substrate supply, ﬁne roots, and temperature control ... · capabilities of their fungal...

Ecology, 92(4), 2011, pp. 892–902� 2011 by the Ecological Society of America

Substrate supply, fine roots, and temperature control proteolyticenzyme activity in temperate forest soils

EDWARD R. BRZOSTEK1

AND ADRIEN C. FINZI

Department of Biology, Boston University, Boston, Massachusetts 02215 USA

Abstract. Temperature and substrate availability constrain the activity of the extracellularenzymes that decompose and release nutrients from soil organic matter (SOM). Proteolyticenzymes are the primary class of enzymes involved in the depolymerization of nitrogen (N)from proteinaceous components of SOM, and their activity affects the rate of N cycling inforest soils. The objectives of this study were to determine whether and how temperature andsubstrate availability affect the activity of proteolytic enzymes in temperate forest soils, andwhether the activity of proteolytic enzymes and other enzymes involved in the acquisition of N(i.e., chitinolytic and ligninolytic enzymes) differs between trees species that form associationswith either ectomycorrhizal or arbuscular mycorrhizal fungi. Temperature limitation ofproteolytic enzyme activity was observed only early in the growing season when soiltemperatures in the field were near 48C. Substrate limitation to proteolytic activity persistedwell into the growing season. Ligninolytic enzyme activity was higher in soils dominated byectomycorrhizal associated tree species. In contrast, the activity of proteolytic and chitinolyticenzymes did not differ, but there were differences between mycorrhizal association in thecontrol of roots on enzyme activity. Roots of ectomycorrhizal species but not those ofarbuscular mycorrhizal species exerted significant control over proteolytic, chitinolytic, andligninolytic enzyme activity; the absence of ectomycorrhizal fine roots reduced the activity ofall three enzymes. These results suggest that climate warming in the absence of increases insubstrate availability may have a modest effect on soil-N cycling, and that global changes thatalter belowground carbon allocation by trees are likely to have a larger effect on nitrogencycling in stands dominated by ectomycorrhizal fungi.

Key words: amino acids; microbial acclimation; mycorrhizal fungi; organic N cycling; proteolyticenzymes; temperate forests; temperature sensitivity of SOM decomposition.

INTRODUCTION

Nitrogen (N) limitation is widespread in the terrestrial

biosphere (e.g., Vitousek and Howarth 1991, LeBauer

and Treseder 2008). There is a rich and deep history of

research on the terrestrial N cycle (e.g., Nadelhoffer et

al. 1985, Aber et al. 1990, Reich et al. 1997), yet there

remain significant uncertainties in our understanding of

the processes that control N availability, particularly

those that regulate the depolymerization of N from soil

organic matter (SOM). Proteolytic enzymes are an

important class of enzymes involved in the breakdown

of protein, a large pool of organic N in SOM, into

amino acids that can be used as a source of N by plants

and microbes (Barraclough 1997, Schulten and Schnit-

zer 1998, Finzi and Berthrong 2005, Gallet-Budynek et

al. 2009). The abundance and activity of proteolytic

enzymes is a principal driver of the within-system cycle

of soil N (Schimel and Bennett 2004), yet the

mechanisms controlling the production and activity of

proteolytic enzymes in most ecosystems remain largely

unknown.

Following a simple model of enzyme kinetics (Fig. 1),

the activity of proteolytic enzymes in soils should be a

function of the interaction between three parameters:

soil temperature, substrate concentration, and the

enzyme pool size (Schimel and Weintraub 2003).

Accordingly, factors that alter some combination of

these three kinetic parameters should alter proteolytic

enzyme activity and the availability of N in soil.

The most immediate effect of an increase in soil

temperature on enzyme activity is to speed up the rate at

which enzymes collide with and breakdown substrates.

Over longer times scales however (e.g., growing season

to years), the environmental conditions under which

enzymes are produced and become active can override

short-term temperature sensitivities. For example, ex-

tracellular enzymes produced in cold soils tend to exhibit

greater temperature sensitivity than those produced in

warm soils because they reach their peak activity at

lower temperatures (Fenner et al. 2005, Koch et al.

2007). By contrast, the lower temperature sensitivity of

enzymes produced under warm conditions has been

linked to a decline in enzyme flexibility that reduces their

Manuscript received 14 September 2010; accepted 12October 2010. Corresponding Editor: J. B. Yavitt.

1 E-mail: [email protected]

892

reaction rate with substrates (Siddiqui and Cavicchioli2006, Bradford et al. 2010).

Despite the abundance of protein in soil organicmatter, substrate limitation of proteolytic enzyme

activity is common in forest soils (Schulten and

Schnitzer 1998, Berthrong and Finzi 2006). Substratelimitation arises because of the physical protection of

protein substrate in soil aggregates and the binding ofproteins by organic molecules (Sollins et al. 2006, Rillig

et al. 2007, McCarthy et al. 2008). Substrate limitation islikely to affect the temperature sensitivity of proteolytic

enzymes (c.f., Davidson and Janssens 2006), though this

interaction has not been widely studied.Evidence of low enzyme activity despite optimum

temperature and substrate conditions suggests that thestanding pool size of enzymes may also be an important

limit on enzyme activity (Wallenstein et al. 2009). Theproduction of enzymes can be limited by the availability

of carbon (C) and N for enzyme synthesis (Allison et al.

2009, Wallenstein et al. 2009) and microbial demandsfor the products of enzyme activity (Craine et al. 2007,

Geisseler and Horwath 2008). Furthermore, onceenzymes are produced, they can be rendered inactive

by their binding to soil organic matter (Joanisse et al.2007), though sorption reactions to clay minerals may

actually stabilize and protect enzymes from degradationby preventing them from becoming substrates them-

selves (Allison 2006).

Temperate forest tree species of the northeastern U.S.differ in many of the factors that control the abundance

and activity of proteolytic enzymes. These species differin litter chemistry, affecting the availability of C and N

for microbial growth, as well as in SOM chemistry (e.g.,

polyphenolic compounds) which leads to variation in

protein binding by organic materials in the soil (Kraus etal. 2003, Talbot and Finzi 2008). Temperate forest trees

also differ in mycorrhizal association. Most tree speciesin the northeast are colonized by ectomycorrhizal

(ECM) fungi, which possess broad enzymatic capabil-

ities allowing for the decomposition of labile andrecalcitrant components of SOM and the transfer of N

to host plants (Chalot and Brun 1998, Talbot et al.2008). A smaller number of species are colonized by

arbuscular mycorrhizal (AM) fungi, which do notappear to have as broad an enzymatic capability as

ECM fungi (but see Hodge et al. 2001). In addition todifferences in enzymatic capability, rates of root

exudation in ECM tree species are higher than AM tree

species (Smith 1976, Phillips and Fahey 2005) stimulat-ing faster rates of C and N mineralization in the

rhizosphere of ECM trees (Phillips and Fahey 2006).Higher rates of SOM decomposition must be mediated

by greater enzyme production and activity in therhizosphere, suggesting that ECM tree roots may

provide soil microbes with at least some of the resources

required in the synthesis of extracellular enzymes.However, the effect of tree roots per se has not been

well established.Proteolytic enzymes are not the only enzyme system

involved in the release of N from SOM. Chitinolytic

enzymes which release amino sugars and ligninolyticenzymes which can release N from recalcitrant sources

may also have a substantial role in mobilizing N(Sinsabaugh 2010). Like proteolytic enzymes, chitino-

lytic and ligninolytic enzyme activities may be controlledby differences between tree species in the enzymatic

capabilities of their fungal symbionts and rates of root

exudation. Further, the greater recalcitrance of SOM in

FIG. 1. Conceptual model of proteolytic enzyme activity in soil. Activity is a function of the interaction among three kineticparameters: temperature, substrate availability, and enzyme pool size. Increases or decreases in kinetic parameters due to variationin the biological, physical, and chemical properties of the soil environment lead to subsequent increases or decreases in proteolyticenzyme activity.

April 2011 893LIMITS TO PROTEOLYTIC ENZYME ACTIVITY

ECM compared to AM soils (Finzi et al. 1998) is likely

to require a broader suite of enzymes to mobilize N from

SOM, including chitinolytic and ligninolytic enzymes

(Allison and Vitousek 2005).

The objective of this study was to investigate the

impact of seasonal changes in temperature and substrate

availability, as well as the presence of tree roots on the

activity of proteolytic enzymes in temperate forest soils.

Because the mobilization of N from SOM requires

multiple enzyme systems, we also examined the response

of chitinolytic and ligninolytic enzyme activities. We

tested two hypotheses: (1) proteolytic enzyme activity is

temperature limited in cold soils and substrate limited in

warm soils, and (2) tree roots promote proteolytic,

chitinolytic, and ligninolytic enzyme activity to a greater

degree in ECM soils than in AM soils.

METHODS

Site and plot description

This research was conducted at two sites, one located

at the Harvard Forest (HF, 42.58 N, 72.188 W) in

Petersham, Massachusetts, USA and the other site

located at the Pisgah State Forest (PSF, 42.878 N,

72.458 W) in Chesterfield, New Hampshire, USA. The

sites have similar land use history and stand age (Foster

1988, 1992). Soils at both sites are inceptisols classified

as Typic Dystrochrepts derived from glacial till overly-

ing granite–schist–gneiss bedrock (USDA Natural

Resources Conservation Service, Web Soil Survey, data

available online).2

Experimental plots dominated by one of four target

tree species were established at each site. Stands of sugar

maple (Acer saccharum) and American beech (Fagus

grandifolia) were located in the PSF. Stands of Eastern

hemlock (Tsuga canadensis) and white ash (Fraxinus

americana) were located in the HF. At each site we

located six replicate, 8 m radius, monodominant plots

that were based on the following criteria: (1) .80% of

the standing basal area was composed of the target tree

species, (2) the fresh litterfall layer was dominated by the

target species, and (3) the core 5 m of the plot containedonly the target tree species (Lovett et al. 2004).

Significant organic horizons were present only in thebeech and hemlock plots. The lack of organic horizon in

the white ash plots may be due to the presence of

earthworms, which are not present in the other plots.Upon plot establishment, we measured background

characteristics of each plot (Table 1). All trees withdiameter at breast height (dbh) .5 cm were measured

and identified to species. Litterfall from each plot was

collected in October and November of 2007. Litterfallwas sorted to species, dried at 608C, and weighed. The

target tree species litter from each plot was ground andanalyzed for %C and %N on an elemental analyzer

(NC2500; CE Elantech, Lakewood, New Jersey, USA).

Soils collected from the top 15 cm of the mineral soilhorizon from each plot and the organic horizon from the

hemlock and beech plots were analyzed for %C and %N,and pH.

To meet our research objectives, we performed

seasonal monitoring of soil enzyme activities in con-junction with two experiments. The first experiment

examined the response of proteolytic enzyme activity to

manipulations of temperature and substrate availabilityat three key seasonal time points. The second experiment

investigated the role of roots in promoting enzymeactivities by using in-growth treatments to either include

or exclude fine roots in the field.

Soil sampling protocol

Soil samples for the first experiment examiningtemperature and substrate limitation to proteolytic

enzyme activity were collected three times in April,

June, and August of 2008. For the seasonal monitoringof enzyme activity, we collected soils in April, June, and

August of 2008 and 2009. We choose the April, June,and August time points because they span the majority

of the growing season including leaf out, peak photo-

synthesis, and leaf area index, and the start of the

TABLE 1. Litterfall characteristics, soil C:N ratios, amino acid concentrations, and inorganic N pools and cycling rates in themineral soil and organic horizons for each tree species.

Common name Latin name

LitterfallC:Nratio

Bulklitterfall(g/m2)

Targetlitter(%) Soil pH

Soil C:Nratio

Amino acidconcentration(lg AA-N/g)

Mineral soil

Sugar maple Acer sacharrum 42.64a (1.24) 427.8a (19.7) 77 4.80a (0.05)a 12.32a (0.69) 1.49a (0.20)White ash Fraxinus americana 44.70a (1.74) 377.2a (19.1) 56 5.10b (0.10)b 13.35a (0.42) 1.80a (0.25)Eastern hemlock Tsuga canadensis 65.05b (1.85) 206.9b (15.8) 67 3.88c (0.08)c 25.66b (0.94) 2.26a (0.51)American beech Fagus grandifolia 52.02c (1.29) 273.5c (4.9) 56 4.44d (0.07)d 19.97c (1.17) 1.93a (0.24)

Organic horizon

Eastern hemlock Tsuga canadensis 3.49a (0.01) 29.91a (1.98) 19.54a (1.82)American beech Fagus grandifolia 3.87b (0.11) 23.86b (0.94) 26.07b (1.63)

Notes: Each value is the mean (with SE in parentheses). Soil N cycling rates and pools are the means of samples taken in April,June, and August of 2008. Superscript letters within a column denote significant differences among species for a given horizon at P, 0.05. Target litter (%) is the mean percentage for each species of the total bulk litterfall mass that is composed of that species’litter. Abbreviations: net N min., net nitrogen mineralization; net nit., net nitrification; AA N, amino acid nitrogen.

2 hhttp://websoilsurvey.nrcs.usda.gov/i

EDWARD R. BRZOSTEK AND ADRIEN C. FINZI894 Ecology, Vol. 92, No. 4

seasonal decline in C uptake at the Harvard Forest,

respectively (Urbanski et al. 2007). At each sampling

point, we collected a 20320 cm sample of the Oe and Oa

horizon in the hemlock and beech plots. Directly

beneath this location we removed the top 15 cm of

mineral soil using a 5 cm diameter soil bulk-density

sampler in each plot. After collection fine roots were

removed from the soils, which were then subsequently

sieved through a 2-mm mesh for mineral soils and an 8-

mm mesh for organic horizons. Subsamples of each soil

were immediately frozen at �808C for later assays of

extracellular enzyme activity. Initial pools of amino acid

and inorganic N were extracted within 18 hours of

sample collection. Experimental manipulations and

assays of proteolytic rates began within 48 hours of

sample collection.

Seasonal monitoring

The 2 mol/L KCl extractable pool sizes of amino

acids, NH4þ and NO3

� were determined for each soil

sample for every sampling date in 2008. Ten and 30 g of

organic horizon and mineral soil sample, respectively,

were extracted in 100 mL of 2 mol/L KCl. In addition,

rates of net N mineralization and net nitrification were

measured by quantifying the change in the 2 mol/L KCl

extractable pool sizes of NH4þ and NO3

� after a 28-day

soil incubation in the lab (Finzi et al. 1998). The

concentration of amino acids for all samples was

quantified using the o-phthaldialdehyde and b-mercap-

toethanol (OPAME) method (Jones et al. 2002).

Concentrations of amino acid N were determined by

comparing the fluorescence of the samples relative to a

standard curve composed of glycine. The concentration

of NH4þ and NO3

� for all samples was quantified using

a flow injection autoanalyzer (Lachat Quickchem 8000;

Hach Company, Loveland, Colorado, USA).

For every soil at each sample date in 2008 and 2009,

we also assayed the potential activities of the chitinolytic

enzyme, n-acetyl-glucosaminidase (NAG), and the

ligninolytic enzyme, phenol oxidase. All the assays were

run using a pH 5.0 sodium acetate buffer at 238C. NAG

activity was determined using a fluorometric microplate

assay, while phenol oxidase activity was determined

using a colorometric microplate assay (Saiya-Cork et al.

2002).

Temperature and substrate limitation

to proteolytic activity

In 2008, we performed an experiment to determine

how temperature and substrate availability affect

proteolytic enzyme activity throughout the growing

season. In this factorial experiment, proteolytic rates

were assayed at two temperatures (field temperature at

time of sampling and 238C) and two substrate levels

(ambient or elevated protein in the form of casein) in

soils from each plot and both soil horizons (n ¼ 144).

This experiment was performed for every sample date in

2008. The temperature of the soil in the field in April,

June, and August were 48C, 158C, and 188C, respectively

(K. Savage, personal communication). Within 12 hours

of soil sample collection, four replicate 2–3 g subsamples

of each soil were placed in their proper incubation

temperature and allowed to equilibrate in closed 50-mL

centrifuge tubes for 12 hours. We choose 12 hours as an

equilibration time to minimize the time between the

assay and the field sampling and to ensure that all of the

samples were at the correct assay temperature.

After equilibration, we assayed proteolysis following a

method modified from Watanabe and Hayano (1995)

and Lipson et al. (1999). To perform the assay under

ambient protein conditions, initial and incubated

subsamples received 10 mL of a 0.5-mmol/L sodium

acetate buffer (pH 5.0) with a small volume of toluene

(400 lL) added to inhibit microbial uptake. The assay of

proteolysis under elevated conditions was performed

similarly, except the sodium acetate buffer contained

0.6% casein. After the reagent addition, the initial

samples were immediately terminated and the incubated

subsamples were placed back into their respective assay

temperature for 4 hours. We choose 4 hours as an

incubation time because previous work has shown

proteolytic enzyme activity to be linear over this period

in temperate forest soils (Finzi and Berthrong 2005,

Rothstein 2009). Enzyme activity in all the initial and

incubated subsamples was terminated through the

addition of 3 mL of a trichloroacetic acid solution.

Proteolytic rates for each experimental treatment were

calculated from the difference between amino acid

concentrations in the incubated and initial subsamples

of each treatment assayed using the OPAME method

described above (Jones et al. 2002).

We assayed the activity of proteolytic enzymes as a

class rather than specific proteolytic enzymes (e.g.,

leucine-aminopeptidase, glycine-aminopeptidase) be-

cause the Watanabe and Hayano (1995) method allowed

us to measure proteolytic rates under ambient and

elevated protein substrate conditions. Assays for indi-

vidual enzymes are conducted under saturating substrate

conditions and therefore could not be used to address

the role of substrate limitation in the depolymerization

of N (Saiya-Cork et al. 2002). In addition, this method

TABLE 1. Extended.

Net N min.(lg N�g�1�d�1)

Net N(lg N�g�1�d�1)

Inorganic N(lg N/g)

0.70a (0.10) 0.60a (0.12) 3.93a 0.29)0.61a (0.13) 0.57a (0.13) 4.41a (0.41)0.10b (0.04) 0.02b (0.02) 2.12b (0.22)0.16b (0.05) 0.05b (0.04) 1.91b (0.14)

3.50a (1.11) �0.05a (0.03) 15.15a (1.30)7.49b (1.66) �0.07a (0.03) 18.14b (2.47)


integrates the activity of a larger suite of proteolytic

enzymes in soils, which may have more relevance

ecologically than more specific enzyme assays. We

acknowledge the loss of plant–microbe interactions

and soil disturbance as methodological artifacts when

employing this assay of proteolytic enzyme activity (c.f.,

Hofmockel et al. 2010). However, these assumptions are

inherent to any assay of soil enzyme activity (Wallen-

stein and Weintraub 2008).

In-growth experiment

We conducted a growing season long field experiment

to investigate the role of roots on proteolysis and

extracellular enzyme activity. This experiment used

mineral soil in-growth cores and organic horizon in-

growth bags (see Plate 1). Organic horizon bags (5 3 10

3 3 cm; Wallander et al. 2001) and PVC core frames for

mineral soil (5 cm diameter 3 15 cm deep) were

constructed using either 1-mm fiberglass mesh or 50-

lm nylon mesh (Sefar Industries, Depew, New York,

USA; Langley et al. 2006). The 50-lm nylon mesh

excludes roots but allows fungal hyphae and heterotro-

phic microbes to explore the soil (herein ‘‘microbes’’),

while the 1mm mesh allows both roots and fungal

hyphae ingrowth (‘‘rootsþmicrobes’’).

The in-growth cores were filled with soil collected

from each plot in April of 2008. The organic and mineral

soil horizons were sieved as previously described. From

these samples we created a single, bulk sample of organic

or mineral soil horizon for each species. The homoge-

nized soil was then placed back into the in-growth

treatments to approximate bulk density (Hendricks et al.

2006).

In May of 2008, one core from each treatment was

placed into each plot and soil horizon (n ¼ 48 mineral

soil cores; n¼ 24 organic horizon bags). In-growth bags

were installed at the interface between the organic

horizon and mineral soil horizon layers. In-growth cores

were installed into the soil by first removing the organic

horizon, then placing the core in the top 15 cm of the

mineral soil horizon and finally replacing the organic

horizon. The cores/bags were harvested during the first

week of October 2008 and then assayed for proteolytic,

NAG, and phenol oxidase enzyme activity.

Statistical analysis

We used three-way nested ANOVA to test for the

effects of mycorrhizal association, tree species nested

within mycorrhizal association, and sample date on the

pools and fluxes of amino acid and inorganic N, NAG

activity, and phenol oxidase activity. The temperature3

substrate limitation experiment was also tested by four-

way ANOVA with the effects of temperature, substrate

availability, mycorrhizal association, and tree species

nested within mycorrhizal association as independent

variables. Finally, differences in enzyme activity in the

in-growth experiment were tested by three-way ANOVA

with mycorrhizal association, tree species nested within

mycorrhizal association, and ingrowth treatment as

independent variables.

Data were analyzed in SAS (2003) with the GLM

procedure. Data were assessed for normality and

homogeneity of variance and log-transformed to meet

model assumptions when needed. For the analysis of the

seasonal enzyme activity, when the interaction with year

was not significant, we streamlined the data presentation

by averaging across years. Statistical tests were per-

formed separately by horizon, given the unequal design

and known ecological differences in the role of these

horizons. Tukey’s studentized range test was used for all

post hoc comparisons of mean differences among species

or experimental treatments.

RESULTS

Seasonal monitoring

The pool size and cycling rates of inorganic N were

significantly higher in the plots dominated by the AM-

associated trees, maple and ash, than by the ECM-

associated trees, hemlock and beech (Table 1). Concen-

trations of amino acids and rates of proteolysis did not

differ between tree species or mycorrhizal association

(Table 1). The organic soil horizons in the hemlock and

beech plots had significantly higher pool sizes and

cycling rates of inorganic and organic N than the

mineral soils (Table 1).

Rates of phenol oxidase activity were significantly

higher (P , 0.0001) in the mineral soils of the ECM

trees than in the AM trees (Fig. 2a). NAG activity did

not differ between tree species (Fig. 2b). Seasonally,

both phenol oxidase and NAG activity declined in the

mineral soil as the growing season progressed (Fig. 2).

Phenol oxidase activity was significantly higher in April

and June than in August for the ECM soils (P , 0.05);

whereas in the AM soils the seasonal decline was not

significant. Across all species, NAG activity was

significantly higher in April than in June and August

(P , 0.05). Both enzyme activities were significantly

greater in the organic horizons than in the mineral soils

of the ECM trees (P , 0.0001, Table 2).

Temperature and substrate limitation

to proteolytic activity

Proteolytic enzyme activity responded significantly to

increases in temperature and substrate availability

across the growing season (Fig. 3). In April when the

soils were cold, there was an interactive effect of

increasing temperature and substrate availability on

proteolytic enzyme activity (P , 0.002, Fig. 3a). As the

soils warmed in June, proteolytic enzyme activity was no

longer temperature sensitive but remained substrate

limited (P , 0.0001, Fig. 3b). By August, there was little

effect of temperature or substrate limitation on proteo-

lytic rates in the soil with the exception of a weak

stimulation by temperature (P , 0.03) in white ash (Fig.

3c).


To account for the difference in the amount ofwarming in April, June, and August, we calculated thepercent change in proteolytic rates between the field and

elevated temperature treatments for each 18C increaseunder ambient protein conditions (Fig. 4). In allspecies, the largest increase in proteolytic rates per

18C increase occurred in April, followed by a steepdecline in the temperature sensitivity in June andAugust (Fig. 4). There was no relationship between

proteolytic enzyme activity and soil moisture (R2 ,

0.01; data not shown).

A qualitatively similar pattern of temperature-by-substrate limitation was observed in the organic horizonof the beech and hemlock soils throughout the growing

season (Table 2). In April, there was an interactive effectof temperature and substrate limitation on proteolyticrates (P , 0.05, Table 2). This interaction was not

significant in June though proteolytic rates remainedtemperature (P , 0.003) and substrate limited (P ,

0.008, Table 2). By August, there was no temperature

limitation, but the organic horizons remained substratelimited (P , 0.03, Table 2).

TABLE 2. Mean (6SE) seasonal extracellular enzyme activity and the response of proteolysis to temperature and substratemanipulations in the organic horizon of the beech and hemlock stands.

Species Month

Seasonal monitoring Temperature (T ) and substrate limitation to proteolysis

Phenol oxidase* NAG Field T Field T þ protein 238C 238C þ protein

Eastern hemlock April 1.97a (0.30) 0.43 (0.05) 10.23 (2.43) 13.03 (3.34) 22.10 (1.83) 37.97 (2.37)June 1.62ab (0.33) 0.47 (0.07) 16.09 (1.56) 27.34 (2.35) 29.59 (2.42) 30.77 (2.91)August 1.14b (0.22) 0.47 (0.06) 16.60 (2.30) 18.47 (1.94) 22.12 (3.36) 23.43 (3.74)

American beech April 4.05a (0.45) 0.48 (0.08) 4.66 (1.50) 24.01 (4.48) 26.36 (3.25) 47.19 (5.00)June 3.16ab (0.51) 0.60 (0.09) 17.80 (4.64) 36.53 (8.08) 38.45 (2.47) 53.25 (13.9)August 2.81b (0.34) 0.63 (0.07) 41.29 (4.62) 51.78 (9.35) 45.39 (2.32) 64.71 (8.86)

Notes: Different lowercase letters indicate significant differences (P , 0.05) between months. Asterisks indicate parameters withsignificant differences (P , 0.05) between species. Phenol oxidase and n-acetyl-glucosaminidase (NAG) activity is expressed aslmol�[g dry soil]�1�h�1. Proteolytic data are expressed in units of lg AA-N�[g dry soil]�1�4 h�1, where AA stands for amino acid.Field temperature in April, June, and August 2008 was 48C, 158C, and 188C, respectively. For Temperature and substrate limitationto proteolysis, the column titles denote the temperature (T ) and protein substrate conditions of the experiment. Proteolytic rateswere assayed at two temperatures (Field T at time of sampling or 238C) with or without added protein substrate (þ protein). Field Tin April, June, and August 2008 was 48C, 158C, and 188C, respectively.

FIG. 2. Seasonal variation in mean (6SE) of (a) phenol oxidase and (b) n-acetyl-glucosaminidase (NAG) activity in the top 15cm of mineral soil for each tree species over two growing seasons. Key to abbreviations: AM, arbuscular mycorrhizal fungi; ECM,ectomycorrhizal fungi; NS, not significant.


In-growth experiment

The exclusion of roots from the mineral soil horizonled to a significant reduction in proteolytic, NAG, and

phenol oxidase activity in the two ECM tree species (P

, 0.05, Fig. 5). In the two AM species, the exclusion ofroots had no effect on enzyme activity. The exclusion of

roots in the organic horizon similarly decreased proteo-lytic (P , 0.06), NAG (not significant), and phenol

oxidase activity (P , 0.04, Fig. 5).

DISCUSSION

The similarity in the temperature and substrate

limitation of proteolytic enzymes across species andmycorrhizal associations (Fig. 3), suggests that there are

common kinetic constraints on proteolytic enzymeactivity. Proteolytic enzyme activity was temperature

sensitive only early in the growing season when the soils

were cold. Enzymes produced under cold conditionstend to have lower temperature optima and greater

physical flexibility, which is thought to make earlyenzymes more temperature sensitive than those pro-

duced under warm conditions (Siddiqui and Cavicchioli

2006, Koch et al. 2007). The significant interaction

between temperature and substrate limitation (Fig. 3)

shows that low concentrations of protein substrate also

limit proteolytic enzyme activity early in the growing

season. The persistence of substrate limitation in to June

in the mineral soil and August in the organic horizon

FIG. 3. Mean (6SE) proteolytic enzyme activity in the top 15 cm of mineral soil for each tree species in response to increases intemperature and additions of protein substrate in (a) April, (b) June, and (c) August of 2008. AA-N is amino acid nitrogen.

FIG. 4. Mean (6SE) percentage change in proteolytic ratesbetween the field and elevated temperature treatments for each18C increase under ambient protein conditions in the top 15 cmof mineral soil for each tree species in April, June, and Augustof 2008.


along with evidence that substrate limitation is wide-

spread among biomes (Hofmockel et al. 2010) suggests

that the availability of protein is likely to constrain the

temperature response of a key component of the N cycle

to climate warming (Davidson and Janssens 2006).

Soil microbes alter their activity in response to shifting

nutrient and temperature regimes, and this may explain

why neither temperature nor substrate availability

limited proteolytic activity late in the growing season

(Figs. 3 and 4). Several different examples help make

this case. For example, low N availability led to a

reduction in the synthesis of enzymes by microbes in

warm, arctic soils (Wallenstein et al. 2009). In the same

forests studied here, increasing soil temperature across

the growing season resulted in declines in microbial

respiration as a consequence of thermal or microbial

community adaptation (Bradford et al. 2008). And, field

rates of root exudation have been shown to decline

throughout the growing season (Weintraub et al. 2007,

Phillips et al. 2008), potentially limiting the supply of

resources for enzyme synthesis by soil microbes. The

decline in temperature and substrate limitation was not

related to seasonal changes in soil moisture but coincides

with important phenological changes including the

FIG. 5. Mean (6SE) response of (a) proteolytic, (b) NAG, and (c) phenol oxidase activity to the two in-growth treatments inthe top 15 cm of mineral soil for each tree species, and in the organic horizons of hemlock and beech. The contribution of roots toenzyme activity was calculated using the equation: Roots¼ (activity in the RootsþMicrobes treatment)� (activity in the Microbestreatment). See Methods for additional description of the in-growth cores.


depletion of carbohydrate stores in roots (Kozlowski

1992) and a decrease in root production (Hendrick and

Pregitzer 1996). Consistent with the response of

proteolytic enzymes, we also found declines in NAG

and phenol oxidase activity over the course of the

growing season (Figs. 2 and 3).

Tree roots of ECM species had a large, positive effect

on extracellular enzyme activity. Tree roots of AM

species had little to no effect (Fig. 5). A number of

processes may explain why the presence of roots

promotes enzyme activity in ECM soils. The production

of root exudates and rates of C and N mineralization in

the rhizosphere are higher in ECM than in AM tree

species (Smith 1976, Phillips and Fahey 2005, 2006),

suggesting that ECM tree roots subsidize microbial

enzyme production. Mycorrhizal and non-mycorrhizal

plants are also capable of producing proteolytic enzymes

(Paungfoo-Lonhienne et al. 2008), raising the possibility

that the ECM tree species studied here release more

proteolytic enzymes directly into the soil. There may

also have been vigorous colonization of new, fine roots

by ECM fungi and this may have stimulated enzyme

activity in the ‘‘roots þ microbes’’ in-growth cores.

Finally, the ECM fungal taxa present in these soils may

not possess long distance foraging capability and instead

concentrate their activity near roots (Agerer 2001).

Regardless of the specific mechanism, ECM tree roots

exert important control over enzyme activity in the soil

by modifying microbial activity.

The similarity in the rates of NAG activity in the

mineral soils and organic horizons of the ECM tree

species in the in-growth experiment (Fig. 5) suggests an

experimental artifact of the treatments. This similarity is

in stark contrast to the consistent, order-of-magnitude

greater NAG activity observed in the field in the organic

horizons than in the mineral soils (Table 2, Fig. 2). The

soils that were used in the in-growth cores and bags were

sieved and homogenized by species prior to the field

incubation. This may have resulted in a large flush of

chitin substrate into these soils as a result of severing

fungal hyphae. Alternatively, roots from the organic

horizon may have proliferated into the in-growth cores

and stimulated greater NAG activity. The artifact

associated with the in-growth cores does not, however,

impact our conclusions because (1) they affected only

NAG activity and not other enzymes, and (2) our

conclusions depend on relative not absolute differences

between treatments.

Based on the observation that AM roots exert little

effect over enzyme activity, a non-root process must

support enzyme activity in soils dominated by AM tree

species. Litter inputs by AM tree species in temperate

forests have a narrow C:N ratio, labile chemistry, and

result in narrow C:N ratio soils (Table 1; Finzi et al.

1998, Lovett et al. 2004) that typically have high rates of

C and N mineralization (Ollinger et al. 2002). Further,

AM soils displayed lower activities of phenol oxidase in

the bulk soil than ECM soils (Fig. 2). Thus, it may be

that the decomposition of labile SOM requires fewer

active enzymes or provides the resources required to

support enzyme activity in AM soils. By contrast, the

litter chemistry of ECM trees is generally recalcitrant

leading to slower rates of decomposition and N cycling.

In ECM soils, it appears that ECM-root inputs are

subsidizing the decomposition of more recalcitrant

organic matter, which typically incurs a larger resource

cost (i.e., more enzyme synthesis, Sinsabaugh 2010).

Much of the research on the response of SOM

decomposition to increasing temperatures has focused

on C mineralization and microbial respiration with a

rather limited set of studies examining N cycling

responses (e.g., Rustad et al. 2001, Koch et al. 2007).

This study links some of the processes thought to control

the response of soil C mineralization to increasing

temperature (e.g., thermal acclimation and substrate

depletion; Bradford et al. 2008, 2010) with those that

mobilize N from SOM (e.g., Melillo et al. 2002). We

show that proteolytic enzyme activity is both tempera-

PLATE 1. (Left) In-growth mineral soil cores prior to installation in May 2008, and (right) an in-growth organic horizon bagupon harvest in October 2008. Photo credits: E. R. Brzostek.


ture and substrate limited, with seasonal declines in both

limitations mirroring seasonal declines in microbial

respiration in the same forest (Bradford et al. 2008).

Further, we show an important linkage between

belowground plant-C flux (e.g., plant roots), SOM

decomposition (e.g., phenol oxidase activity) and the

depolymerization of N from SOM (e.g., NAG and

proteolytic enzyme activity) in stands dominated by

ECM trees. Given the widespread occurrence of N

limitation in terrestrial ecosystems and the importance of

N supply to primary production (Vitousek and Howarth

1991), future research should couple the C- and N-cycle

responses to increasing soil temperatures to better

understand temperate forest responses to climate change.

ACKNOWLEDGMENTS

We thank Colin Averill, Joy Cookingham, CatherineAchorn, Verity Salmon, Poliana Lemos, and Branden Riderfor laboratory and field assistance. In addition, we thank theHarvard Forest and the New Hampshire Department ofResources and Economic Development for granting us accessto perform field research. This work was supported by a grantfrom the NSF (DEB-0743564). Also, E. R. Brzostek wassupported by a Northern Forest Scholar fellowship from theNortheastern States Research Cooperative, a joint program ofthe University of Vermont, the University of Maine, and theNorthern Research Station, USDA Forest Service.

LITERATURE CITED

Aber, J. D., J. M. Melillo, and C. A. McClaugherty. 1990.Predicting long-term patterns of mass-loss, nitrogen dynam-ics, and soil organic-matter formation from initial fine litterchemistry in temperate forest ecosystems. Canadian Journalof Botany 68:2201–2208.

Agerer, R. 2001. Exploration types of ectomycorrhizae: aproposal to classify ectomycorrhizal mycelial systems ac-cording to their patterns of differentiation and putativeecological importance. Mycorrhiza 11:107–114.

Allison, S. D. 2006. Soil minerals and humic acids alter enzymestability: implications for ecosystem processes. Biogeochem-istry 81:361–373.

Allison, S. D., D. S. LeBauer, M. R. Ofrecio, R. Reyes, A. M.Ta, and T. M. Tran. 2009. Low levels of nitrogen additionstimulate decomposition by boreal forest fungi. Soil Biologyand Biochemistry 41:293–302.

Allison, S. D., and P. M. Vitousek. 2005. Responses ofextracellular enzymes to simple and complex nutrient inputs.Soil Biology and Biochemistry 37:937–944.

Barraclough, D. 1997. The direct or MIT route for nitrogenimmobilization: a N-15 mirror image study with leucine andglycine. Soil Biology and Biochemistry 29:101–108.

Berthrong, S. T., and A. C. Finzi. 2006. Amino acid cycling inthree cold-temperate forests of the northeastern USA. SoilBiology and Biochemistry 38:861–869.

Bradford, M. A., C. A. Davies, S. D. Frey, T. R. Maddox, J. M.Melillo, J. E. Mohan, J. F. Reynolds, K. K. Treseder, andM. D. Wallenstein. 2008. Thermal adaptation of soilmicrobial respiration to elevated temperature. EcologyLetters 11:1316–1327.

Bradford, M. A., B. W. Watts, and C. A. Davies. 2010.Thermal adaptation of heterotrophic soil respiration inlaboratory microcosms. Global Change Biology 16:1576–1588.

Chalot, M., and A. Brun. 1998. Physiology of organic nitrogenacquisition by ectomycorrhizal fungi and ectomycorrhizas.FEMS Microbiology Reviews 22:21–44.

Craine, J. M., C. Morrow, and N. Fierer. 2007. Microbialnitrogen limitation increases decomposition. Ecology88:2105–2113.

Davidson, E. A., and I. A. Janssens. 2006. Temperaturesensitivity of soil carbon decomposition and feedbacks toclimate change. Nature 440:165–173.

Fenner, N., C. Freeman, and B. Reynolds. 2005. Observationsof a seasonally shifting thermal optimum in peatland carbon-cycling processes; implications for the global carbon cycleand soil enzyme methodologies. Soil Biology and Biochem-istry 37:1814–1821.

Finzi, A. C., and S. T. Berthrong. 2005. The uptake of aminoacids by microbes and trees in three cold-temperate forests.Ecology 86:3345–3353.

Finzi, A. C., N. Van Breemen, and C. D. Canham. 1998.Canopy tree soil interactions within temperate forests: specieseffects on soil carbon and nitrogen. Ecological Applications8:440–446.

Foster, D. R. 1988. Disturbance history, community organiza-tion and vegetation dynamics of the old-growth PisgahForest, southwestern New Hampshire, USA. Journal ofEcology 76:105134.

Foster, D. R. 1992. Land-use history (1730–1990) andvegetation dynamics in Central New England, USA. Journalof Ecology 80:753–772.

Gallet-Budynek, A., E. Brzostek, V. L. Rodgers, J. M. Talbot,S. Hyzy, and A. C. Finzi. 2009. Intact amino acid uptake bynorthern hardwood and conifer trees. Oecologia 160:129–138.

Geisseler, D., and W. R. Horwath. 2008. Regulation ofextracellular protease activity in soil in response to differentsources and concentrations of nitrogen and carbon. SoilBiology and Biochemistry 40:3040–3048.

Hendrick, R. L., and K. S. Pregitzer. 1996. Temporal anddepth-related patterns of fine root dynamics in northernhardwood forests. Journal of Ecology 84:167–176.

Hendricks, J. J., R. J. Mitchell, K. A. Kuehn, S. D. Pecot, andS. E. Sims. 2006. Measuring external mycelia production ofectomycorrhizal fungi in the field: the soil matrix matters.New Phytologist 171:179–186.

Hodge, A., C. D. Campbell, and A. H. Fitter. 2001. Anarbuscular mycorrhizal fungus accelerates decompositionand acquires nitrogen directly from organic material. Nature413:297–299.

Hofmockel, K. S., N. Fierer, B. P. Colman, and R. B. Jackson.2010. Amino acid abundance and proteolytic potential inNorth American soils. Oecologia 163:1069–1078.

Joanisse, G. D., R. L. Bradley, C. M. Preston, and A. D.Munson. 2007. Soil enzyme inhibition by condensed littertannins may drive ecosystem structure and processes: the caseof Kalmia angustifolia. New Phytologist 175:535–546.

Jones, D. L., A. G. Owen, and J. F. Farrar. 2002. Simplemethod to enable the high resolution determination of totalfree amino acids in soil solutions and soil extracts. SoilBiology and Biochemistry 34:1893–1902.

Koch, O., D. Tscherko, and E. Kandeler. 2007. Temperaturesensitivity of microbial respiration, nitrogen mineralization,and potential soil enzyme activities in organic alpine soils.Global Biogeochemical Cycles 21:11.

Kozlowski, T. T. 1992. Carbohydrate sources and sinks inwoody-plants. Botanical Review 58:107–222.

Kraus, T. E. C., R. A. Dahlgren, and R. J. Zasoski. 2003.Tannins in nutrient dynamics of forest ecosystems: a review.Plant and Soil 256:41–66.

Langley, J. A., S. K. Chapman, and B. A. Hungate. 2006.Ectomycorrhizal colonization slows root decomposition: thepost-mortem fungal legacy. Ecology Letters 9:955–959.

LeBauer, D. S., and K. K. Treseder. 2008. Nitrogen limitationof net primary productivity in terrestrial ecosystems isglobally distributed. Ecology 89:371–379.


Lipson, D. A., S. K. Schmidt, and R. K. Monson. 1999. Linksbetween microbial population dynamics and nitrogen avail-ability in an alpine ecosystem. Ecology 80:1623–1631.

Lovett, G. M., K. C. Weathers, M. A. Arthur, and J. C.Schultz. 2004. Nitrogen cycling in a northern hardwoodforest: Do species matter? Biogeochemistry 67:289–308.

McCarthy, J. F., J. Ilavsky, J. D. Jastrow, L. M. Mayer, E.Perfect, and J. Zhuang. 2008. Protection of organic carbon insoil microaggregates via restructuring of aggregate porosityand filling of pores with accumulating organic matter.Geochimica et Cosmochimica Acta 72:4725–4744.

Melillo, J. M., P. A. Steudler, J. D. Aber, K. Newkirk, H. Lux,F. P. Bowles, C. Catricala, A. Magill, T. Ahrens, and S.Morrisseau. 2002. Soil warming and carbon-cycle feedbacksto the climate system. Science 298:2173–2176.

Nadelhoffer, K. J., J. D. Aber, and J. M. Melillo. 1985. Fineroots, net primary production, and soil-nitrogen availability:a new hypothesis. Ecology 66:1377–1390.

Ollinger, S. V., M. L. Smith, M. E. Martin, R. A. Hallett, C. L.Goodale, and J. D. Aber. 2002. Regional variation in foliarchemistry and N cycling among forests of diverse history andcomposition. Ecology 83:339–355.

Paungfoo-Lonhienne, C., T. G. A. Lonhienne, D. Rentsch, N.Robinson, M. Christie, R. I. Webb, H. K. Gamage, B. J.Carroll, P. M. Schenk, and S. Schmidt. 2008. Plants can useprotein as a nitrogen source without assistance from otherorganisms. Proceedings of the National Academy of SciencesUSA 105:4524–4529.

Phillips, R. P., Y. Erlitz, R. Bier, and E. S. Bernhardt. 2008.New approach for capturing soluble root exudates in forestsoils. Functional Ecology 22:990–999.

Phillips, R. P., and T. J. Fahey. 2005. Patterns of rhizospherecarbon flux in sugar maple (Acer saccharum) and yellow birch(Betula allegheniensis) saplings. Global Change Biology11:983–995.

Phillips, R. P., and T. J. Fahey. 2006. Tree species andmycorrhizal associations influence the magnitude of rhizo-sphere effects. Ecology 87:1302–1313.

Reich, P. B., D. F. Grigal, J. D. Aber, and S. T. Gower. 1997.Nitrogen mineralization and productivity in 50 hardwoodand conifer stands on diverse soils. Ecology 78:335–347.

Rillig, M. C., B. A. Caldwell, H. A. B. Wosten, and P. Sollins.2007. Role of proteins in soil carbon and nitrogen storage:controls on persistence. Biogeochemistry 85:25–44.

Rothstein, D. E. 2009. Soil amino-acid availability across atemperate-forest fertility gradient. Biogeochemistry 92:201–215.

Rustad, L. E., J. L. Campbell, G. M. Marion, R. J. Norby,M. J. Mitchell, A. E. Hartley, J. H. C. Cornelissen, and J.Gurevitch. 2001. A meta-analysis of the response of soilrespiration, net nitrogen mineralization, and abovegroundplant growth to experimental ecosystem warming. Oecologia126:543–562.

Saiya-Cork, K. R., R. L. Sinsabaugh, and D. R. Zak. 2002. Theeffects of long term nitrogen deposition on extracellularenzyme activity in an Acer saccharum forest soil. Soil Biologyand Biochemistry 34:1309–1315.

SAS Institute. 2003. SAS version 9. SAS Institute, Cary, NorthCarolina, USA.

Schimel, J. P., and J. Bennett. 2004. Nitrogen mineralization:challenges of a changing paradigm. Ecology 85:591–602.

Schimel, J. P., and M. N. Weintraub. 2003. The implications ofexoenzyme activity on microbial carbon and nitrogenlimitation in soil: a theoretical model. Soil Biology andBiochemistry 35:549–563.

Schulten, H. R., and M. Schnitzer. 1998. The chemistry of soilorganic nitrogen: a review. Biology and Fertility of Soils26:1–15.

Siddiqui, K. S., and R. Cavicchioli. 2006. Cold-adaptedenzymes. Annual Review of Biochemistry 75:403–433.

Sinsabaugh, R. L. 2010. Phenol oxidase, peroxidase andorganic matter dynamics of soil. Soil Biology and Biochem-istry 42:391–404.

Smith, W. H. 1976. Character and significance of forest treeroot exudates. Ecology 57:324–331.

Sollins, P., C. Swanston, M. Kleber, T. Filley, M. Kramer, S.Crow, B. A. Caldwell, K. Lajtha, and R. Bowden. 2006.Organic C and N stabilization in a forest soil: evidence fromsequential density fractionation. Soil Biology and Biochem-istry 38:3313–3324.

Talbot, J. M., S. D. Allison, and K. K. Treseder. 2008.Decomposers in disguise: mycorrhizal fungi as regulators ofsoil C dynamics in ecosystems under global change.Functional Ecology 22:955–963.

Talbot, J. M., and A. C. Finzi. 2008. Differential effects ofsugar maple, red oak, and hemlock tannins on carbon andnitrogen cycling in temperate forest soils. Oecologia 155:583–592.

Urbanski, S., C. Barford, S. Wofsy, C. Kucharik, E. Pyle, J.Budney, K. McKain, D. Fitzjarrald, M. Czikowsky, andJ. W. Munger. 2007. Factors controlling CO2 exchange ontimescales from hourly to decadal at Harvard Forest. Journalof Geophysical Research-Biogeosciences 112:25.

Vitousek, P. M., and R. W. Howarth. 1991. Nitrogen limitationon land and in the sea: how it can occur. Biogeochemistry13:87–115.

Wallander, H., L. O. Nilsson, D. Hagerberg, and E. Baath.2001. Estimation of the biomass and seasonal growth ofexternal mycelium of ectomycorrhizal fungi in the field. NewPhytologist 151:753–760.

Wallenstein, M. D., S. K. McMahon, and J. P. Schimel. 2009.Seasonal variation in enzyme activities and temperaturesensitivities in Arctic tundra soils. Global Change Biology15:1631–1639.

Wallenstein, M. D., and M. N. Weintraub. 2008. Emergingtools for measuring and modeling the in situ activity of soilextracellular enzymes. Soil Biology and Biochemistry40:2098–2106.

Watanabe, K., and K. Hayano. 1995. Seasonal-variation of soilprotease activities and their relation to proteolytic bacteriaand Bacillus spp in paddy field soil. Soil Biology andBiochemistry 27:197–203.

Weintraub, M. N., L. E. Scott-Denton, S. K. Schmidt, andR. K. Monson. 2007. The effects of tree rhizodeposition onsoil exoenzyme activity, dissolved organic carbon, andnutrient availability in a subalpine forest ecosystem. Oeco-logia 154:327–338.


substrate supply, ﬁne roots, and temperature control ... · capabilities of their fungal...

Documents