This article appeared in a journal published by Elsevier. The attached copy is furnished to the author for internal non-commercial research and education use, including for instruction at the authors institution and sharing with colleagues. Other uses, including reproduction and distribution, or selling or licensing copies, or posting to personal, institutional or third party websites are prohibited. In most cases authors are permitted to post their version of the article (e.g. in Word or Tex form) to their personal website or institutional repository. Authors requiring further information regarding Elsevier’s archiving and manuscript policies are encouraged to visit: http://www.elsevier.com/copyright
Author's personal copy
Soil Biology & Biochemistry 42 (2010) 353e359
Contents lists available at ScienceDirect
Soil Biology & Biochemistry journal homepage: www.elsevier.com/locate/soilbio
Temporal variation in pools of amino acids, inorganic and microbial N in a temperate grassland soil Charles R. Warren*, Maria T. Taranto School of Biological Sciences, Heydon-Laurence Building A08, The University of Sydney, NSW 2006, Australia
a r t i c l e i n f o
a b s t r a c t
Article history: Received 21 July 2009 Received in revised form 2 October 2009 Accepted 10 November 2009 Available online 26 November 2009
Plants can take up intact amino acids, even in competition with soil microbes, yet we lack detailed information on which amino acids dominate the soil and whether amino acid composition varies seasonally. This study tested the hypotheses that 1) the pool of amino acid N is generally larger than inorganic N; 2) temporal changes in the concentration of amino acid N is related to changes in the size of the microbial N pool; and 3) amino acid N is dominated by simple, neutral amino acids during warm months, whereas during cold months the amino acid N is dominated by more complex aromatic and basic amino acids. Approximately every month for two years we collected soil from a temperate, subalpine grassland in the Snowy Mountains of Australia. We quantified exchangeable pools of amino acids, nitrate and ammonium in 1 M KCl extracts. Microbial N was quantified by chloroform fumigation. Averaged across the 21 monthly samples, nitrate was 13% of the quantified pool of soluble non-protein N, ammonium was 34% and amino acid N was 53%. These data are consistent with our hypothesis that the pool of amino acid N is larger than inorganic N. There was substantial variation between months in concentrations of amino acids and inorganic N, but no clear temporal pattern. Microbial N did not vary between months, and thus changes in amino acid N were unrelated to microbial N. Principal components analysis indicated multivariate groupings of the different pools of N that were broadly indicative of function and/or biosynthetic relationships. Thus PCA identified a grouping of aromatic amino acids (Phe and Try) with amino acids derived from oxaloacetate (Asp, Ala, Val, Leu, Ile), and a second group comprising microbial N, nitrate and glycine. The pool of exchangeable amino acid N was dominated by Arg (26% of amino N) Val (20%) Gln (18%), Try (8%) and Asn (8%). Contrary to our hypothesis, the composition of the amino acid pool did not vary in a consistent way between months, and there was no evidence simple amino acids were relatively more abundant in warm months and complex amino acids in cool months. Ó 2009 Elsevier Ltd. All rights reserved.
Keywords: Amino acid Nitrate Ammonium Organic N Sub-alpine Temperate
1. Introduction The past decade has seen a resurgence of interest in the role of dissolved organic N (DON) in ecosystem N cycling (Schimel and Bennett, 2004). DON comprises everything from recalcitrant high molecular weight compounds such as proteins, DNA and proteinepolyphenol complexes, through to low molecular weight compounds such as amino acids. Amino acids are particularly important because a range of studies have shown that vascular plants can directly take up significant quantities of amino acids, and bypass the supposed bottleneck of N mineralization (Chapin et al., 1993; Jones and Darrah, 1993; Warren, 2006). It has even been
* Corresponding author. Tel.: þ61 (0) 2 9351 2678; fax: þ61 (0) 2 9351 4119. E-mail address:
[email protected] (C.R. Warren). 0038-0717/$ e see front matter Ó 2009 Elsevier Ltd. All rights reserved. doi:10.1016/j.soilbio.2009.11.017
argued that turnover of amino acids may dominate N flux in some ecosystems (Jones and Kielland, 2002). We still know considerably less about the pools of amino acids than inorganic N. In very general terms we know that free amino acids are typically less than 1% of the pool of DON (Stevenson, 1982; Christou et al., 2006), while concentrations of amino acids in the soil solution are in the order of 0.1e50 mM (Monreal and Mcgill, 1985; Jones et al., 2002; Yu et al., 2002; Christou et al., 2006). Fewer studies have quantified concentrations of individual amino acids, yet such information is critical for developing hypotheses and understanding soil N dynamics. For example, acidic, neutral and basic amino acids diffuse through the soil at different rates (Owen and Jones, 2001) while plants take up different amino acids at different rates (Persson and Nasholm, 2001). Clearly many of the questions regarding the importance of amino acids in ecosystem N cycling require knowledge of the concentrations of individual amino acids.
Author's personal copy
354
C.R. Warren, M.T. Taranto / Soil Biology & Biochemistry 42 (2010) 353e359
Of those studies quantifying individual amino acids, most have found that the pool of amino acids is dominated by a small number (generally 5) of abundant amino acids (Schulten and Schnitzer, 1997; Werdin-Pfisterer et al., 2009). Amino acids may vary over the course of a year, yet few studies have examined temporal trends. Temporal variation in concentrations of individual amino acids has been reported from arctic and alpine ecosystems (Kielland, 1995; Weintraub and Schimel, 2005) and heathlands (Abuarghub and Read, 1988). In contrast, in boreal forests concentrations of amino acids did not vary (Werdin-Pfisterer et al., 2009); but this may be because soils were only sampled three times and thus temporal variation was missed. Temporal variation in amino acids is probably due to seasonal shifts in the activity of plants and soil microbes. This is because the pool of free amino acids in the soil is affected by processes that vary over the course of a year such as plant and microbial uptake (Weintraub and Schimel, 2005), root exudation (Jones et al., 2005b), turnover of roots and mycorrhizae (Ruess et al., 2006), activity of exoenzymes responsible for depolymerisation of high molecular weight organic N into amino acids (Abuarghub and Read, 1988), and litter inputs (Schimel et al., 2004; Schmidt and Lipson, 2004). A key emerging theme is that temporal variation in the pool of microbial N may determine soil pools of amino acids (and inorganic N) (Weintraub and Schimel, 2005). In alpine soils, for example, decreases in microbial biomass during spring were correlated with increases in pools of free amino acids (Lipson et al., 1999). That microbial biomass might regulate pools of free amino acids is hardly surprising given that microbial N is commonly the largest dynamic N pool in ecosystems (Bardgett et al., 2007) and the microbial biomass is a strong competitor for amino acids (Warren, 2009). This study examines temporal variation in pools of amino acid N, inorganic N and microbial biomass in soils of a sub-alpine temperate grassland dominated by C3 grasses and sedges. Inorganic N and amino acids were quantified in 1 M KCl extracts of soil, and thus represent exchangeable pools. Previous studies at this site have shown that soil respiration varies by almost an order of magnitude between summer and winter (C. Warren unpublished data), and thus it would seem reasonable to expect temporal variation in microbial biomass and pools of inorganic and amino N. By collecting samples approximately every month for two years we were able to test the following hypotheses: 1) the pool of amino acids is generally larger than inorganic N; 2) temporal changes in the concentration of amino acids is related to changes in the size of the microbial N pool; 3) the amino pool is dominated by simple, neutral amino acids during warm months (due to rapid plant growth and large inputs of labile C and N), whereas during cold months the amino pool is dominated by more complex aromatic and basic amino acids (due to smaller inputs of labile C and N).
hiemata (soft snow-grass), which together have a cover-abundance of approximately 50%. Less abundant grasses are Austrodanthonia alpicola (crag wallaby-grass) and Austrostipa nivicola (alpine speargrass). Common sedges are Carex incomitata (hillside sedge) and Carex breviculmis (common sedge-grass). Above-ground dry mass of plants varies between 500 and 1000 g m2, 60e70% of which is grasses and sedges with the remaining 30e40% being herbs and shrubs. The dry mass of roots to a depth of 30 cm varies between 600 and 1000 g m2 (C. Warren unpublished data). The soil is a humic umbrosol (World Reference Base) derived from Silurian Mowomba granodiorite (approximately 433 1.5 million years old). From 0 to 30 cm the soil is a well-drained sandy loam without coarse fragments > 2 mm. Below 30 cm there are abundant coarse fragments of granodiorite. In the upper 30 cm, pH (H2O) is 4.5, organic C (Walkley and Black) is 12e17%, and total N is 0.2e0.3%. Soil samples were collected approximately every month from March 2007 to January 2009. Samples were collected less frequently over winter because roads were not cleared of snow and thus vehicular access was impossible. On each sampling occasion, a 5-cm diameter corer was used to collect 12 replicate soil samples (0e15 cm). Samples were collected randomly from within 50 50 m permanent plots. Soils were sieved to 2 mm in the field and stored in plastic bags and kept cool until they were extracted 24e48 h later. A weather station was established at the field site in May 2007. Volumetric soil water content was measured at depths of 30, 10 and 5 cm with a standing wave probe (MP 406, ICT International, Armidale, Australia), soil temperature was measured at the same depths with a type-T thermocouple. Data were stored as halfhourly averages on a data logger (Smart Logger, ICT International).
2. Materials and methods
2.3. Analysis of individual amino acids by capillary electrophoresis
2.1. Study site, experimental design and soil collection
Amino acids in 1 M KCl extracts were analysed by capillary electrophoresis, essentially as described previously (Warren, 2008). This method quantified 16 of the 20 protein amino acids plus gaminobutyric acid (GABA). The method could not reliably quantify cysteine, methionine, proline or serine. Leucine and isoleucine could not be separated from each other, so data are presented as leucine or isoleucine (Xle). A 5-mL aliquot of undiluted 1 M KCl extract was derivatised in a vial containing 100 nmol of dried 3-(2-furoyl)quinoline-2-carboxyaldehyde (FQ), 5 mL of methanol, 1 mL of internal standard (0.1 mM nor-leucine), and 10 mL of 10 mM KCN in 10 mM sodium tetraborate (pH 9.2). Derivatization was completed by incubating tubes at 55 C for 60 min 30 mL of water was then added to stop the
This study was conducted in a sub-alpine grassland in the Snowy Mountains of Australia (36 060 S; 148 320 E) at altitudes from 1500 to 1600 m above sea level. The mean annual maximum temperature is 13 C, while the mean annual minimum is 0.5 C. Annual precipitation is in the order of 1200 mm. The mean duration of snowcover is 2e3 months, though this varies among years. The vegetation in the region is a mosaic of grassland, woodland with grassy understorey, and woodland with shrubby understorey. All measurements in this study were made on sub-alpine grassland. The vegetation is dominated by perennial C3 grasses and sedges. The dominant grasses are Poa costiniana (bog snow-grass) and Poa
2.2. Extraction of soil samples One sub-sample of soil was dried (72 h at 80 C) to determine gravimetric soil water content. Two sub-samples were used to determine microbial biomass via the chloroform fumigation technique (Brookes et al., 1985). One sub-sample was immediately extracted with 0.5 M K2SO4 (8.0 g FW soil: 40.0 mL K2SO4), while the other was fumigated with CHCl3 for four days in an evacuated desiccator and then extracted with 0.5 M K2SO4 (8.0 g FW soil: 40.0 mL K2SO4). Aliquots of the fumigated and unfumigated K2SO4 extracts were digested to nitrate via persulfate oxidation (Cabrera and Beare, 1993). The fourth 8.0 g sub-sample was extracted with 1 M KCl (8.0 g FW soil: 40.0 mL KCl). In all cases, extracts were shaken end-to-end at 100 rpm for 90 min, centrifuged (3200 g, 10 min, 20 C) and then filtered through Whatman #1 filter paper. Extracts were stored at 80 C before analysis. For each set of samples four blanks were carried through extraction and analysis procedures.
Author's personal copy
C.R. Warren, M.T. Taranto / Soil Biology & Biochemistry 42 (2010) 353e359
In general, the exchangeable pool of nitrate was smaller than the pool of ammonium which was in turn smaller than the pool of amino acid N (Fig. 2). The most notable seasonal trend in nitrate was that concentrations were low during winter, while from spring to autumn there were several small maxima in nitrate concentrations. There were large differences between months in concentrations of ammonium and amino acid N, and trends in ammonium and amino acid N generally tracked each other. The most notable feature of the temporal trends was that they differed between the two years. In 2007 there were high concentrations of ammonium and amino acid N from winter to early spring (JuneeSeptember), whereas in 2008 concentrations were low and steadily decreased over winter. In December 2007 and January 2008 (summer) concentrations of amino acid N and ammonium were very low, whereas such low concentrations were not seen from December 2008eJanuary 2009. The primary similarity between years was that in both 2007 and 2008 there was a spike in nitrate and ammonium during mid-late spring (6 November 2007, 23 October 2008). Capillary electrophoresis detected 14 amino acids, but some were present at small concentrations and/or in a minority of samples. In order of decreasing abundance the detected amino acids were: Arg (26% of amino N pool), Val (20%), Gln (18%), Try 40
water Tsoil Tair
35
35
Jan 09
Dec 08
Oct 08
Nov 08
Sep 08
Jul 08
Aug 08
-5
Jun 08
-5
Apr 08
0
May 08
0
Mar 08
5
Jan 08
5
Feb 08
10
Dec 07
15
10
Oct 07
20
15
Nov 07
25
20
Sep 07
25
Jul 07
30
Aug 07
30
Volumetric water content (%)
40
Jun 07
Microbial biomass N was estimated as the difference in N between samples fumigated with CHCl3 and the initial (unfumigated) samples. A recovery coefficient was not used to account for incomplete lysis of microbes by CHCl3 and incomplete extraction by K2SO4. This is unavoidable because the relative extractability of the soil is unknown, and introduction of unknown factors was not desired. Hence, numbers presented here likely underestimate N in microbial biomass. Univariate analyses of relationships among air temperature, soil temperature, water content and pools of N were based on monthly means. Input variables were chosen on the basis that they were not logically or mathematically auto-correlated and would provide the most ‘information’, and thus only a sub-set of variables were used. Multivariate statistics were performed essentially as described previously (Warren et al., 2006). Principal component analysis (PCA) was based on a sub-set of 16 variables. We included only those amino acids that were present each month. We excluded the total pool of amino acids as an input variable because it is logically and mathematically auto-correlated with the concentration of individual amino acids. An acceptable principal component solution was determined based on visual examination of the Scree plot and the Kaiser criterion (all eigenvalues greater than 1). Component scores and PC loadings were determined after Varimax axis rotation so as to maximise the variance of the squared loadings (Johnson and Wichern, 1992). Once the optimal PC solution had been found, linear regression was used to determine if the arrangement of monthly samples in multivariate space was related to abiotic variables. Data are presented as the mean (one standard error) of 9e12 replicates. All statistics were performed with SPSS (release 16.0.1, SPSS Inc. Chicago IL USA).
3.2. Temporal variation in inorganic N, amino acid N and microbial N
Apr 07
2.5. Calculations and statistics
Over the two years for which weather data are available, the absolute maximum air temperature was 31 C while the absolute minimum was 13 C (Fig. 1). Daily minima were below freezing for 9e10 months of the year, whereas daily maxima generally remained above zero. There were 60 days of snowcover in 2007 and 68 in 2008. Soil temperatures followed a similar annual pattern as air temperature (Fig. 1). During the periods of snowcover, soil temperatures changed little between days. Moreover, under snowcover there was negligible diurnal variation in soil temperature (data not shown). Soil water content was consistently 30e40% (volumetric) from April to October, whereas from early November to late March there were periods when water content was reduced to 20% or less (Fig. 1).
May 07
Ammonium was measured by a modification of the tartratenitroprusside-hypochlorite method (Baethgen and Alley, 1989), nitrate was measured after reduction to nitrite with vanadium(III) and quantification with the Griess reagent (Miranda et al., 2001), while total amino acids were measured after reaction with ninhydrin (Jones et al., 2002). All analyses were performed on a monochromator-based microplate reader (Synergy 2, BioTek, Winooski, USA).
3.1. Climate
Mar 07
2.4. Measurement of nitrate, ammonium and total amino acids
3. Results
Temperature (oC)
labelling reaction and bring the total sample volume to w50 mL. Amino acids were separated and quantified with a commercially available CE system (P/ACE MDQ, Beckman-Coulter, CA, USA) equipped with an argon-ion laser and laser-induced fluorescence (LIF) detector. The argon-ion laser provided excitation light at 488 nm (c.f. excitation maximum of FQ ¼ 480 nm), while fluorescence emission was collected by a 590/20 nm band-pass filter, with a notch filter used to attenuate the background fluorescence. Separations were performed in uncoated fused silica capillaries (50 cm effective length 75 mm i.d. 365 mm o.d.). The run buffer consisted of 20 mM sodium tetraborate and 20 mM SDS at a pH of 9.2. Before each sample injection the capillary was flushed with 0.1 M HCl for 3 min, NaOH for 2 min and run buffer for 3 min. The original reference (Warren, 2008) did not include the rinse with HCl, but experience showed that this was necessary to achieve optimum reproducibility. Samples were injected by pressure (0.5 psi 5 s) and separated by applying a voltage of 25 kV for 12 min with the capillary thermostated at 25 C. LIF data were collected at 8 Hz and the detector was auto-zeroed 3 min into every run.
355
Fig. 1. Monthly averages of air temperature (Tair), soil temperature at 10 cm (Tsoil) and volumetric soil water content at 10 cm (water) for a temperate sub-alpine grassland in the Snowy Mountains of Australia. The shaded area denotes the period of permanent snowcover. Data were collected as half-hourly averages, but for clarity are shown as monthly averages.
Author's personal copy
356
C.R. Warren, M.T. Taranto / Soil Biology & Biochemistry 42 (2010) 353e359 500
100% NH4
NO3
450
Amino
400
80%
Amino acid N (mol %)
N (nmol g -1 DM)
350 300 250 200 150 100
Acidic 60%
Neutral Aromatic 40%
Basic
20%
50
0%
Fig. 2. Pools of exchangeable nitrate, ammonium and amino acid N in soil. Nitrate, ammonium and amino acids were determined by extraction of soil with 1 M KCl and subsequent colorimetric quantification. Data are means of 12 replicates. Error bars are one standard error.
(8%), Asn (8%), Phe (4%), Ala (3%), His (3%), GABA (3%), Gly (2%), Xle (1%), Glu (1%), Thr (
