lable at ScienceDirect

Soil Biology & Biochemistry 43 (2011) 835e844

Contents lists avai

Soil Biology & Biochemistry

journal homepage: www.elsevier .com/locate/soi lb io

Seasonal variation in soluble soil carbon and nitrogen across a grasslandproductivity gradient

Mark Farrell a,b,c,*, Paul W. Hill a, John Farrar a, Richard D. Bardgett b, Davey L. Jones a

a School of the Environment, Natural Resources and Geography, College of Natural Sciences, Bangor University, Gwynedd LL57 2UW, UKb Soil and Ecosystem Ecology Laboratory, Lancaster Environment Centre, Lancaster University, Bailrigg, Lancaster LA1 4YQ, UKcCSIRO Land and Water, PMB 2, Glen Osmond, SA 5064, Australia

a r t i c l e i n f o

Article history:Received 12 November 2010Received in revised form20 December 2010Accepted 22 December 2010Available online 12 January 2011

Keywords:Dissolved organic matterFertiliserNitrogen mineralisationOligopeptidePlant productivitySoluble organic nitrogenSoil organic matter

* Corresponding author. CSIRO Land and Water, PMAustralia. Tel.: þ61 (0) 8 8303 8664; fax: þ61 (0) 8 8

E-mail address: mark.farrell@csiro.au (M. Farrell).

0038-0717/$ e see front matter � 2011 Elsevier Ltd.doi:10.1016/j.soilbio.2010.12.022

a b s t r a c t

Understanding the fate and turnover of the pools that comprise dissolved organic nitrogen (DON) in soil iskey to determining its role in ecosystem functioning. We investigated seasonal changes of dissolvedorganic carbon (DOC) and nitrogen (DON) concentrations within four molecular weight (MW) size frac-tions across an altitudinal gradient (from lowland to montane systems), and quantified individual aminoacids and amino acid constituents of oligopeptidic-N, as well as nitrate and ammonium. We tested twoideas: first, that DON is more abundant than DIN in low-productivity relative to high-productivitygrassland ecosystems; and second, that the abundance of peptides and amino acids is likewise greater inlow- than high-productivity grassland. The most productive site had a history of inorganic fertiliserapplication, and hence in this site alone DIN was more abundant than DON. Plant productivity varied3-fold between the other sites, and DON was generally at higher concentrations in the sites of lowerproductivity both in absolute terms as well as relative to DIN, with a large increase observed in spring. Thefraction containing the highest concentration of the DON had a MW of >100 kDa, and in summer andautumn this fraction was more abundant at the lowest productivity site. We conclude that relationshipsbetween the abundance of DON relative to DIN and ecosystem productivity is dependent on season, andhence more complex than previously suggested, and that peptides are a dynamic and potentially nutri-tionally significant component of DON.

� 2011 Elsevier Ltd. All rights reserved.

1. Introduction

An ever-increasing body of evidence from many ecosystemsshows that a small fraction of dissolved organic nitrogen (DON) insoils: free amino acids (FAAs), may be directly assimilated by plants(Kielland, 1994; Näsholm et al., 1998; Nordin et al., 2001; Weigeltet al., 2003, 2005), circumventing the microbially mediated andpotentially rate-limiting mineralisation step. Whilst DON is a largefraction of total dissolved nitrogen (TDN) in most agricultural soils,it still represents only ca. 30% of the TDN pool (Christou et al., 2005).However, in nutrient limited soils, DON can make up the bulk ofTDN, and is therefore the N species of greatest interest in naturalecosystems in terms of N cycling and potential plant uptake.

DON is notone pool, but a heterogeneousmixture of compounds,themost studied of which are the free amino acids (FAAs) (Kielland,

B 2, Glen Osmond, SA 5065,303 8550.

All rights reserved.

1995; Bardgett et al., 2003; Jones et al., 2004, 2009; Näsholm et al.,2009; Geisseler et al., 2010), which are generally < 5% of total DON(Jan et al., 2009; Jones et al., 2009). Of the remaining un-charac-terised DON, 15N NMR studies have indicated that the majority ispeptidic or proteinaceousmoieties (Kögel-Knaber, 2006;Nannipieriand Eldor, 2009). Jämtgard et al. (2010) quantified soil-solutionproteins to be in the order of 50 times the concentration of soil-solution amino acids in agricultural soils, forming a much largerreservoir of potentially utilisable N. However, Jan et al. (2009)observed protein mineralisation rates in an agricultural soil to be50-times slower than those of an amino acid mixture e furtheremphasising the need to focus on purely directly-assimilatable N toascertain a complete picture of immediate N availability. As the highrate of turnover of FAAs (Jones et al., 2009)mirrors turnover rates oforganic acids such as oxalate and citrate, it has been suggested thata split of the DON pool between LMW (<1 kDa) and HMW (>1 kDa)compounds be used when investigating the rate of turnover andecological role of the DON pool (Jones et al., 2004). The LMW frac-tionwould therefore be limited to peptide chain lengths of ca. 2e12amino acids.

M. Farrell et al. / Soil Biology & Biochemistry 43 (2011) 835e844836

Direct utilisation of peptides by microbes may be limited bytheir Stoke’s radius (the effective size of themolecule), with a cutoffroughly equivalent to 650 Da (Payne, 1980; Weiss et al., 1991).A 1 kDa filter may therefore select peptides that could potentiallybe assimilated intact, rather than considering the hydrolysable-DON pool as a whole. The ericoid mycorrhizae are able to directlymobilise and assimilate oligopeptides intact and translocate this Nto plant roots (Bajwa and Read, 1985). It is therefore clear that inorder to increase our understanding of the nutritive function of theLMW DON pool, quantification of oligopeptidic-N is required.Whilst several authors have quantified combined amino acids aspeptides or proteins using acid hydrolysis followed by HPLC or GCdetection (Yu et al., 2002; Amelung et al., 2006; Roberts and Jones,2008; Jämtgard et al., 2010), to our knowledge only Isnor andWarman (1990) did molecular weight (MW) fractionation ofextracted peptides to <5 kDa prior to hydrolysis. Smolander andKitunen (2002) investigated size fractionation of DON in a podzolin a forest ecosystem, and found the fraction with the highestconcentration to be <1 kDa, with most DOC in the 10e100 kDafraction. This appears to be the only study that has quantified sizefractionation of DON, albeit in this case in a water extract.

There are very few studies quantifying individual amino acidpools and their changes either between soil types or over time, andthose that do exist are on radically different timescales: eitherhours in laboratory studies (Jones et al., 2009) or months for fieldwork (Bardgett et al., 2002), and we cannot yet causally link thesetimescales. Basic, neutral, and acidic amino acids diffuse throughthe soil at different rates (Owen and Jones, 2001), and plants takeup individual amino acids at different rates (Persson and Näsholm,2001; Weigelt et al., 2005; Harrison et al., 2007). Jones et al. (2009)demonstrate that rates of bulk FAA turnover are conserved acrossa global latitudinal gradient whenwater and temperature restraintsare removed, whereas Warren and Taranto (2010) argue thatquestions regarding the role of amino acids in ecosystem N cyclingrequire amino acids to be considered individually. Amino acidconcentrations may vary seasonally (Kielland,1995;Weintraub andSchimel, 2005) or remain relatively constant (Werdin-Pfistereret al., 2009). Whilst Warren and Taranto (2010) found large varia-tions in DIN and FAA concentrations between months in sub-alpinegrasslands; no overall seasonal patterns emerged.

It is therefore clear that DON is not a single component of soil N,that FAAs are not the only ecologically significant part of DON, andthat even FAAs should possibly be considered individually. Here,our main objective was to determine seasonal variation in the sizeand form of DON and DOC, and in particular to test the notion thatthe availability (and hence potential plant use) of DON (in the formof peptides and amino acids) relative to DIN is greatest in lowproductivity ecosystems. This was tested using an altitudinalgradient along which primary productivity decreased withincreased altitude and vegetation changed from Lolium perennedominated pasture at 15m above sea level (asl) to Festuca ovina andTricophorum cespitosum dominated upland grassland at 710 m asl.We hypothesise, following Smolander and Kitunen (2002), that thebulk of DON will be <10 kDa, whereas the bulk of DOC willbe >10 kDa. Further, we predict distinct seasonal changes due toseasonal plant growth and climatic patterns, and the direct andindirect effects these have on the soil. Given the differing infor-mation about seasonal variations in the soil AA pool (Werdin-Pfisterer et al., 2009; Warren and Taranto, 2010), we hypothesisethat FAAs and peptidic-AAs vary between sites and seasons due todifferences in DON turnover rates and plant inputs. Due to the factthat peptides may be broken down to FAAs extracellularly by soilmicrobes, we predict correlation between the FAA- and oligo-peptidic-AA pools, in that patterns in FAAs remain conserved tothose observed in the constituent peptidic-AAs.

2. Materials and methods

2.1. Site characterisation

An altitudinal gradient was established on grassland up a north-facing slope above Abergwyngregyn, Gwynedd, UK (53�14’N;4�10’W); above-ground net primary productivity (ANPP) decreasedwith increased elevation. Five sites were selected with different soilcharacteristics along this gradient, with four randomly positionedreplicate plots (10 m2) at each site. Although our study is based ona single elevation gradient, it was selected to include a broad range ofvegetation, soil and climatic conditions representative of wet,lowland, sub-montane, and montane regions of western Britain(Rodwell, 1992). The soils along this gradient cover 90% of the totalarea ofWales (Rudeforth et al.,1984). In order tominimise the sourcesof error associatedwith the use of a single gradient,we usedmultiple,randomly-located sites at each position along the elevation gradient.Mean annual temperature ranged from9.8 �C at Site 1 to 6.5 �C at Site5, with annual rainfall ranging from 800 mm at Site 1 to 2300mm atSite 5. Vegetation was recorded as percentage cover using 1 m2

quadrats at each plot, and ranged from L. perenne L./Trifolium repens L.dominated pasture at Sites 1 and 2 to Agrostsis canina L./Agrostiscapillaris L./Anthoxanthum odoratum L./Potentilla erecta (L.) Rauschelgrassland at Site 3, to F. ovina L./Juncus effusus L. dominatedmoorlandat Site 4, with F. ovina/T. cespitosum (L.) Hartman dominated moor-land at Site 5 (see Supplementary Information S1 for species list).Above-ground standing biomass and net annual primary produc-tivity (ANPP) were determined between April and October 2009according to Vile et al. (2006). Briefly, above-ground vegetation washarvested at the start of the growing season, and at the end of thegrowing season from an adjacent area fromwhich large grazers wereexcluded by cages. The difference was expressed as a mean growthrate in g dry matter m�2 d�1. In order to account for anthropogenicinputs of atmospheric N, which at the higher sites could confoundinterpretation,we also estimated atmosphericNdeposition at eachofthe sites by interpolation from the maps in NEGTAP (2001), witha 23% increase to account for organic N deposition (Gonzalez Benitezet al., 2009) not quantified in the NEGTAP survey.

2.2. Soil sampling and analysis

Soil samples were taken from the top 15 cm of each plot inDecember 2008, April 2009, July 2009 and October 2009, sealed insample bags and refrigerated at 4 �C until analysis. Soil pH andelectrical conductivity (EC; 1:2 v/v soil:distilled water) weredetermined with standard electrodes. Moisture content wasdetermined after drying at 80 �C for 72 h, and organic matter as losson ignition at 450 �C for 16 h. Bulk density was determined using100 cm3 cores (Rowell, 1994). Total C and N of both soil and vege-tation were determined using a Carlo Erba NA 1500 ElementalAnalyzer (Thermo Fisher Scientific, Milan, Italy). Extractablepotassium (K) and phosphorus (P) were extracted using a 0.5 Macetic acid (1:5 w/v) shaken for 1 h, then centrifuged for 10 min at3220 g before passing through a Whatman 42 filter (Quevauviller,1998). Potassium was analysed by flame emission spectroscopy(Sherwood 410 flame photometer: Sherwood Scientific, Cambridge,UK) and P by the colourimetric method of Murphy and Riley (1962).

Soil solution was extracted within 6 h of collection using thecentrifugal-drainage method of Giesler and Lundström (1993).Briefly, ca.1 kg field-moist soil was centrifuged at 3220 g for 30min,in containers drilled to facilitate drainage of the soil solution intoa collection vessel below. This method extracts plant-availablewater, in contrast to other extraction methods which may over-estimate pool sizes (Jones and Willett, 2006). Soil solutions werethen sterile-filtered (0.2 mm Whatman GDX Sterile PES filters) into

M. Farrell et al. / Soil Biology & Biochemistry 43 (2011) 835e844 837

sterile tubes and stored at 4 �C until analysis. The soil solution wasanalysed for soluble C and N using a TOC-V-TN analyzer (ShimadzuCorp., Kyoto, Japan), and soluble phenolics were analysed colouri-metrically after Swain and Hillis (1959). Microbial C and N weredetermined by chloroform fumigationeextraction according toVoroney et al. (2008), followed by analysis for C and N using theTOC-V-TN analyzer. NO3

� and NH4þ were analysed colourimetrically

using the methods of Miranda et al. (2001) and Mulvaney (1996)respectively. Total FAAs were analysed after 15-fold pre-concen-tration using vacuum-centrifugation and reconstitution in 100 mL18 MU water by anion-exchange chromatography using a DionexICS-3000 (Dionex, Sunnyvale, USA) in integrated amperometrymode using an ED40 electrochemical detector, AminoPac columnmaintained at 30 �C, and an alkaline mobile phase of NaOH andC2H3NaO2 (Clarke et al., 1999). A working minimum detection limitof 0.5 mMwas used due to noise and impurities in the samples, andarginine, glutamine, asparagine and proline were not determined,potentially leading to an underestimation of the total FAA andoligopeptidic-AA pools. The sub-1 kDa peptide fraction (oligopep-tides of up to ca. 8-12 AA units e i.e. those that could be feasiblytaken up intact by microbes) were separated by molecular weightfiltration under positive pressure of N2 (Amicon 8010 stirred celland Millipore YM1 filter membranes; Millipore Corp., Bedford,USA) (Smolander and Kitunen, 2002), and then hydrolysed with6 M HCl in an Ar atmosphere at 105 �C for 16 h (Amelung et al.,2006). The HCl was removed by vacuum-centrifugal evaporationbefore samples were reconstituted in 100 mL 18 MU water andanalysed by the same procedure as FAAs. Peptidic-N was thencalculated by subtraction of FAA-N. Fifteen individual AAs werequantified as FAAs, whilst thirteen were quantified in the hydro-lysates, due to the degradation of Histidine and Methionine duringhydrolysis. Also, any glutamine and asparagine present in thesamples is likely to have been degraded to glutamic and asparticacids respectively (Amelung et al., 2006), and is therefore the sumsof Glu þ Gln, and Asp þ Asn are presented here as Glx and Asxrespectively. 10 kDa and 100 kDa membranes (YM10 and YM100respectively) were used to give four size fractions of DOC/N andphenolics, namely < 1 kDa, 1e10 kDa, 10e100 kDa, and >100 kDa.

2.3. Statistical analyses

Differences in soil properties along the altitudinal gradient wereassessed using ANOVA with Tukey’s HSD post hoc test (SPSS v17.0,

Table 1General site and soil properties (samples collected December 2009). Values are mean �

Site 1 2

Eutric cambisol Dy

Location (UK grid reference) SH 653729 SHAltitude (m.a.s.l.) 15 40Estimated N deposition (kg ha�1 y�1) 12 12Plant diversity (Shannon-H) 0.64 � 0.09a 1.5Above-ground net primary productivity (g dw m�2 d�1) 5.39 � 0.36a 2.7*Above-ground C g m�2 95.2 � 4.7a 36*Above-ground N g m�2 6.60 � 0.66a 2.0Soil moisture content (%) 29.1 � 1.0a 45Soil organic matter (%) 7.8 � 0.5a 12Bulk density (g cm�3) 1.09 � 0.01c 0.6pH 6.92 � 0.28a 5.8Electrical conductivity (mS cm�1) 103 � 27 ab 45Extractable P (mg kg�1) 38 � 4a 5 �Extractable K (mg kg�1) 122 � 11 ab 36Total soil C (g kg�1) 34�3a 68Total soil N (g kg�1) 3.4 � 0.3a 6.5Microbial C (g kg�1) 3.9 � 0.5a 3.6Microbial N (g kg�1) 0.54 � 0.08a 0.4

* standing biomass at end of 2009 growing season.

SPSS Inc., Chicago, IL). In order to investigate changes between sizefractions and over time for DOC, DON, and phenolics, repeatedmeasures ANOVA was carried out whereby sampling date and sizefraction were the repeated measures, and site was a fixed factor.Where sphericity was not met, the HuynheFeldt correction wasapplied to allow univariate results to be used (Townend, 2002).Principal components analysis was used to group FAAs andpeptidic-AAs to condense the number of factors and to produceordination plots. Where a constituent AA was not present at thattime, that variable was omitted from the analysis.

3. Results

3.1. Site characteristics

The five sites along the productivity gradient differed greatly intheir plant ad soil characteristics (Table 1). Above-ground netprimary productivity (ANPP) varied 6-fold, although this waslargely attributed to ANPP being significantly greater at Site 1 thanat sites 3e5; plant diversity was greatest at Site 3. The sites athighest altitude (4 and 5) had moisture contents in excess of 60%,reflecting higher rainfall, reduced evapotranspiration, and highorganic matter content. Soil pH decreased steadily along theelevation gradient from 6.9 at Site 1, to around pH 4 at the highestsites (4 and 5). Here, both soil C and N contents were highest ona mass basis, with sites 1 and 2 having the lowest concentrations(P < 0.05).

3.2. Temporal variability of DIN and DON along theproductivity gradient

Inorganic N, and in particular nitrate, dominated the soluble N inthe soil solution of the lowest altitude, high-productivity site (Site1), whereas the DON was the dominant form of dissolved N acrossall other sites (Sites 2e5). NH4

þ was found at concentrations<3mg N L�1 across all five sites throughout the sampling year. Totaldissolved N was much greater in spring than at other times of theyear, mainly due to changes in DON rather than DIN; hence DONincreased relative to DIN at this time (Fig. 1). Beyond these findings,things were more complex. In summer, DIN was at very lowconcentrations in all except the two low altitude, higher produc-tivity sites.

SEM; n ¼ 4.

3 4 5

stric gleysol Cambic podzol Haplic podzol Fibric histosol

651725 SH 656716 SH 655701 SH 668680320 530 71020 26 31

5 � 0.03b 2.22 � 0.05c 1.41 � 0.14b 1.69 � 0.07b8 � 0.83 ab 0.92 � 0.34b 0.84 � 0.05b 1.45 � 0.92b.7 � 9.7b 65.1 � 4.9 ab 54.3 � 3.0 ab 68.3 � 18.1 ab2 � 0.48b 2.24 � 0.17b 2.08 � 0.19b 1.79 � 0.44b.7 � 1.9b 44.3 � 2.4b 65.5 � 1.7c 87.8 � 0.7d.6 � 0.7a 22.3 � 2.0b 50.9 � 3.3c 94.6 � 1.3d49 � 0.106b 0.451 � 0.083b 0.359 � 0.040 ab 0.0763 � 0.0096a9 � 0.05b 4.76 � 0.04c 4.12 � 0.20c 4.17 � 0.02c� 8a 36 � 4a 127 � 23b 52 � 3a1a 12 � 1a 44 � 12a 323 � 19b� 8a 76 � 8 ab 185 � 80 ab 273 � 86b� 11 ab 112�7bc 238 � 29c 405 � 12d� 0.8 ab 9.7 � 0.5bc 11.8 � 2.0c 23.0 � 1.1d� 0.6a 0.7 � 0.3a 2.9 � 0.3a 22.7 � 2.0b3 � 0.08a 0.17 � 0.05a 0.17 � 0.05a 2.69 � 0.30b

a b

dc

Fig. 1. Size fractionation of DON and DIN concentrations across the five sites at: a) winter; b) spring; c) summer; and d) autumn sampling times. Values are mean � SEM; n ¼ 4.

M. Farrell et al. / Soil Biology & Biochemistry 43 (2011) 835e844838

3.3. Concentrations and trends in size fractions of DON, free aminoacid- and peptidic-N

We fractionated the DON into four size classes, and >100 kDacommonly accounted for the majority of the DON. Of the others,the <1 kDa DON fraction was generally most abundant, with theintermediate size fractions contributing much less to the total DON.This was particularly obvious in spring where the total amount ofDON was ca. 4 times higher than during the other seasons(F3,45 ¼ 24.5, P< 0.001). The significant interaction between seasonand size fraction (F9,135 ¼ 8.52, P ¼ 0.001) demonstrated that theseasonal effect was greater in some size fractions, most notablythe<1 kDa fraction in spring. No differences in total DON across thesites (F4,15 ¼1.46, P ¼ 0.265) were detected, and this was consistentbetween seasons (F12,45 ¼ 1.48, P ¼ 0.226), indicating seasonalvariations in total DON were similar across the altitude gradient.

We examined in more detail the FAAs and oligopeptidesof <1 kDa. We did not analyse further the DON size classesof >1 kDa, as molecules of this size are considered too large forintact microbial uptake (Payne, 1980). Due to the low concentra-tions, we present FAA- and oligopeptidic-N in terms of the TDN inthe soil solution on a w/w basis (Supplementary Information S2and S3). It should be noted that not all potential individual soilamino acids (or oligopeptidic amino acids) could be quantified inthis study, as the very low soil-solution concentrations of theseanalytes can prove problematic, potentially leading to a slightunderestimation of the total FAA-N and oligopeptidic-N pools.When considered as a single pool, total FAA-N concentrations

varied between ca. 428 and 3531mg N kg�1 TDN between sites andseasons, while total oligopeptidic (<1 kDa) N varied between 3496and 18626 mg N kg�1 TDN. There were no obvious trends in theproportion of peptidic-N either between sites or throughout theseasons, with the greatest average concentrations in spring andautumn. However, as an average across all sites and seasons,peptidic-N was present in concentrations six times that of FAA-N.The lowest total concentrations of FAA-N were found at the lowertwo sites (sites 1 and 2) with the greatest plant productivity, albeitwith seasonal variability.

Individually, the amino acids glycine, lysine and threoninedominated the FAA pool across the altitude gradient and over time,whilst alanine, aspartate/asparagine, glutamate/glutamine, lysine,threonine and valine were all prominent in the oligopeptide poolacross sites and seasons. Principal component analyses werecarried out on both the FAAs and peptidic-AAs to investigatemultivariate trends in relative AA concentration between sites andacross the seasons, and the resultant ordination plots for compo-nents 1 and 2 (accounting for an average of 51% of the total variancefor the FAA dataset and 75% of the total variance for the peptidic-AAdataset) are presented as Figs. 2 and 3 respectively. There is muchvariability of FAAs within some sites, and generally lower variabilityfor peptidic-AA. However, Fig. 2 indicates that the patterns ofindividual FAA abundance shifted between seasons, with Sites 1and 2 being grouped separate from Sites 3, 4 and 5 in winter;whereas Sites 1, 2 and 3 were closely grouped in spring, with Sites 4and 5 more separate, whilst exhibiting much higher within-sitevariability. In summer, Sites 1 and 4 were separate from a central

a b

dc

Fig. 2. Principal components analysis of free individual amino acid-N concentrations across the productivity gradient at: a) winter [PC1 ¼ 32% variance, PC2 ¼ 19% variance];b) spring [PC1 ¼ 47% variance, PC2 ¼ 14% variance]; c) summer [PC1 ¼ 22% variance, PC2 ¼ 20% variance]; and d) autumn [PC1 ¼ 26% variance, PC2 ¼ 24% variance].

M. Farrell et al. / Soil Biology & Biochemistry 43 (2011) 835e844 839

cluster of the remaining three sites, whilst at the final autumnsampling, only Site 3 was obviously separate in terms of bothprincipal components 1 and 2. However, much greater variabilitywas observed across PC1. The amino acids strongly positively andnegatively associated with principal components 1 and 2 variedgreatly between seasons, with no single amino acid conserved toa principal component across the four sampling times.

As with the individual FAAs, component peptidic-AAs varied intheir relative abundance between sites across the seasons. Inwinter, all five sites showed relatively discrete amino acid distri-butions, with Sites 2 and 4 being most closely related. In spring,Sites 3 and 5weremore separate than the loose clustering of Sites 1,2 and 4. In summer, only Site 3 differed obviously from the clusterof the other four sites, whilst in autumn, the fertilised soil wasseparate from two loose groupings of Sites 2 and 3, and Sites 4 and5. Compared to the individual FAA principal component loadings,the component peptidic-AAs alanine, aspartate/asparagine, gluta-mate/glutamine, glycine, isoleucine, leucine and phenylalaninewere strongly positively associatedwith PC1 across all four seasons,indicating that these AAs were consistently important whendescribing the variance in peptidic-AA concentrations betweenseasons, whereas the concentrations of the remaining six compo-nent peptidic-AAs (cysteine, lysine, serine, threonine, tyrosine andvaline) were of less importance across the whole year.

3.4. Size fractionation of DOC and phenolics

DOC concentrations varied from 32 mg L�1 to 257 mg L�1 acrossthe sites and seasons (Fig. 4), whilst the phenolic component

consistently ranged from below 2 mg L�1 in the lowest altitudesites, to 3e10 mg L�1 at the highest (Fig. 5). With the exception ofSite 1, the majority of DOC had anMW>100 kDa across the seasons(Fig. 4). For Site 1, the concentration of DOC across seasons followedthe pattern [<1 kDa] > [>100 kDa] > [1e10 kDa] ¼ [10e100 kDa],whereas the pattern [>100 kDa] > [<1 kDa] > [1e10 kDa] ¼[10e100 kDa] was conserved across Sites 2e5. There were signifi-cant differences in DOC concentration along the altitude gradient(F4,15 ¼ 5.88, P ¼ 0.005), between seasons (F3,45 ¼ 21.0, P < 0.001)and fractions (F3,45 ¼ 98.1, P < 0.001). There was also a significantinteraction between season and size fraction (F9,135 ¼ 5.32,P¼ 0.001): the seasonal effect was greater in some size fractions, sothat the overall increase in DOC in spring did not result in a uniformincrease of DOC in each size fraction. Significant interactionsbetween size fraction and site (F12,135 ¼ 6.06, P < 0.001) wereobserved, indicating that the distribution of DOC between the sizefractions varied across the gradient. However, because there waslittle seasonal interactionwith site (F12,135¼1.97, P¼ 0.071), overallseasonal patterns remained relatively uniform across the gradient.

With the exception of Site 1, total DOC and the four size fractionsshowed no obvious trend along the altitude gradient, whilst thephenolic component of DOC showed a significant (F4,15 ¼ 43.3,P < 0.001) 5e10-fold increase towards Site 5 that was consistentacross the four seasons. There were significant differences inphenolic concentration with season (F3,45 ¼ 19.5, P < 0.001) andbetween fractions (F3,43 ¼ 43.3, P < 0.001). As with DOC, there wasa significant interaction between season and size fraction(F9,135 ¼ 10.6, P < 0.001), so the seasonal effect is greater in somesize fractions, particularly at Site 1. Generally, the distribution of

a b

c d

Fig. 3. Principal components analysis of hydrolysed oligopeptidic individual amino acid-N concentrations across the productivity gradient at: a) winter [PC1 ¼ 38% variance,PC2 ¼ 30% variance]; b) spring [PC1 ¼ 58% variance, PC2 ¼ 15% variance]; c) summer [PC1 ¼ 65% variance, PC2 ¼ 13% variance]; and d) autumn [PC1 ¼ 62% variance, PC2 ¼ 20%variance].

M. Farrell et al. / Soil Biology & Biochemistry 43 (2011) 835e844840

phenolics followed the trend [>100 kDa] > [<1 kDa] >

[1e10 kDa] > [10e100 kDa], however, at Site 1 across all seasons,and in the autumn across all sites, the smallest fraction dominatedthe phenolics.

4. Discussion

Through sampling an altitudinal gradient of grassland sites, wehave shown that DON is relatively more abundant than DIN in lowproductivity, unfertilised grassland of higher altitudes, than in highproductivity, intensively managed lowland grassland. We have alsoshown that peptides of <1 kDa are a quantitatively significant partof the DON pool whatever the management, productivity or alti-tude of the site, and that these peptides show only modest seasonaldynamics in their relative abundance. Next we consider our find-ings in more detail, in the context of soil N status and dynamicsacross the gradient of above-ground grassland productivity.

4.1. Seasonal variations in dissolved nitrogen

In our study, NO3� was the largest component of the TDN pool

only at the low altitude, high-productivity and fertilised grasslandsite in winter, summer and autumn, with concentrations beingslightly reduced in spring, a reduction masked by the large increasein DON. Across the higher altitude, lower productivity sites (i.e.2e5), DONwas always the dominant pool, which is consistent withthe hypothesis that DON is the quantitatively dominant pool of low

productivity ecosystems (Schimel and Bennett, 2004; Christou et al.,2005; Kranabetter et al., 2007; Näsholm et al., 2009). For instance,Christou et al. (2005) noted large variations in the distribution ofDON,NO3

� andNH4þ across various land uses, butwithNO3

� generallydominating the TDN pool in high-productivity systems, and NH4

þ

only contributing marginally to the TDN pool in any site. With theDOC and phenolic concentration data reported here, we detected noobvious seasonal trends; however, in line with the DOC data, weobserved a sharp increase in DON in spring across all sites. Wetentatively ascribe this spring increase in DON to increased N fluxesat the start of the growing when soils warm, soil microbes becomemore active, and plant growth increases. This is consistent withstudies of alpine grasslands, where a pulse in soluble N in the formof protein has been detected (Lipson and Schmidt, 2004), andcoincides with a peak in soil protease activity, which facilitates thesupply of amino acids for plant uptake at a time of high plantnitrogen demand (Raab et al., 1999; Bardgett et al., 2005).

4.2. The significance of oligopeptidic-N

As an average across the gradient and all seasons, we found thatpeptidic-N was present in concentrations as high as six times thatof FAA-N, representing 9 g peptidic-N kg�1 TDN. However, given anestimated average chain length of 5e6 AAs in the <1 kDa fraction,this could represent an equal concentration of penta-/hexa-peptides to FAAs in terms of individual molecules. As the oligo-peptidic-N pool is so poorly studied, it is difficult to assess whether

a b

c d

Fig. 4. Size fractionation of DOC across the five sites at: a) winter; b) spring; c) summer: and d) autumn sampling times. Values are mean � SEM; n ¼ 4.

M. Farrell et al. / Soil Biology & Biochemistry 43 (2011) 835e844 841

the large increase in peptidic-N over FAA-N is representative of soilsglobally. Indeed, there is a dearth of knowledge about individualamino acids in soils, including which are quantitatively dominant.Further, those at higher concentrations may not be of greaterimportance to microbial and/or plant nutrition since their flux maybe less than others at lower concentrations. Another possibility isthat the concentration of individual FAAs is a function of thecomposition of proteins deposited in the soil.

With the exception of Isnor and Warman (1990), there appearsto be no literature explicitly measuring oligopeptidic-N concen-trations, either as component hydrolysed amino acids, or as a bulkmeasurement. We considered five greatly differing temperategrassland soils across an altitude gradient e representative ofa broad range of soil vegetation and climate conditions of lowland,sub-montane and montane regions of western Britain (Rodwell,1992) e which each support discrete plant communities and aresubject to differing intensities of land management. Sites 3, 4 and 5were at higher altitudes and of low productivity (Table 1), and insuch environments DON (or at least FAAs) plays a pivotal role inplant N nutrition (Streeter et al., 2000; Bardgett et al., 2003;Weigelt et al., 2003, 2005). As this pool of peptidic-N is in muchgreater concentrations than FAA-N across all our sites, we ask: whatis its significance? Schimel and Bennett (2004) discussed theemerging view that plants successfully compete with soil microbesfor FAA-N across many ecosystems. Specifically, they proposed thatthe rate-limiting step in plant N uptake is not the microbial min-eralisation of soil organic matter, but rather the degradation of SOMto monomeric compounds. Given that some mycorrhizal fungi are

capable of taking up peptides intact in vitro (Bajwa and Read, 1985),and recently non-mycorrhizal plants have been shown to take upintact proteins (Paungfoo-Lonhienne et al., 2008); it seemsreasonable to suggest that the proposed bottleneck may not bebetween SOM and monomers, but between proteins released oncell death and peptides that can be taken up by plant roots andassociated symbionts. Further, some microbes favour peptides asa nutrient source on an energetic basis (Matthews and Payne,1980), so competition in the rhizosphere for DON may occursooner in the biodegradation process than FAA-N.

4.3. Distribution of free- and oligopeptidic-amino acids

The literature suggests that LMWDON compounds such as FAAsand oligopeptidic-AAs accumulate more in low productivitysystems where DON turnover is considered to be slower (Schimeland Bennett, 2004). Production of FAAs and oligopeptidic-AAsprobably increases as temperatures increase in spring and HMWDON is degraded, together with increased input as root exudatesfrom the active plant community. However, since both FAAs andsmall peptides can turnover in hours, it is premature to ascribechanges in concentration in the soil solution measured overmonths to any specific change in flux. Our analysis of individualFAAs and constituent oligopeptidic-AAs showed variation betweenthe five sites, and the relative distribution of AAs varied across thefour seasons. However, despite the significant increases of DOC andDON in spring, there were no trends in total concentrations ofeither the FAAs or oligopeptidic-AAs either in spring or across the

a b

dc

Fig. 5. Size fractionation of dissolved phenolics across the five sites at: a) winter; b) spring; c) summer; and d) autumn sampling times. Values are mean � SEM; n ¼ 4.

M. Farrell et al. / Soil Biology & Biochemistry 43 (2011) 835e844842

altitude gradient. Most of the work on individual amino acids hasused water or K2SO4 extractions as opposed to direct measure-ments on the soil solution. This further confounds comparisons ofour data with existing work (Jones et al., 2005; Jones and Willett,2006) and the true representativeness of any study of labilecompounds involving extractions at room temperature (Rousk andJones, 2010). Yu et al. (2002) used soil solution to assess changes inFAA distribution between individual AAs in soils across theEcological Staircase on the northern California coast, and noteddifferences between soils under different plant species. Theyproposed that ornithine, lysine and tryptophan may have beentaken up preferentially by Pinus muricata due to their absencecompared to soils under Cupressus pygmaea. Indeed, McKane et al.(2002) found that plant species composition in the arctic tundracoexist on the basis of uptake of different chemical forms of N, withdominant species preferentially using the most abundant N formsand the least dominant species using the least abundant N forms.However, 15N labelling studies in grasslands similar to those usedhere show that while co-existing plant species vary markedly inuptake rates of different chemical forms of N, they all preferentiallyuptake inorganic N over more complex amino acids, and moresimple amino acids over complex ones (Harrison et al., 2007).Further studies are clearly needed to test the significance of pref-erential uptake of N on the basis of chemical form, and individualrates of cycling.

Of the 15 FAAs quantified in our study, glycine, lysine andthreonine accounted for 70% of the total FAA-N pool when averaged

across sites and seasons.When considered individually, glycinewasgenerally more prevalent in winter, whereas threonine was ingreater quantities in autumn. Overall, Fig. 2 illustrates that thedistribution of FAAs differed both between sites and seasons, andFig. 3 demonstrates further variability between sites and seasonswhen constituent oligopeptidic-AAs are considered. Here, alanine,aspartate/asparagine, glutamate/glutamine, lysine, threonine andvaline accounted for 75% of the total oligopeptidic-N. With theexception of dialanine, which forms the backbone of peptidoglycanin bacterial cell walls (Kitamura et al., 2009), it is unlikely thatsignificant homopeptides exist in protein residues, possiblyexplaining the greater number of component AAs. Althougha bacterially-derived pool of homopeptidic alanine may be present,current techniques for identifying intact oligopeptides in soilsolution are limited (Roberts and Jones, 2008). Lastly, combinedFAA- and oligopeptidic-N of <1 kDa on average constituted only 1%of the TDN pool across all our sites. Whilst this is lower than somestudies investigating extractable N (Warren and Taranto, 2010), it iswithin the same order of magnitude of FAA-N determined ina global sample of soils (ca. 0.4 mg N l�1: Jones et al., 2009).

4.4. Molecular weight of DOC and DON is important

In a maize-cropped soil, bulk soluble organic matter fitteda two-pool model where one pool turned over rapidly (<1 d) andthe second pool took ca. 80 d to decompose (Gregorich et al., 2003),and Kalbitz et al. (2003) concluded that DOM is not necessarily the

M. Farrell et al. / Soil Biology & Biochemistry 43 (2011) 835e844 843

most biodegradable fraction of organic matter. By using MWfiltration to size-fractionate the DOC and DON content in the soilsolution of five soil types along a productivity gradient, we havedemonstrated that the greatest proportion is within the largest sizefraction (>100 kDa) at all sites studied. The only exception was forDOC at the low altitude, high-productivity Site 1, which regularlyreceives fertiliser inputs and is periodically tilled and reseeded.Kalbitz et al. (2003) noted that DOC from agricultural soils miner-alised quicker than that from forest soils, and whilst all our soils arefrom grass-dominated sites, only Site 1 is intensively managed. Ofthe MW fractions we studied, the next largest fraction of DOC andDON was in the <1 kDa fraction. The implication is that the DOCand DON pools can usefully be split into two pools: one consistingof large, highly complex molecules of DOC and DON which mayturnover with a half-time of weeks to months, consistent withseasonal changes in concentration of TDN, and a second poolconsisting of low MW organic molecules which could turnoverrapidly (in hours; Boddy et al., 2007). Haynes (2005) noted thatonly 10e40% of DOM is readily degradable, and our amountof the <1 kDa fraction is of this order, again suggesting thatthe <1 kDa fraction could be used as a surrogate for the amount ofrapidly turning over C in soil solution. Most work characterisingindividual compounds within the DOC and DON pools has indeedfocussed on LMW compounds such as organic acids, sugars, aminosugars, and amino acids (van Hees et al., 2005; Roberts et al., 2007),which are all known to be rapidly used, but only represent a smallproportion of the total DOM pool.

The largest component of the DOC and DON in all but the mostproductive (Site 1) of soils is the>100 kDa fraction, and the fact thatwe have identified little in the way of an intermediate pool(1e100 kDa) implies that DOC/N >100 kDa is released from thedecaying organisms, and the resultant intermediate polymers areeither taken up directly or degraded rapidly to short-chain poly-mers of <1 kDa. Marschner and Kalbitz (2003) found a much largerproportion of DOC in the <1 kDa fraction than that of their1e10 kDa fraction, however, no information is given about largersize fractions. This is in contradiction to the earlier study ofSmolander and Kitunen (2002), who found DOC to be relativelyevenly distributed between fractions, with the 10e100 kDa sizecontaining most DOC across most treatments in their study. Theirstudy also found that the <1 kDa fraction in their soils containedthe highest proportion of DON, whereas in our study, the propor-tion of DON in the <1 kDa fraction is lower than that of DOC.Smolander and Kitunen (2002) worked with forest soils, whereasours are from grasslands. If DOM is more degradable in agriculturalsoils (Kalbitz et al., 2003), we might have expected a higherproportion of DON to be present in the <1 kDa fraction than wasfound in the present study.

4.5. Seasonal variations in dissolved organic carbon

Whilst no obvious seasonal trends were observed for either bulkDOC or phenolics, spring sampling found a two-fold increase intotal concentrations of DOC, with an increase of two- to four-fold inthe <1 kDa fraction. This is confirmed by the significant interactionbetween fraction and season (F9,135 ¼ 5.32, P ¼ 0.001), and wetentatively ascribe this pulse to a build-up of labile compounds overwinter due to lower microbial activity. Plant seasonal cycles alter Cand N availability in soils (Bardgett et al., 2002, 2005; Marschnerand Kalbitz, 2003; Chu and Grogan, 2010; Kaiser et al., 2010), andmicrobial decomposition processes in soils are sensitive to theavailability of labile C and N (Schimel and Weintraub, 2003). Incontrast to bulk DOC, the phenolic constituent showed an obviousincrease along the altitude gradient, with the total amount at thelowest productivity, high altitude site (Site 5) always being in the

order of ten-times greater than that at the lowest altitude, high-productivity site (Site 1). Given the range of soil types, and the largeincrease in organicmatter along our altitudinal gradient, we ascribethe trend to an increase in ligninous material and reduced rates ofdecomposition analogous to highly organic soils.

In summary, we have shown that DON is relatively more abun-dant than DIN when plant productivity is lower, and that DON mayconsidered as two functional pools, one of<1 kDa and turning overmuch more rapidly than larger fractions which are dominant butmay cycle more slowly and account for seasonal changes in DONabundance. Additionally, peptides of <1 kDa are a quantitativelysignificant part of the DON pool that show only modest seasonaldynamics in their relative abundance. Collectively, ourfindings raisethe possibility that oligopeptidic-N may provide a previouslyunrecognised pool of readily-assimilatable N for microbes andplants; however, further studies are required to verify this.

Acknowledgements

The authors gratefully acknowledge funding from NERC (Grantnumber NE/E017304/1) awarded to RDB, DLJ and JF, and technicalassistance from Helen Quirk of Lancaster University and FrancisGuyver, Jonathan Roberts, Llinos Hughes, Mark Hughes and SarahChesworth of Bangor University.

Appendix. Supplementary material

Supplementary material related to this article can be found atdoi:10.1016/j.soilbio.2010.12.022.

References

Amelung, W., Zhang, X., Flach, K.W., 2006. Amino acids in grassland soils: climaticeffects on concentrations and chirality. Geoderma 130, 207e217.

Bajwa, R., Read, D.J., 1985. The biology of mycorrhizal in the Ericaceae. IX. Peptidesas nitrogen sources for the ericoid endophyte and for mycorrhizal and non-mycorrhizal plants. New Phytologist 101, 459e467.

Bardgett, R.D., Streeter, T.C., Cole, L., Hartley, I.R., 2002. Linkages between soil biota,nitrogen availability, and plant nitrogen uptake in a mountain ecosystem in theScottish Highlands. Applied Soil Ecology 19, 121e134.

Bardgett, R.D., Streeter, T.C., Bol, R., 2003. Soilmicrobes compete effectivelywith plantsfor organic-nitrogen inputs to temperate grasslands. Ecology 84, 1277e1287.

Bardgett, R.D., Bowman, W.D., Kaufmann, R., Schmidt, S.K., 2005. A temporalapproach to linking aboveground and belowground ecology. Trends in Ecologyand Evolution 20, 634e641.

Boddy, E., Hill, P.W., Farrar, J., Jones, D.L., 2007. Fast turnover of low molecularweight components of the dissolved organic nitrogen pool of temperategrassland field soils. Soil Biology & Biochemistry 39, 827e835.

Christou, M., Avramides, E.J., Roberts, J.P., Jones, D.L., 2005. Dissolved organicnitrogen in contrasting agricultural systems. Soil Biology & Biochemistry 37,1560e1563.

Chu, H., Grogan, P., 2010. Soil microbial biomass, nutrient availability and nitrogenmineralization potential among vegetation-types in a low arctic tundra land-scape. Plant and Soil 329, 411e420.

Clarke, A.P., Jandik, P., Rocklin, R.D., Liu, Y., Avdalovic, N., 1999. An integratedamperometry waveform for the direct, sensitive detection of amino acids andamino sugars following anion-exchange chromatography. Analytical Chemistry71, 2774e2781.

Geisseler, D., Horwarth, W.R., Joergensen, R.G., Ludwig, B., 2010. Pathways ofnitrogen utilization by soil microorganisms e A review. Soil Biology &Biochemistry 42, 2058e2067.

Giesler, R., Lundström, U.S., 1993. Soil solution chemistry e the effects of bulkingsoil samples and spatial variation. Soil Science Society of America Journal 57,1283e1288.

Gonzalez Benitez, J.M., Cape, J.N., Heal, M.R., van Dijk, N., Diez, A.V., 2009. Atmo-spheric nitrogen deposition in SoutheEast Scotland: quantification of theorganic nitrogen fraction in wet, dry and bulk deposition. Atmospheric Envi-ronment 43, 4087e4094.

Gregorich, E.G., Beare, M.H., Stoklas, U., St-Georges, P., 2003. Biodegradability ofsoluble organic matter in maize-cropped soils. Geoderma 113, 237e252.

Harrison, K.A., Bol, R., Bardgett, R.D., 2007. Preferences for different nitrogen formsby coexisting plant species and soil microbes. Ecology 88, 989e999.

Haynes, R.J., 2005. Labile organic matter fractions as central components of thequality of agricultural soils: an overview. Advances in Agronomy 85, 221e268.

M. Farrell et al. / Soil Biology & Biochemistry 43 (2011) 835e844844

Isnor, R.A., Warman, P.R., 1990. Amino acid composition of soil peptides chroma-tographed by high performance liquid chromatography on C18 and C8 columns.Biology and Fertility of Soils 10, 213e217.

Jämtgard, S., Näsholm, T., Huss-Danell, K., 2010. Nitrogen compounds in soil solu-tions of agricultural land. Soil Biology & Biochemistry 42, 2325e2330.

Jan, M.T., Roberts, P., Tonheim, S.K., Jones, D.L., 2009. Protein breakdown representsa major bottleneck in nitrogen cycling in grassland soils. Soil Biology &Biochemistry 41, 2272e2282.

Jones, D.L., Willett, V.B., 2006. Experimental evaluation of methods to quantifydissolved organic nitrogen (DON) and dissolved organic carbon (DOC) in soil.Soil Biology & Biochemistry 38, 991e999.

Jones, D.L., Shannon, D., Murphy, D.V., Farrar, J.F., 2004. Role of dissolved organicnitrogen (DON) in soil N cycling in grassland soils. Soil Biology & Biochemistry36, 749e756.

Jones, D.L., Healey, J.R., Willett, V.B., Farrar, J.F., Hodge, A., 2005. Dissolved organicnitrogen uptake by plants e an important N uptake pathway? Soil Biology &Biochemistry 37, 413e423.

Jones, D.L., Kielland, K., Sinclair, F.L., Dahlgren, R.A., Newsham, K.K., Farrar, J.F.,Murphy, D.V., 2009. Soil organic nitrogen mineralization across a global lat-itudinal gradient. Global Biogeochemical Cycles 23, GB1016.

Kaiser, C., Koranda, M., Kitzler, B., Fuchsleuger, L., Schnecker, J., Schweiger, P.,Rasche, F., Zechmeister-Boltenstern, S., Sessitsch, A., Richter, A., 2010. Below-ground carbon allocation by trees drives seasonal patterns of extracellularenzyme activities by altering microbial community composition in a beechforest soil. New Phytologist 187, 843e858.

Kalbitz, K., Schmerwitz, J., Schwesig, D., Matzner, E., 2003. Biodegradation of soil-derived dissolved organic matter as related to its properties. Geoderma 113,273e291.

Kielland, K., 1994. Amino-acid-absorption by arctic plants e implications for plantnutrition and nitrogen cycling. Ecology 75, 2373e2383.

Kielland, K., 1995. Landscape patterns of total free amino acids in arctic tundra soils.Biogeochemistry 31, 85e98.

Kitamura, Y., Ebihara, A., Agari, Y., Shinkai, A., Hirotso, K., Kuramitsu, S., 2009.Structure of D-alanine-D-alainine ligase from Thermus thermophilus HB8:cumulative change and enzyme-ligand interactions. Acta ChrystallographicaSection D 65, 1098e1106.

Kögel-Knaber, I., 2006. Chemical structure of organic N and organic P in soils. In:Nannipieri, P., Smalla, K. (Eds.), Nucleic Acids and Proteins in Soils. Spring-ereVerlag, Berlin, pp. 23e48.

Kranabetter, J.M., Dawson, C.R., Dunn, D.E., 2007. Indices of dissolved organicnitrogen, ammonium and nitrate across productivity gradients of boreal forests.Soil Biology & Biochemistry 39, 3147e3158.

Lipson, D.A., Schmidt, S.K., 2004. Seasonal changes in an alpine soil bacterialcommunity in the Colorado Rocky Mountains. Applied & EnvironmentalMicrobiology 70, 2867e2879.

Marschner, B., Kalbitz, K., 2003. Controls of bioavailability and biodegradability ofdissolved organic matter in soils. Geoderma 113, 211e235.

Matthews, D.M., Payne, J.W., 1980. Transmembrane transport of small peptides.Current Topics in Membranes and Transport 14, 331e425.

McKane, R.B., Johnson, L.C., Shaver, G.R., Nadelhoffer, K.J., Rastetter, E.B., Fry, B.,Giblin, A.E., Kielland, K., Kwiatkowski, B.L., Laundre, J.A., Murray, G., 2002.Resource-based niches provide a basis for plant species diversity and domi-nance in arctic tundra. Nature 415, 68e71.

Miranda, K.M., Espey, M.G., Wink, D.A., 2001. A rapid, simple spectrophotometricmethod for simultaneous detection of nitrate and nitrite. Nitric Oxide-Biologyand Chemistry 5, 62e71.

Mulvaney, R.L., 1996. Nitrogen e inorganic forms. In: Sparks, D.L., Page, A.L.,Helmke, P.A., Loeppert, R.H., Soltanpour, P.N., Tabatabai, M.A., Johnston, C.T.,Sumner, M.E., Bartels, J.M., Bigham, J.M. (Eds.), Methods of Soil Analysis Part 3:Chemical Methods. Soil Science Society of America Inc, Madison, WI, pp.1123e1184.

Murphy, J., Riley, J.P., 1962. A modified single solution method for the determinationof phosphate in natural waters. Analytica Chimica Acta 27, 31e36.

Nannipieri, P., Eldor, P., 2009. The chemical and functional characterization of soil Nand its biotic components. Soil Biology & Biochemistry 41, 2357e2369.

Näsholm, T., Ekblad, A., Nordin, A., Giesler, R., Hogberg, M., Hogberg, P., 1998. Borealforest plants take up organic nitrogen. Nature 392, 914e916.

Näsholm, T., Kielland, K., Ganeteg, U., 2009. Uptake of organic nitrogen by plants.New Phytologist 182, 31e48.

NEGTAP., 2001. Transboundary Air Pollution: Acidification, Eutrophication andGround-Level Ozone in the UK. Available online. http://www.freshwaters.org.uk/resources/documents/negtap_2001_final_report.pdf (Accessed 24.02.10).

Nordin, A., Hogberg, P., Näsholm, T., 2001. Soil nitrogen form and plant nitrogenuptake along a boreal forest productivity gradient. Oecologia 129, 125e132.

Owen, A.G., Jones, D.L., 2001. Competition for amino acids betweenwheat roots andrhizosphere microorganisms and the role of amino acids in plant N acquisition.Soil Biology & Biochemistry 33, 651e657.

Paungfoo-Lonhienne, C., Lonheinne, T.G.A., Rentsch, D., Robinson, N., Christie, M.,Webb, R.I., Gamage, H.K., Carroll, B.J., Schenk, P.M., Schmidt, S., 2008. Plants canuse protein as a nitrogen source without assistance from other organisms.Proceedings of the National Academy of Science USA 105, 4524e4529.

Payne, J.W.,1980. Transport and utilization of peptides by bacteria. In: Payne, J.W. (Ed.),MicroorganismsandNitrogenSources. JohnWiley&Sons,Chichester,pp.211e256.

Persson, J., Näsholm, T., 2001. Amino acid uptake, a widespread ability amongboreal forest plants. Ecology Letters 4, 434e438.

Quevauviller, P., 1998. Method Performance Studies for Speciation Analysis. TheRoyal Society of Chemistry, Cambridge, UK.

Raab, T.K., Lipson, D.A., Monson, R.K., 1999. Soil amino acid utilization amongspecies of the Cyperaceae: plant and soil processes. Ecology 80, 2408e2419.

Roberts, P., Jones, D.L., 2008. Critical evaluation of methods for determining totalprotein in soil solution. Soil Biology & Biochemistry 40, 1485e1495.

Roberts, P., Bol, R., Jones, D.L., 2007. Free amino sugar reactions in soil in relation tosoil carbon and nitrogen cycling. Soil Biology & Biochemistry 39, 3081e3092.

Rodwell, J.S., 1992. British Plant Communities, Volume 3 Grasslands and MontaneCommunities. Cambridge University Press, Cambridge, UK.

Rousk, J., Jones, D.L., 2010. Loss of low molecular weight dissolved organic carbon(DOC) and nitrogen (DON) in H2O and 0.5 M K2SO4 soil extracts. Soil Biology &Biochemistry 42, 2331e2335.

Rowell, D.L., 1994. Soil Science Methods & Applications. Pearson Education Ltd.,Harlow, UK.

Rudeforth, C.C., Hartnup, R., Lea, J.W., Thompson, T.R.E., Wright, P.S., 1984. Soils andTheir Use in Wales. Soil Survey of England and Wales, Harpenden, UK.

Schimel, J.P., Bennett, J., 2004. Nitrogen mineralization: challenges of a changingparadigm. Ecology 85, 591e602.

Schimel, J.P., Weintraub, M.N., 2003. The implications of exoenzyme activity onmicrobial carbon and nitrogen limitation in soil: a theoretical model. SoilBiology & Biochemistry 35, 549e563.

Smolander, A., Kitunen, V., 2002. Soil microbial activities and characteristics ofdissolved organic C and N in relation to tree species. Soil Biology & Biochem-istry 34, 651e660.

Streeter, T.C., Bol, R., Bardgett, R.D., 2000. Amino acids as a nitrogen source intemperate upland grasslands: the use of dual labelled (13C, 15N) glycine to testfor direct uptake by dominant grasses. Rapid Communications in Mass Spec-trometry 14, 1351e1355.

Swain, T., Hillis, W.E., 1959. The phenolic constituents of Prunus domestica. I. Thequantitative analysis of phenolic constituents. Journal of the Science of Foodand Agriculture 10, 63e68.

Townend, J., 2002. Practical Statistics for Environmental and Biological Scientists.Wiley, Chichester, UK.

van Hees, P.A.W., Jones, D.L., Finlay, R., Godbold, D.L., Lundström, U.S., 2005. Thecarbon we do not see e the impact of low molecular weight compounds oncarbon dynamics and respiration in forest soils: a review. Soil Biology &Biochemistry 37, 1e13.

Vile, D., Shipley, B., Garnier, E., 2006. Ecosystem productivity can be predicted frompotential relative growth rate and species abundance. Ecology Letters 9,1061e1067.

Voroney, R.P., Brookes, P.C., Beyaert, R.P., 2008. Soil microbial biomass C, N, P, and S.In: Carter, M.R., Gregorich, E.G. (Eds.), Soil Sampling and Methods of Analysis,second ed.. CRC Press, Boca Raton, FL (Chapter 049).

Warren, C.R., Taranto, M., 2010. Temporal variation in pools of amino acids, inor-ganic and microbial N in a temperate grassland soil. Soil Biology & Biochemistry42, 353e359.

Weigelt, A., King, R., Bol, R., Bardgett, R.D., 2003. Inter-specific variability in organicnitrogen uptake of three temperate grassland species. Journal of Plant Nutritionand Soil Science 166, 606e611.

Weigelt, A., Bol, R., Bardgett, R.D., 2005. Preferential uptake of soil nitrogen forms bygrassland plant species. Oecologia 142, 627e635.

Weintraub, M.N., Schimel, J.P., 2005. The seasonal dynamics of amino acids and othernutrients in the Alaskan Arctic tundra soils. Biogeochemistry 73, 359e380.

Weiss, M.S., Abele, U., Weckesser, J., Welte, W., Schiltz, E., Schulz, G.E., 1991.Molecular architecture and electrostatic properties of bacterial porin. Science254, 1627e1630.

Werdin-Pfisterer, N.R., Kielland, K., Boone, R.D., 2009. Soil amino acid compositionacross a boreal forest successional sequence. Soil Biology & Biochemistry 41,1210e1220.

Yu, Z., Zhang, Q., Kraus, T.E.C., Dahlgren, R.A., Anastasio, C., Zasoski, R.J., 2002.Contribution of amino acid compounds to dissolved organic nitrogen in forestsoils. Biogeochemistry 61, 173e198.