Can plants absorb and utilize phosphate buffer extractable soil organic nitrogen without its prior mineralization?

Abstract

Several researchers have hypothesized that organic nitrogen extracted using neutral phosphate buffer, termed phosphate buffer extractable organic nitrogen (PEON), is absorbed and utilized by some plant species as a nitrogen source without prior mineralization. Earlier reports have suggested that leaf lettuce (Lactuca sativa L.) depends mainly on inorganic nitrogen, whereas qing-geng-cai (Brassica chinensis L.) can absorb PEON. We tested leaf lettuce and qing-geng-cai growth under aseptic conditions using three nitrogen sources under different light conditions to investigate this hypothesis. Nitrate, amino acids or purified PEON were supplied as the sole nitrogen source. Five different levels of light intensity were used to detect the contribution of carbon to plant growth supplied from PEON. Leaf lettuce growth was promoted only by nitrate, whereas growth of qing-geng-cai was promoted by nitrate and amino acids. The positive effect of amino acids on qing-geng-cai was pronounced at higher light intensity. However, PEON imparted no significant effect on either of the plants under any of the light conditions. Our results do not support the hypothesis that PEON is used as a nitrogen source by plants.

Key words:

• assimilation
• inorganic carbon
• nitrogen compound
• inorganic nitrogen
• size exclusion

INTRODUCTION

Soil organic nitrogen (N) has recently been widely recognized as a N source that can be absorbed directly and used by various plants (Schimel and Bennett 2004). This recognition has spurred a re-evaluation of the traditional view of N dynamics in terrestrial ecosystems, which is based on the assumption that only inorganic N is available to plants (Lipson and Näsholm 2001; Schimel and Bennett 2004). Among organic N types, low molecular weight constituents, mainly amino acids, have been studied intensively (Lipson and Näsholm 2001). The ability of plants to take up intact amino acids as a N source has been demonstrated in laboratory and field experiments.

Unlike low molecular weight organic N, high molecular organic N has attracted less attention as a N source because of a general understanding that plant roots are incapable of directly absorbing and using high molecular weight organic matter (Jones et al. 2005). However, the results of some studies have suggested that the soil organic N of high molecular weight matter might also be taken up directly by plant roots (Koga et al. 2001; Matsumoto et al. 1999, 2000c; Okamoto and Okada 2004; Yamagata et al. 2001). These studies have specifically addressed soil organic N extracted by neutral phosphate buffer, which has been termed phosphate buffer extractable organic nitrogen (PEON) by Matsumoto et al. (2000a). In fact, PEON has been described as “a protein-like N compound with a uniform molecular weight of 8,000–9,000” (Matsumoto and Ae 2004). It is considered to originate from the cell walls of bacteria and can be extracted irrespective of soil type (Matsumoto et al. 2000b).

The hypothesis of direct uptake of PEON was derived originally from the fact that the N uptake responses of plants to organic fertilization differed among species. Previous pot experiments have shown repeatedly that certain species assimilate more N in an organically fertilized treatment (low inorganic N and high PEON concentrations) than in an inorganically fertilized treatment (high inorganic N and low PEON concentrations) (Matsumoto et al. 1999, 2000c; Okamoto and Okada 2004; Yamagata et al. 2001). The difference in the ability of plants to uptake N under low inorganic N conditions has been found to coincide with the difference in the rate at which PEON decreased from the level in the culture solution over a 3-h culturing (Okamoto and Okada 2004). A considerable amount of PEON was detected in plant roots after 6-h culturing with a solution containing biotin-labeled PEON (Koga et al. 2001). The xylem sap of qing-geng-cai was analyzed using high-performance liquid chromatography with a size exclusion column. A peak similar to that of PEON was found in plants grown in a soil culture, but not from those plants grown in an inorganic solution culture (Matsumoto et al. 2000c). The amount of N uptake by qing-geng-cai and carrot increased when PEON was supplied aseptically (Matsumoto et al. 2000c).

If intact PEON were also an organic N source to be absorbed and used by plants, as hypothesized in the studies described above, the paradigm change in the terrestrial N cycle could be much more substantial than that of amino acids; the quantity of PEON in the soil is much higher than that of free amino acids (Lipson and Näsholm et al. 2001; Matsumoto et al. 2000c). However, little evidence directly supports the contribution of PEON to plant growth; moreover, nothing is known about how efficiently plants can use PEON as a source of N.

Another issue exists concerning PEON utilization by plants. Both N and carbon (C) might contribute to plant growth when plants absorb organic matter. Experiments investigating the plant uptake of dual-labeled (¹³C and ¹⁵N) amino acids have shown that plants assimilate C as well as N (Inagaki and Kohzu 2005; Näsholm et al. 2000). Plant dry weight increased by up to 30% when plant roots were supplied with dissolved inorganic C (Bialczyk et al. 1996; Vapaavuori and Pelkonen 1985). Root-absorbed inorganic C was transferred to shoots through the xylem sap in the form of organic compounds and constituted up to 10% of the CO₂-assimilating photosynthetic activity of plants (Cramer and Richards 1999). PEON also contains carbon and its C:N ratio is approximately 8 to 15 (Aoyama 2006; Matsumoto and Ae 2004). If PEON were absorbed directly, an appreciable amount of C would also be incorporated through the roots; can it also be a C source for plants?

We tested the growth of two vegetable species under aseptic conditions with three N sources to investigate whether plants use PEON as a source of N and C. We used leaf lettuce (Lactuca sativa L.) and qing-geng-cai (Brassica chinensis L.; “chingensai” in Japanese) because the results of previous experiments suggested that leaf lettuce depends mainly on inorganic N, whereas qing-geng-cai is capable of absorbing PEON (Matsumoto et al. 1999, 2000c). The experiments were conducted under five different levels of light intensity to detect, indirectly, the contribution of C from the organic supplements to plant growth.

MATERIALS AND METHODS

Phosphate buffer extractable organic nitrogen purification

A soil sample was collected from a fallow field at the National Agricultural Research Center for Tohoku Region (Andosol, clay loam, pH 5.7, total C of 5.14%, total N of 0.37%). The air-dried soil (2.4 kg) was shaken for 1 h with 6 L phosphate buffer (1/15 mol L⁻¹, pH 7.0). The suspension was centrifuged at 12,250 g for 10 min. The supernatant was passed through a glass-fiber filter (GA200; Advantec Toyo Kaisha; Tokyo, Japan) and concentrated approximately 10-fold using a rotary evaporator (N-1000; Tokyo RikaKikai; Tokyo, Japan). The concentrate was filtered through a cellulose acetate membrane filter (0.2 µm; Advantec Toyo Kaisha). The filtered sample was mounted (10 mL per loading) on a preparative size exclusion chromatography (PSEC) column (50 cm with 5 cm diameter, packed with Toyopearl HW-55F; Tosoh Corporation; Tokyo, Japan). Then PSEC was conducted with a mobile phase of 50 mmol L⁻¹ phosphate buffer at a rate of 1.5 mL min⁻¹. Using a fraction collector (DC1500; Tokyo RikaKikai), 12 fractions per cycle were collected with a fraction size of 18 mL. These fractions differed in their apparent molecular weight and hydrophobicity (Miyazawa and Murayama 2007). Therefore, the fractions that had apparent molecular weights ranging from 6,000 to 11,000 Da based on the protein standard were retained for further purification. The PSEC procedure was repeated until a fraction of approximately 650 mL was obtained. The fraction was concentrated approximately threefold using the rotary evaporator. Approximately 100 mL of the concentrate was dialyzed using a cellulose membrane (3,500 MW CO, Cellu Sep; Membrane Filtration Products; Seguine, TX, USA) in a 5 L container with distilled water. The water was changed daily and dialysis was stopped when no inorganic N or phosphorus were detected in the dialysate water. The final concentration of PEON-N in the retentate was measured using an alkaline persulfate oxidation technique (Cabrera and Beare 1993). A persulfate reagent mixture for oxidation was added to 50-fold diluted samples and autoclaved at 120°C for 30 min. Then the nitrate concentration was measured using Cataldo's method (Cataldo et al. 1975). The residual phosphorus concentration was measured using the molybdenum blue method. The residual concentration of inorganic N was determined using an auto-analyzer (AAII, BL TEC K.K.; Tokyo, Japan). The dialysis procedure was conducted twice; consequently, two retentates that contained PEON at slightly different concentrations were obtained. The concentrations of PEON-N were 37.2 and 35.5 mg L⁻¹, and those of phosphorous were 1.3 and 5.4 mg L⁻¹. No inorganic N was detected. The former PEON solution was used for the light intensity treatments L1, L2 and L3, and the latter for L4 and L5. No degradation of PEON was observed during the purification procedures (Miyazawa and Murayama 2007).

Aseptic culture experiment

Test tubes (25 mm × 120 mm) containing 3 g of vermiculite were sealed (MilliWrap; Millipore Corporation; Billeriea, MA, USA) and autoclaved at 120°C for 20 min. Autoclaved Hoagland solution without N was dispensed 4.5 mL per test tube. For N treatment, 4.5 mL of the following solutions were added to the test tubes: (1) control (distilled water), (2) nitrate, (3) amino acids (4) PEON. The N concentrations of the nitrate solution and the amino acid solution were adjusted to be equal to that of the PEON solution. The amino acid solution consisted of 15 amino acids based on the amino acid composition of PEON reported by Higuchi (1982). The molecular weights of the amino acids ranged from 75 to 181. Because the dialyzed PEON solution still contained some phosphorus, the amount of phosphorus in the treatments was adjusted by adding phosphate buffer to the control, nitrate and amino acid solutions. For sterilization, the control and nitrate solutions were autoclaved, and the amino acid and PEON solutions were filtered through MILLEX-GS (0.22 µm; Millipore Corporation). Leaf lettuce (Lactuca sativa L. cv. Green) and qing-geng-cai (Brassica chinensis L. cv. Choyo) seeds were surface sterilized in 0.75% hypochlorous acid solution with a drop of TWEEN20 for 15 min and rinsed with sterilized water. One seed was sown in each test tube with four replicates per treatment. Subsets of these seeds were germinated on MS agar media to ensure the validity of sterilization. The experiment was conducted in growth chambers with five different light intensities: L1 (0 µmol m⁻² s⁻¹), L2 (16 µmol m⁻² s⁻¹), L3 (160 µmol m⁻² s⁻¹), L4 (250 µmol m⁻² s⁻¹) and L5 (500 µmol m⁻² s⁻¹). The growth chambers used were the NK system (LPH-1000MP; Nippon Medical and Chemical Instruments; Osaka, Japan) for L1, L2 and L3, and Koitotron (KG-106SLD/S; Koito Industries; Yokohama, Japan) for L4 and L5 with growth conditions of 25/15°C (day/night) with a 12-h photoperiod. The growth period was 14 days except for L1 (L1 was terminated at 10 days, which was before the plants reached the top of the test tube). The fresh biomass and height of each plant were measured. The root biomass was also measured in L4 and L5. Plant tissues were combined for each treatment and dried at 70°C for 24 h. The dry weight was measured per treatment, rather than per sample, to avoid artifacts caused by measuring ultra-light samples.

Statistical procedures

Statistical analyses were conducted using SAS 9.1 (SAS Institute, Cary, NC, USA). A one-way anova was conducted for the N source effect at each light intensity level, followed by one-tailed Dunnett's procedures to test the significance of differences between the control and each N treatment, or Tukey's procedures to test all pairwise comparisons (maintaining the α level at 0.05). The dry weight of the plants was not analyzed statistically because the values were obtained by dividing the weight of a composite sample by the number of replications.

RESULTS

Total fresh biomass was greater in the nitrate treatment compared to that of the control, especially at higher light intensities, in both species (Fig. 1a,b). Qing-geng-cai responded to the amino acid treatment as well. The biomass of both species under PEON treatment did not differ from the control at any light intensity. Total dry biomass also showed a similar pattern (Fig. 1c,d). The height of the plants generally decreased as the light intensity increased (Fig. 2a,b). The effect of the N source resembled that of the biomass: leaf lettuce responded to nitrate only; qing-geng-cai responded to both nitrate and amino acids. The positive effect of amino acids on qing-geng-cai was pronounced at higher light intensities (Fig. 2b). The above-ground and root biomass were further investigated in L4 and L5 (Fig. 3a,b). Both above-ground and root biomass showed a pattern similar to that of total biomass. The difference between the nitrate and amino acid treatments in the above-ground biomass of qing-geng-cai was not statistically significant in L5 (Fig. 3b).

DISCUSSION

Purified PEON showed no effect on the growth of leaf lettuce and qing-geng-cai under the aseptic conditions used in our experiment, which suggests that these plants did not use PEON. This result is contradictory to the results reported by Matsumoto et al. (2000c); when PEON solution was added to N-free MS liquid medium qing-geng-cai grew better than the control. However, these researchers used N-free MS medium as the control, which should have had the highest osmotic pressure among all the treatments. We used N-free Hoagland solution (half-strength because treatment solutions of the same volume were added) for all treatments, including the control. An appropriate salt concentration may be necessary when evaluating plant growth to avoid the

Figure 1  Total fresh weight (FW) and dry weight (DW) of leaf lettuce (a,c) and qing-geng-cai (b,d) under four nitrogen treatments and at five light intensities (L1–L5). The total FW data represent the mean ± standard deviation (n = 4), and the total DW represent the mean values. An asterisk indicates that the value differs significantly from the control at each light intensity treatment (Dunnett's procedure). PEON, phosphate buffer extractable organic nitrogen.

Figure 2  Height of (a) leaf lettuce and (b) qing-geng-cai under four nitrogen treatments and at five light intensities (L1–L5). The data represent the mean ± standard deviation (n = 4). An asterisk indicates a significant difference from the control at each light intensity treatment (Dunnett's procedure).

effect of salinity stress, which might have been induced by water evaporation during the experiment. Another difference between the two studies is the PEON purification procedure. We discarded PEON fractions that were out of the range of 6,000–11,000 Da, whereas Matsumoto et al. (2000c) did not further separate PEON fractions. However, it is unlikely that the discarded fraction in our experiment was the actual N supply for plants. The fractions we incorporated were the major fractions, representing 85% of total PEON. If the remaining 15%, at most, had been the N source utilizable by plants, the amount of N supplied would have been insufficient to account for the additional plant N uptake assumed to be provided by PEON in Matsumoto et al. (2000c).

Assuming that PEON was not responsible for promoting plant N acquisition unless it was mineralized, can any other hypothesis explain the phenomenon of the different responses of plant species to organic fertilization shown in this series of pot experiments? In the previous pot experiments, carrot, qing-geng-cai and spinach assimilated more N when they were supplied with organic materials (a mixture of rice bran and rice straw, or rapeseed cake) than when supplied with

Figure 3  Above-ground and root fresh weights (FW) of (a) leaf lettuce and (b) qing-geng-cai under four nitrogen treatments and at two light intensities (L4–L5). The data represent the mean ± standard deviation (n = 4). Nitrogen treatments with different letters at each light intensity treatment indicate a significant difference (Tukey's procedure). The graph legend is the same as that used in Fig. 2.

ammonium sulfate. In contrast, pimento and leaf lettuce assimilated more N with ammonium sulfate (Matsumoto et al. 1999, 2000c). Sorghum and rice also responded better when rice bran or a mixture of rice bran and straw was supplied, whereas maize and pearl millet grew better when ammonium nitrate was supplied (Okamoto and Okada 2004). The inorganic N content was much higher in the pots supplied with ammonium sulfate or nitrate, whereas PEON content was higher in the pots supplied with organic materials. The well-known effect of organic materials on soil physical properties was considered negligible in above mentioned pot experiments because the soil culture consisted mostly of vermiculate (80%). What then could have caused the better growth of some species under lower inorganic N conditions?

One reason that might partly explain this phenomenon is the difference in the use of amino acids by these plants. In our experiment, leaf lettuce and qing-geng-cai differed in their responses to amino acids. Qing-geng-cai might have an advantage compared to leaf lettuce when amino acid N is supplied through organic materials. Amino acid utilization by rice as a N source is well known (Mori and Nishizawa 1979). Another possibility is the difference in the N mineralization rate. Both PEON and dissolved organic N are considered to be immobilized and stabilized by soil minerals (Ito and Ae 2000; Kalbitz et al. 2005). Some plant species are known to be able to release H⁺and OH⁻ and/or excrete organic anions, which can drastically change the chemical conditions and nutrient availability in the rhizosphere (Bertrand et al. 1999; Jones 1998). The difference in the ability of desorption of organic N might have affected N mineralization, resulting in different responses to organic materials among species. However, these suppositions still do not explain why some plants grew better under organically fertilized treatments even if they had much less inorganic N available than their inorganically fertilized counterparts.

Another speculation worth exploring is the contribution of C supplied by organic materials. In fact, C supply through roots can contribute to C assimilation of plants, especially when plants are exposed to high temperature, high irradiation and salinity stress, which endanger low stomatal conductance (Cramer and Richards 1999). In our experiment, the above-ground biomass of qing-geng-cai under the amino acid treatment showed a tendency to increase as the light intensity increased, and the biomass approached that obtained under the nitrate treatment. Although our light intensity level was not extremely high, even at the highest level (500 µmol m⁻² s⁻¹), the amino acids might have contributed partially to plant C acquisition when C demand was high. If the organic materials supplied had significantly increased the C concentration in an artificial soil condition, it may have overridden the effect of the inorganic N availability. Dissolved inorganic C has also been reported to stimulate N uptake (Bialczyk 1994; Cramer and Lips 1995).

Our results did not support the hypothesis that PEON is used directly by plants as a N source. However, this is not to deny the possibility of PEON uptake by plant roots. Macromolecules have long been known to be absorbed directly by roots (McLaren et al. 1960; Nishizawa and Mori 1977). Thus, PEON might have been taken up in our experiment, but did not affect plant growth. Furthermore, our results with young seedlings under aseptic condition provide only a limited indication of PEON utilization by plants in situ. As plants develop and encounter stress events, plant roots might respond by activating strategies for nutrient acquisition, which our experiment might not have been able to detect. For example, invagination of cortex cells of plant roots, which incorporated polysaccharides into root cells, was stimulated by desiccation stress (Nishizawa and Mori 1984, 1986).

Recent studies have indicated that PEON is not a homogeneous compound, but rather a humic substance (Aoyama 2006; Miyazawa and Murayama 2007). Numerous technical difficulties arise in demonstrating organic N utilization by plants in situ, even in studies using simple amino acids (Jones et al. 2005). Soil organic matter, such as PEON, whose composition is yet to be identified, requires further studies before any conclusion can be reached as to its actual contribution to plant nutrient acquisition.

We would like to thank Dr Noriharu Ae, Dr Nobuhisa Koga, Dr Miwa Okamoto and Dr Kensuke Okada for kindly providing advice on the experimental procedures, and Toshiyuki Kimura and Masahiro Suzuki for valuable technical advice on the PEON purification procedures.