noninoculated plants (Fig. 1). Similarly, plants on soil amended with leaves grew faster compared to plants on soil without leaf amendment. After 4 months of growth, plants on noninoculated, nonamended soil were the smallest plants measuring an average of 26 ± 7 cm high, while alders on inoculated, amended soil were the largest plants measuring 38 ± 7 cm high. Plants on amended soil were taller than plants on inoculated soil (Table 1). No significant differences were found between plants growing on inoculated and noninoculated soils. However, plants on soils that had been amended and incubated with dried leaves in the soil were significantly taller than plants from soils incubated without leaves (Fig. 1; Table 1). A pattern similar to that of plant height measurements was found for plant dry weights. The lowest dry weights were obtained from plants grown on noninoculated, nonamended soil [2.9 ± 1.3 g of plant material (dry weight)] and the highest ones from inoculated, amended soil plants [5.6 ± 1.6 g of plant material (dry weight)]. Plants from soils either amended with dried leaves or inoculated with pure Frankia cultures were 4.3 ± 1.7 g and 4.1 ± 2.5 g (dry weight), respectively. Again, higher values were determined for plants that were grown on soils amended with leaves. The C/N ratios in leaves and roots from the different treatments were determined based on measured C and N concentrations that ranged from 14 ± 1.4 to 16 ± 1.4 to 19 ± 2.3 to 21 ± 3.1, respectively (data not shown). Plant size was directly correlated with ¹³C values in plant biomass, since the ¹³C values of plants growing on noninoculated, nonamended soil were significantly lower with 30.5 ± 0.8 for leaves and 30.1 ± 0.7 for roots. In contrast the values of plants growing on inoculated, amended soil were (29.2 ± 1.1 and 28.7 ± 1.3 for leaves and roots, respectively) (Table 1).

View larger version (22K): [in this window] [in a new window]   FIG. 1.   Averaged plant heights (X ± SD; n = 15 per treatment) of Alnus glutinosa cultivated on noninoculated, nonamended soil (); noninoculated soil amended with 1% A. glutinosa leaf litter (); nonamended soil inoculated with Frankia strains AgB1.9, ArI3, and Ag45/Mut15 each at an estimated density of 107 cells g of soil1 (dry weight) (); and inoculated, amended soil ().

Frankia populations in root nodules. The total weight of nodule lobes per plant was not significantly different in all treatments but the number and size of these lobes varied (Table 2). The number of nodule lobes formed on plants grown on inoculated soil was higher than on plants that had been grown on noninoculated soil. In contrast, the largest nodules were found on noninoculated plants and lobes from plants grown on inoculated soil were significantly smaller. On noninoculated soils, the difference between amended and nonamended soil regarding lobe number and size was not significant (Table 2). However, on inoculated soils a significant difference in the numbers of nodule lobes was found between amended and nonamended soil.

3

3

View larger version (27K): [in this window] [in a new window]   FIG. 2.   Correlation between N2 fixation and relative abundance of frankiae in root nodules of Alnus host infection group IV (top []) and group IIIa (bottom []).

	DISCUSSION

Top Abstract Introduction Materials and Methods Results Discussion References

The experimental set-up of this study consisted of different pretreatments of the soil before planting of A. glutinosa seedlings that included (i) inoculation with pure Frankia cultures and (ii) amendment with alder leaf litter. The addition of leaf litter changed soil properties as demonstrated by the large decrease in NO₃ concentrations over time compared to nonamended soils and their subsequent return to the original values. Amending soil with organic material added carbon and nitrogen resources to the soil system supporting the activities of different functional groups of microorganisms. During the initial mineralization of the organic material, prevailing NO₃ was presumably immobilized in microbial cells. The NO₃ concentration might have increased while alder leaves degraded from activity of nitrifying bacteria in the highly oxic soils. Thus, the addition of organic material not only supplied potential carbon and nitrogen sources for Frankia populations in soil but also changed additional abiotic and biotic components of the soil.

Plant growth enhancement in this study was less pronounced than in our previous study in which seedlings were planted directly after inoculation with the same strains used in the present study and cultivated with similar water availability (36). The studies also differed with respect to the overall growth conditions because the previous study was conducted under natural light and temperature regimes and this study used controlled light and temperature conditions. Nevertheless, this study detected the significant effects of leaf litter amendments on plant growth performance, including generally increasing the height and weight of plants. The positive effect of leaf litter amendments on plant growth performance might be due to alterations in physicochemical characteristics of the soils that were not detected in our chemical analyses of the pore waters. During the initial incubation of soil amended with leaf litter, but without plants, mineralization and nitrification processes might have resulted in better availability of nitrogen for seedlings although the NO₃ concentration at the time of planting was comparably high with about 13 ± 3 µmol g of soil¹ (dry weight) in all treatments.

In a previous study, we had demonstrated that high NO₃ concentrations in pore water of nonamended, noninoculated soil were not sufficient for optimal plant growth during a 4-month-growth period (36). Leaves of plants grown on these soils were chlorotic and displayed a much higher C/N ratio (25 ± 3) than those of plants grown on nonamended, inoculated soils (16 ± 2) (36). In this study, however, C/N ratios in plant leaves of all treatments were comparably low (14 to 16) based on comparable C and N contents. These values did not indicate any nitrogen limitation but leaves of plants grown on noninoculated, nonamended soil in this study appeared slightly less green than those of plants from the remaining treatments. In contrast to amended soils, NO₃ and NO₂ concentrations in pore water of nonamended soils did not change significantly during the incubation without plants. Therefore, additional factors besides the assumed physicochemical parameters generated after leaf litter amendment and incubation must be assumed to affect nitrogen availability and supply for plants.

Inoculation also had a positive effect on plant growth performance but the effect was less pronounced than in treatments with the leaf litter amendment. It is known that inoculation with Frankia strains might improve plant growth performance by enhancing nodule formation on the host plant and by increasing nitrogen availability and supply (17, 36, 47, 48). Plant growth performance in this study correlated to only small differences in nodulation, i.e., the nodule lobe weight showed no significant differences between treatments and the number of nodule lobes obtained was only higher for plants on inoculated, amended soil (Table 2). Total lobe weight of and lobe numbers on plants from nonamended treatments were comparable with those obtained in the previous study (36) except that the total lobe weight of nodules on plants grown on the noninoculated soil was much larger than in the previous study [160 ± 76 mg versus 38 ± 43 mg (fresh weight)] (36). Despite only small differences in nodulation, percentages of nitrogen in plants originating from N₂ fixation increased from 33% ± 19% when plants were grown on nonamended, noninoculated soil to 78% ± 24% when plants were grown on amended, inoculated soil. These results indicate differences in structure and N₂-fixing activity of Frankia populations in root nodules of plants grown on soils with leaf litter amendments and with inoculation.

Frankia populations in root nodules on plants grown on noninoculated, nonamended soils represented the population generally found in nodules harvested from plants at the field site (53). This population accounted for only a part of the indigenous Frankia population present in this soil that was known to harbor at least three subgroups, IIIa, IIIb, and IV, of the Alnus host infection group (53). In contrast to our previous study in which most of the nodules were found to contain no frankiae (36), nearly all nodule lobes analyzed in this study harbored vesicle-forming frankiae of subgroup IIIa. This could explain the much lower C/N values and greener leaves of plants in our recent study compared to the previous study but may not explain why the percentage of nitrogen in plant leaves originating from N₂ fixation was low with 33% ± 19% of the total nitrogen in the plant.

Leaves and roots from plants grown on amended only, inoculated only, or amended and inoculated soil displayed higher amounts of nitrogen originating from N₂ fixation than those from plants grown on nonamended, noninoculated soil. They were also significantly ¹³C-enriched, displaying more positive ¹³C values than the plant material grown on nonamended, noninoculated soil. ¹³C values can provide long-term information on the influence of the nitrogen source on gas exchange characteristics and water balance of the plant since ¹³C-discrimination by plants was found to be negatively correlated with their efficiency of water use (3, 38). The lower ¹³C-discrimination rate of alder plants grown on amended only, inoculated only, or amended and inoculated soil, therefore indicated improved efficiency of water use by plants with high amounts of nitrogen originating from N₂-fixation. This finding was in contrast to the assumption that due to the high respiratory requirements for driving N₂-fixing, the efficiency of water use of nitrogen N₂-fixing plants such as legumes or Casuarina is lower than that of plants utilizing mineral nitrogen from soil (25, 31). In these studies, a lower ¹³C-enrichment in N₂-fixing plants than in plants grown on NO₃ and NH₄⁺ indicated a lower efficiency of water use by N₂-fixing plants.

The nodulation capacity of a soil was suggested to be controlled largely by the physiological status of the inhabiting Frankia populations, as indicated by infectivity (34, 35). In our study, leaf litter amendment and subsequent incubation for 6 weeks without plants resulted in a large shift of Frankia populations in root nodules from subgroup IIIa to subgroup IV and a concomitant increase in N₂ fixation accounting for 58% ± 25% of the total nitrogen in the plant. Since leaves were harvested directly from the tree and subjected to high temperature treatment at 120°C for 3 days before milling and amendment to soils, any potential introduction of additional Frankia populations into the soils can be excluded. The large shift in nodule-forming Frankia populations after leaf litter amendment and incubation for 6 weeks without plants must therefore be due to changes in the activity of the indigenous Frankia populations in soil. Leaf litter amendment and the incubation conditions applied clearly favored growth of a Frankia population of group IV of the Alnus host infection group over that of frankiae of group IIIa under saprophytic conditions without plants. This was evident for both indigenous as well as introduced populations and correlated with an increase in specific N₂-fixing activity since the amount of nitrogen in plants originating from N₂ fixation increased significantly with higher percentages of nodules harboring frankiae of group IV (Fig. 2).

In nodules of plants grown at field sites, Frankia populations identified by in situ hybridization as belonging to group IIIa usually were of the spore (+) type while those of group IV did not form spores (52). Although this correlation might be accidental since only a limited number of sites and nodules were analyzed, they compare favorably with earlier studies in which spore (+) and spore () type Frankia populations were found to coexist on the same root system (30). In artificial medium, the Frankia population of the spore () type exhibited larger specific N₂-fixing activity than the population of the spore (+) type. In contrast to our results, however, the introduction of a Frankia population of the spore () type into soil harboring an indigenous population of the spore (+) type did not result in the development of significant numbers of nodules of the spore () type nor in an increase of the N₂ fixation rate. Plant growth of A. glutinosa was much more enhanced after inoculation with a Frankia strain obtained from nodules of the spore () type than after inoculation with homogenates of nodules of the spore (+) type or a mixture of both (29).

In summary, our study has shown that introduced Frankia strains incubated in leaf-litter-amended and nonamended soil for several weeks in the absence of plants remained infective and competitive for nodulation with the indigenous Frankia populations on the host plant Alnus glutinosa. Inoculation into and incubation in soil without host plants generally supported subsequent plant growth performance and increased the percentage of nitrogen acquired by the host plants through N₂-fixation. Further studies, however, need to address long-term effects of such an inoculation on nodulation activity. The potential activation of indigenous Frankia populations through increasing the availability of nutrients rather than through inoculation with pure cultures warrants investigation as well.