Warming alters routing of labile and slower-turnover carbon through distinct microbial groups in boreal forest organic soils

at SciVerse ScienceDirect

Soil Biology & Biochemistry 60 (2013) 23e32

Contents lists available

Soil Biology & Biochemistry

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

Warming alters routing of labile and slower-turnover carbon throughdistinct microbial groups in boreal forest organic soils

Susan E. Ziegler a,*,1, Sharon A. Billings b,2, Chad S. Lane a,1,4, Jianwei Li b,2,5, Marilyn L. Fogel c,3

aDepartment of Earth Sciences, Memorial University, 300 Prince Phillip Drive, St. John’s, NL A1B 3X5, CanadabDepartment of Ecology and Evolutionary Biology, Kansas Biological Survey, University of Kansas, Lawrence, KS 66047, USAcGeophysical Laboratory, Carnegie Institution of Washington, 5251 Broad Branch Road NW, Washington, DC 20015-1305, USA

a r t i c l e i n f o

Article history:Received 8 August 2012Received in revised form2 January 2013Accepted 6 January 2013Available online 31 January 2013

Keywords:Soil C cyclingClimate warmingMicrobial substrate useBoreal forestStable carbon isotopesPhospholipid fatty acidsOrganic soils

* Corresponding author.E-mail addresses: [email protected] (S.E. Ziegler), sh

[email protected] (C.S. Lane), [email protected](M.L. Fogel).

1 Tel.: þ1 709 864 2669.2 Tel.: þ1 785 864 1560.3 Tel.: þ1 202 478 8900.4 Currently at the Department of Geography and

Science, University of North Carolina-WilmingtTel.: þ1 910 962 3466.

5 Currently at the Department of Botany and Micrhoma, Norman, OK 73019-0390, USA.

0038-0717/$ e see front matter � 2013 Elsevier Ltd.http://dx.doi.org/10.1016/j.soilbio.2013.01.001

a b s t r a c t

Our understanding of the mechanisms driving the response of soil organic carbon (SOC) pools towarming, though critical for predicting climate feedbacks, remains limited. Here we report results froma warming experiment using O-horizon soils from two mesic, boreal forest sites with contrasting climateregimes. We replaced extant Oi soil horizons, or litterfall C, with another coniferous Oi possessinga distinct d13C signature, and tracked the net incorporation of the replaced Oi and, by difference, Oea-Cinto soil microbial phospholipid fatty acids (PLFA) following 120-day incubations at 15 �C and 20 �C. Wedemonstrate how regional climate (site effects) and experimental warming (temperature effects) in-fluence microbial incorporation of Oi versus slower-turnover Oea SOC pools. Microbial biomass, esti-mated from total PLFA, increased by 32e60% with temperature and was 20e42% higher within soils fromthe warmer versus cooler site, congruent with increased mineralization in those soils. The proportion ofGram-positive bacterial PLFA-C derived from Oi-C more than doubled and coincided with a reduction inthe incorporation of Oi-C into fungal relative to bacterial PLFA with warming and in soils from thewarmer site. Mirroring the relative decrease in fungal incorporation of Oi-C, warming led to an increaseof 22e31% in the proportion of fungal PLFA-C derived from the Oea-C, consistent with the increasedincorporation of this slower-turnover SOC pool in soils from the warmer site. The increase in microbialbiomass and shift in routing of Oi and Oea pools through PLFA indicate that warming preferentiallyincreases fungal mineralization of more slow-turnover C pools in these boreal organic soils. Shifts inmicrobial substrate routing and biomass increases with warming observed here underscore the potentialimportance of changing proportions of microbial biomass remnant contributions to SOC pools withclimate warming.

� 2013 Elsevier Ltd. All rights reserved.

1. Introduction

Climate warming is linked to increased mineralization rates ofsoil organic carbon (SOC; Lloyd and Taylor, 1994; Kirschbaum,1995;Davidson and Janssens, 2006; Sierra et al., 2012) with some studies

[email protected] (S.A. Billings),(J. Li), [email protected]

Geology, Center for Marineon, NC 28403-5928, USA.

obiology, University of Okla-

All rights reserved.

suggesting greater relative increases in mineralization of morecomplex, slow-turnover SOC pools (Hartley and Ineson, 2008;Craine et al., 2010; Janssens and Vicca, 2010; Li et al., 2012). Evensmall increases in mineralization of Earth’s large reservoir of rela-tively slow-turnover SOC would represent a significant positivefeedback to climate warming, a fact that has prompted manystudies of soil warming. Multiple challenges have hampered at-tempts to develop a mechanistic understanding of the influence ofwarming on microbial substrate choice and associated patterns ofmineralization, however. These challenges include: (1) tremendousvariation in the composition and associated reactivity of multipleSOC substrates, (2) potential changes in microbial structure andfunction with warming, and (3) changing physical and environ-mental soil attributes such as organo-mineral interactions, nutrientavailability, and soil moisture with temperature (Davidson andJanssens, 2006; Conant et al., 2011). Some of these features maygovern differences between apparent temperature sensitivities of

S.E. Ziegler et al. / Soil Biology & Biochemistry 60 (2013) 23e3224

decay and those at least qualitatively predicted by enzyme kinetics(Davidson and Janssens, 2006; Bradford and Watts, 2010; Conantet al., 2011), but the mechanisms controlling the varied responseof soil C processing to warming remain unclear.

Any mechanistic study attempting to unravel these mysteriesmust consider the longevity of any observed effects of warming toproject their potential ramification on climate feedbacks (Janzen,2004). In situ warming experiments are useful for quantifyingSOC responses to warming on time scales of years (Oechel et al.,2000; Luo et al., 2001; Rustad et al., 2001; Melillo et al., 2002),but are difficult to extrapolate across more expansive time scalesrelevant for anthropogenic climate change. Differences in theturnover and geochemical attributes of soil organic matter profilesacross natural temperature gradients (Trumbore et al., 1996; Leifeldet al., 2009; Li et al., 2012) highlight the utility of such gradients forunderstandingmore long-term influences of climate on soil organicmatter characteristics. The multiple factors that vary in addition tomean annual temperature (MAT), however, often make it difficultto determine the physical or biological processes responsible forvarying soil composition across climate gradients (Nakatsubo et al.,1997; Steinberger et al., 1999). Coupling investigations of soils fromvarying climate regimes with laboratory investigations of the samesoils can permit insight into the biogeochemical processesresponsible for changes in soil organic matter pools with climatewarming on longer time scales (Fissore et al., 2009;Wu et al., 2009;Schindlbacher et al., 2010).

Consistent with observed variation in soil organic matter com-position with latitude (Cannone et al., 2008), warming can induceproportionally greater increases in exo-enzyme activities asso-ciatedwith the breakdown of relatively slow-turnover SOC (Li et al.,2012). Changes with warming in microbially mediated SOC decaymay also be linked to changes in the relative abundances of themicrobial groups competing for resources liberated upon decay(Zogg et al., 1997; Andrews et al., 2000; Biasi et al., 2005; Zhanget al., 2005; Frey et al., 2008). Such shifts in microbial dynamicswith warming could have significant implications for SOC decay aswell as SOC formation. For example, phenol oxidases are effectivecatalysts for the decay of phenolic components, including thosederived from lignin (Sinsabaugh, 2010). Activities of these andother oxidative enzymes apparently can increase to a greater de-gree than activities of hydrolytic enzymes with soil warming(Wallenstein, 2011; Li et al., 2012). These oxidative enzymes aregenerated primarily by fungi and actinobacteria in soil (Duran et al.,2002; Claus, 2003; Rabinovich et al., 2004; Kirby, 2005). It remainsunclear, however, whether changes in exo-enzyme activities withtemperature are linked to changes in activity of physiologicallydifferent microbial groups relevant to formation of SOC compoundsderived from that community’s biomass. This is despite the grow-ing recognition of the large proportion of natural organic matterpools comprised of microbially-derived compounds (Ogawa et al.,2001; Tremblay and Benner, 2006), including SOC pools(Amelung et al., 2001; Simpson et al., 2007; Potthoff et al., 2008).

A recent study suggests that turnover of biomass derived fromdifferent microbial groups may not vary (Throckmorton et al.,2012). However, other work suggests that different microbialgroups may contribute to the input, composition and turnover ofSOC pools to varying degrees due in part to variation in their cellwall composition (Liang et al., 2008; Strickland and Rousk, 2010;Fernandez, 2011). For example, recent evidence suggests that chi-tin, a polymer of glucosamine andmajor fungal cell wall material, isquite labile (Fernandez, 2011). On the other hand, some sugars,including amino sugars associated with bacterial peptidoglycan,can be preferentially preserved relative to fungal glucosamine insoils (Amelung et al., 2001). To the extent that warming results inaltered biomass production rates of distinct microbial groups, an

important and relatively unstudied response to soil warming maybe shifts in relative abundances of microbially-derived compoundsimportant for soil organic matter formation. Thus, much like theneed to ascertain changes in plant inputs with climate warming(Kirschbaum, 2000; Gu et al., 2004), we need to constrain howmicrobial inputs to SOC poolsmay change. Insight into the potentialfor changes in sources of these inputs directly associated with soilwarming, however, can be gained on shorter time scales byinvestigating the impact of warming on the activity or soil substrateuse by relevant microbial groups.

Herewe explore the influence of warming on routing of two SOCpools with distinct turnover rates through microbial groupsexhibiting different physiologies and biogeochemical functionsfrom two boreal forests with differing climate regimes. Tracinga unique C isotopic signature in replaced Oi sub-horizons added tosoil mesocosms, we investigated the proportion of Oi materialincorporated into bacterial and fungal phospholipid fatty acids(PLFA) during a 120 day warming incubation, versus SOC derivedfrom deeper, slower-turnover Oea sub-horizons. Comparing theseresults with the respiratory and enzymatic data from these sameincubations, as reported in Li et al. (2012), we addressed twoquestions: (1) does experimental warming induce changes insubstrate routing through distinct microbial groups relevant toboth SOC decay and formation? and (2) do changes in SOC routingwith experimental warming match processes and geochemistryobserved in soils native to a warmer forest relative to soils from anotherwise similar, but cooler, forest?

2. Study site description and context

Soil samples were collected from Salmon River (SR) and GrandCodroy regions of the Newfoundland and Labrador Boreal Ecosys-tem Latitudinal Transect (NL-BELT). This transect spans 5.5 degreesof latitude from the southwest corner of the island of Newfound-land to the most northern extent of balsam fir (Abies balsamea (L.)Mill.) dominated forests in central-eastern Labrador and has beendelineated into four distinct regions (for map see Li et al., 2012). TheNL-BELT, part of the Canadian Forest Service’s (CFS) National Net-work of Latitudinal and Elevational Transects (NNLET), is a researchplatform established by the CFS in collaboration with the New-foundland and Labrador Forestry Service to facilitate researchfocused on climate change impacts on boreal forest ecosystems. Allfour regions of the transect consist of forested study sites approx-imately 60e65 years old underlain by soils consisting of relativelythick (w10 cm) organic horizons and moderately well drained Bhorizons characterized as Orthic Humo-Ferric Podzols. The twosites exploited in this study, though similar in forest stand class andelevation (14.0 and 13.1 masl), differ primarily in temperatureregime. For further site characteristics please see Li et al. (2012).

Working within the NL-BELT provides a platform for inves-tigating warming impacts on a globally significant C pool e O ho-rizons e in a region expected to experience 4e7 �C increases intemperature this century (IPCC, 2007). The high moisture exhibitedby these mesic, C-rich forest soils mitigates the influence of sub-strate limitation on SOC warming responses enabling the isolationof temperature effects on microbial processes. Radiocarbon, ele-mental and stable isotopic composition of the soil profiles at thesestudy sites suggest greater microbial processing rates and possiblygreater loss of older and more slow-turnover SOC at the warmersite (Li et al., 2012). Furthermore, warming induced losses of slow-turnover Oea C via respiration and leaching as dissolved organicC to a greater extent than Oi C and also induced a greater relativeincrease in phenol oxidase activity as compared with enzymesassociated with the use of more labile C substrates. Here, weexplore mechanisms likely responsible for the patterns reported in

S.E. Ziegler et al. / Soil Biology & Biochemistry 60 (2013) 23e32 25

Li et al. (2012) by tracking the flow of Oi and Oea C through themicrobial communities in these soils.

3. Methods

3.1. Experimental set up and sampling procedures

Details of soil processing and incubation procedures are repor-ted in Li et al. (2012). Briefly, soil samples were collected from boththe SR and GC sites in July 2009. We established three circular plots(30 m diameter) at each of the two sites with three subplots (10 mdiameter) distributed evenly within each. Within each of the threesubplots we collected intact O-horizons using a rectangular frame(700 cm2). This sampling process generated a total of 18 O horizonsamples from three plots at each of two sites (GC, SR), which wereshipped to the University of Kansas and stored at 4 �C until analysis.

We characterized the O horizon samples by separating theminto three separate subhorizons (Oi, Oe and Oa) based upon color,texture, and root density. Following this separation, sub-horizonswere homogenized separately for elemental and stable isotopeanalyses (for details see Li et al., 2012). The Oeawas recreated usingthe original proportions of the separated Oe and Oa horizons andthe Oi subhorizon material removed from these O-horizon sampleswas replaced with an equivalent mass of loblolly pine (Pinus taedaL.) litterfall grown with 13C-depleted CO2 at the Duke Free AirCarbon Enrichment site (Hendrey et al., 1999; Lichter et al., 2008)exhibiting a d13C of �34.9& (see Li et al., 2012 for details). Westerilized the pine litterfall with a pressure cooker (Wright andJawson, 2001) to ensure that any microbial activity observed wasderived frommicroorganisms native to NL-BELT sites. We recognizethat this sterilization may have caused some release of microbialorganic matter but assume this to be a relatively minor input,particularly given the 120-day incubation time and high microbialactivity and turnover over this time period. The resulting homog-enized Oi was 49.4% C and 0.56% N (molar C:N of 103). Thed13C value of replaced Oi material relative to Oea sub-horizons(d13C of �28.5&, Li et al., 2012) enabled us to trace it into micro-bial PLFA. By replacing native Oi material, we hoped to minimizepriming effects often observed when substrate supplies are aug-mented beyond extant levels (Horvath, 1972; Dalenberg and Jager,1989; Kuzyakov et al., 2000).

We added the 13C-deplete Oi to the mesocosm cores and mixedthe replaced Oi and Oea sub-horizons together on the 1st and 62ndday of the 120 day incubation (details in Li et al., 2012). We refer tothese incubations as the Oiþ Oea treatment. An equivalent numberof cores were also incubated without this added Oi material and arereferred to as the Oea only treatment. The soil cores were broughtto 75% water holding capacity and incubated at 15 �C and 20 �C(uniformity � 0.5 �C; stability � 0.2 �C) in an Isotemp Low Tem-perature incubator (Fisher Scientific; see Li et al., 2012 for furtherdetails). These temperatures represent those commonly observedin situ during the warmest weeks of the growing season at thewarmest site (GC) and are at the highest end of the range observedat SR. We sealed all jars with an airtight lid equipped with a septumand aerated each every two to three days monitoring moisture lossduring this aeration and adding water to maintain moisture asneeded. We destructively subsampled soil on day 120 for PLFAanalyses.

3.2. Phospholipid fatty acids (PLFA) analyses

We performed phospholipid fatty acid (PLFA) analyses onlyophilized subsamples of soil (0.5 gdw). We split one homoge-nized subsample (1.0 gdw) in half and spiked one half witha phospholipid standard (1,2-diheptadecanoyl-sn-glycero-3-

phosphocholine; Avanti Polar Lipids) to determine phospholipidrecovery. We used a modified Bligh-Dyer extraction method toisolate lipids from the soil, followed by solid phase extraction withsilicic acid to isolate the PLFA from the neutral and glycolipids(White and Ringelberg, 1998). We converted the extracted phos-pholipid fatty acids to their associated fatty acid methyl esters(FAMEs) for the gas chromatography analyses using a single-stepsaponification and methylation procedure (Findlay, 2004). Thederivitized FAMEswere than purified using reverse-phase C18 resinsolid phase extraction, and each sample was then spiked witha known quantity of ethyl arachidate and ethyl tetracosanoate asquantification standards. A small aliquot of the methanol used inthe methylation of each sample batch was analyzed for d13C on anAurora 1020 TOC analyzer (O.I. Analytical) coupled to a DeltaV isotope ratio mass spectrometer (Thermo Scientific) and used tocorrect for the change in the fatty acid d13C imparted by the addedmethyl C (Ziegler et al., 2005).

The FAMEs were identified using an Agilent 6890 gas chroma-tograph interfaced with an Agilent 5973 inert mass selective de-tector (Agilent Technologies) in conjunction with known standardsand the NISTmass spectral library and quantified using a Varian CP-3800 gas chromatograph equippedwith a flame ionization detector.The d13C of the FAMEs were determined using an Agilent 6890 gaschromatograph interfaced with a Finnigan MAT 252 (MemorialUniversity) or, for some samples, a Trace gas chromatograph inter-faced to a Thermo Finnigan Delta XL (Geophysical Laboratory, Car-negie Institution of Washington) stable isotope mass spectrometer.Both instruments are interfaced with a gas chromatographecom-bustion (GC/C) III (Thermo Scientific).We assessed the precision andaccuracy of d13C PLFA analyses on each instrument by repeated in-jections (�2 per sample) and comparison to a commercially avail-able bacterial acid methyl ester standard mixture (ester mixture“F8”) supplied by Indiana University. Additionally, we conductedoverlapping sample analyses to confirm consistency of resultsbetween the two gas chromatographecombustioneIRMS systems.We conducted all gas chromatographic analyses using the sameoperating parameters, capillary column (BPX-70; 70% cyanopropylpolysilphenylene-siloxane; 50 m, 0.32 mm internal diameter,0.32 mm film thickness; SGE Analytical Science), and each gaschromatograph employed a split/splitless injector operated insplitless mode.

We employed PLFA analysis to track relative differences in theactivities of broadly separated functional groups relevant to soilformation and soil organic matter fate. Specifically, we quantifiedOi-C incorporation into bacterial, Gram-positive bacterial, Gram-negative bacterial, actinobacterial, and fungal PLFA. With theexception of 18:1u9t and 18:1u9c, weused the terminally branchedPLFA and monounsaturated PLFA to investigate relative abundanceand Oi-C incorporation into Gram positive and Gram negative or-ganisms, respectively (Ringelberg et al., 1989; Zelles et al., 1992;White et al.,1996).We used 10Me18:0 and 18:2u6 to investigate theroles of actinobacteria and fungi, respectively (Federle et al., 1986;Frostegard and Baath, 1996). Bacterial PLFA were determined fromthe sum of the Gram positive and Gram negative PLFA and used togenerate a relative estimate of the ratio of fungal to bacterial bio-mass (F:B) in these soils by dividing the concentration of 18:2u6 bythe sum of bacterial PLFA. Given their presence in a wide array ofmicrobial groups, the PLFA 18:1u9t and 18:1u9c and were eval-uated separately, grouped as nonspecific along with 16:0, and notused to assess the biomass or substrate incorporation for any par-ticular microbial group. We did, however, use these data to test forassociation with 18:2u6 to determine specificity for fungi in thesesoils in order to evaluate our conservative approach. The PLFA18:1u9t is sometimes cited, along with 18:1u9c, as a fungal PLFA(e.g. Butler et al., 2011) given their variation with 18:2u6 in many

S.E. Ziegler et al. / Soil Biology & Biochemistry 60 (2013) 23e3226

soils (Frostegard et al., 2011). However, these PLFA are not exclusiveto fungi in all environments (Frostegard et al., 1993; Frostegard andBaath, 1996; Olsson, 1999); we therefore did not define them asfungal PLFA. We used the sum of all polyunsaturated PLFA to assesseukaryotic PLFA, and 20:4u6 to assess protozoan PLFA (White et al.,1996). Lastly, we assessed the sum of all PLFA to estimate relativechanges inmicrobial biomass and substrate C incorporation into themicrobial community as a whole.

Employing both the quantity and C isotope composition of PLFA,we calculated the fraction of total Oi derived C (Oi-C) incorporatedinto individual PLFA. This measure enabled us to determine if therewere any shifts in the net incorporation of litterfall C acrossmicrobial groups, and to assess potential changes in the composi-tion of microorganisms using Oi-C in these soils with elevatedtemperature and climate regime. We used the d13C of the PLFA inthe Oiþ Oea and the Oea only soils to determine the quantity of Oi-C incorporated, and by difference, the Oea-C incorporated intothese PLFA at day 120. The fraction of PLFA-C derived from theadded Oi-C was determined using the following equation:

FPLFAOi�C ¼�d13CPLFAOi�Oea � d13CPLFAOea�only

��d13COi � d13COea

� (1)

where d13CPLFAOi þ Oea is the d13C of individual PLFA from soilsreceiving Oi replacement (Oi þ Oea), d13CPLFAOea-only is the d13C ofindividual PLFA from soils receiving no Oi material (Oea only),d13COi is the d13C of the Oi replacement, and d13COea is the d13C of theOea material. This approach assumes that the functioning of themicrobial communities in both soil treatments (Oi-only andOi þ Oea) was similar, a feature we test as described below. Weestimated the absolute quantity of Oi-C incorporated into individ-ual, isotopically resolved PLFA using:

Absolute Oi� C incorporated ¼ FPLFAOi�C*½PLFA� (2)

where absolute Oi-C is presented as mg Oi-C g SOC�1 and [PLFA] asmg PLFA-C g SOC�1.

In addition to calculating absolute Oi-C incorporated into totalPLFA and individual groups of PLFA as described above, we alsocalculated the ratio of absolute Oi-C incorporated into fungal PLFArelative to Oi-C incorporated into bacterial PLFA to assess the rela-tive use of Oi-C by these groups. Throughout the manuscript werefer to this ratio as the proportion of Oi-C routed through fungalrelative to bacteria PLFA. Further, we used Eqs. (1) and (2) to esti-mate two additional parameters depicting substrate flow throughthese soils’microbial communities. First, to assess the proportion ofthe total Oi-C incorporated into a single or group of PLFA,wedividedthe absolute Oi-C incorporated into an individual PLFA by the sumofthe absolute Oi-C incorporated in all PLFA resolved. These individualvalues were than summed for all individual PLFA relevant to a par-ticular group to obtain the fraction of Oi-C incorporated into thetotal PLFA pool that could be attributed to bacterial, Gram negativebacterial, Gram positive bacterial, and fungal components of thecommunity. Second, to determine the proportion of Oi relative toOea-C incorporated into individual or groups of PLFA (mg Oi-C mgPLFA-C�1), we divided the absolute Oi-C incorporated into a PLFA orgroup of PLFA by the total quantity of that PLFA or group.

Our interpretation of these calculations requires a couple ofassumptions: 1) all the PLFA turned over completely by 120 daysand 2) within the Oiþ Oea soils the d13C of PLFA-C derived from theOea-C was the same as that found in the parallel incubations of theOea-only soil. Thus, we assume that the added Oi-C did not alter theOea-C pool accessed by the soil microbes or change the relationshipbetween d13C of the Oea-C and the d13C-PLFA associated with the

use of that pool. The first assumption has been readily investigatedas part of the establishment of the use of PLFA for studying livingmicrobial biomass in sediments and soils where turnover of PLFAhas been found to be relatively rapid in aerobic environments (halftime 2e10 days; White et al., 1979). It is almost certain that soilPLFA turned over several times during this incubation, therefore, itis important to point out that these estimates of Oi-C incorporationrepresent the overall net incorporation of this C pool integratingboth direct and indirect uptake of the Oi-C. This is particularlyimportant in the case of microheterotrophic eukaryotes or proto-zoan PLFA where increased turnover might mean a greater totalOi-C incorporation into these PLFA. The second assumption is moredifficult to directly assess. We cannot independently quantify theimpact of Oi-C inclusion on the relationship between Oea-C sub-strate use and d13C-PLFA of extant microbial communities. The useof the Oea only soil as a control, however, does enable us to accu-rately compare patterns of Oi vs. Oea substrate use across tem-peratures and sites. The use of such a control enabled us tointegrate variation in the relationship between d13C-PLFA and Oea-C substrate use attributed to both site and temperature. Therefore,changes in substrate use by groups identified via PLFA with site ortemperature are comparable using this approach. However, we usecaution when directly comparing absolute Oi or Oea-C incorpo-ration among the identified microbial groups using d13C PLFAwithin a given site or temperature.

3.3. Statistical approaches

We tested the effect of site (GC or SR), temperature (15 �C or20 �C), and their interactions on the absolute concentration (mgPLFA g SOC�1) and weight % PLFA by microbial grouping (bacteria,Gram negative bacteria, Gram positive bacteria, actinobacteria,fungi, protozoan, and total PLFA) and the ratio of fungal to bacterialPLFA (F:B) in all soils using a generalized linear model (GLM;Lindsey, 1997). As a secondary measure, we tested the influence ofOi replacement on these same PLFA metrics including its inter-actionwith site and temperature to determine potential differencesin soil microbial communities incubated with and without thereplaced Oi material. This aided our interpretation of the Oi-Cincorporation datasets. The only other direct testing of treatmenteffects in the Oea only soils was for d13C of PLFA in the soils col-lected on day 120 from all the Oea only treatment samples, todetermine the effect of site, temperature, or their interaction on thenatural abundance d13C of PLFA.

The d13C-PLFA results for the Oi þ Oea and the Oea only treat-ments were tested for significant differences using paired t-testsprior to calculating any of the three Oi-C incorporation metrics. Thecalculated values of absolute Oi-C incorporated into PLFA, fractionof Oi-C incorporated into PLFA and proportion of PLFA-C derivedfrom Oi-C were all tested for the effect of site, temperature andtheir interaction using a GLM for the above-specified PLFA group-ings. The only exceptions were the F:B ratio in the case of both thefraction of Oi-C incorporated into PLFA and proportion of PLFA-Cderived from Oi-C to avoid redundancy, and the total PLFA in thecase of the fraction of Oi-C incorporated into PLFA because it is usedin the calculation of this metric.

For all the effects testing described above, we employed ChiS-quare (c2) statistics from analysis of deviance within the GLMframework to assess significance. The assumptions of homoge-neous and normal errors for each GLM were checked by plottingresiduals versus fits, and by plotting residuals as probability plots(Lindsey, 1997). In the cases where assumptions were violated, were-ran the models using an exponential error distribution. Theexponential distribution is a special case of the gamma distribution,and like the gamma distribution its effect is to reduce the weight

S.E. Ziegler et al. / Soil Biology & Biochemistry 60 (2013) 23e32 27

given to observations exhibiting greater error (Lindsey,1997). In theGLMs using the exponential error distribution (1 out of 45 total), weassessed plots of deviance residuals vs. predicted values to deter-mine if the assumptions of the model were met (Hoffmann, 2004).We performed post-hoc t-tests when we found significant in-teractions to determine the significance of temperature effectswithin each site. All GLMs and t-tests were calculated using JMP 8.1(SAS; Cary, North Carolina). We considered results to be statisticallysignificant using an alpha level of 0.05.

4. Results

4.1. Phospholipid fatty acid quantity and distribution

Total PLFA and most PLFA groupings exhibited an increase inconcentration with temperature within the Oi þ Oea soils (Figs. 1aand 2a, b; see Appendix 1 for all statistical results). Gram negativebacteria and fungi, however, exhibited increases with temperaturewithin the Oi þ Oea soils that were only marginally significant(0.05 < p < 0.07; Fig. 2). The F:B ratio decreased with temperaturein the Oi þ Oea soils (Fig. 1b).

Changes in the relative quantities of PLFA (weight %) by groupwithin the Oi þ Oea soils occurred with higher temperature, butwere limited to the bacterial PLFA, andwere not observed for fungalor eukaryotic PLFA. Furthermore, bacterial PLFAweight % and, morespecifically, Gram positive bacterial PLFA, increased with temper-ature while Gram negative bacterial PLFA weight % decreased.

Fig. 1. Total concentration of phospholipid fatty acids (PLFA; a) and the ratio of con-centration of fungal to bacterial PLFA from the Salmon River (SR) and Grand Codroy(GC) soils incubated at 15 �C or 20 �C and collected after 120 days. Light bars refer toresults for the Oea only (or control) soils while the dark bars represent PLFA concen-trations from the Oi þ Oea soils. Data are presented in units of mg PLFA per g soilorganic carbon (mg PLFA g SOC�1) and as the mean (n ¼ 3) � the standard error.

Site effects for the absolute quantities of PLFA in the Oi þ Oeasoils were limited to the protozoan PLFA, which exhibited a greaterconcentration in the warmer GC soils (Fig. 2b). The bacterial PLFA,and more specifically the Gram negative bacterial and actino-bacteria PLFA, all exhibited a greater weight % in the cooler SR siterelative to warmer GC site soils. In contrast, the protozoan PLFAexhibited a greater weight % in GC soils (the warmer site) relative toSR soils (the cooler site), congruent with the absolute quantities ofthis PLFA.

Total PLFA concentration ranged from 1220 � 58 for the SROi þ Oea soils incubated at 15 �C to 2347 � 344 mg PLFA g SOC�1 forthe GC Oi þ Oea soils at 20 �C. Total PLFA content of these soils didnot vary between the Oea only (controls) and Oi þ Oea soils but,congruent with the Oi þ Oea only results, increased overall withtemperature (Fig. 1a; Appendix 2 for full statistical results). Theimpact of temperature was far more ubiquitous across PLFA groupsrelative to site effects, with quantities of PLFA increasing withtemperature in all groups except actinobacteria (Fig. 2; Appendix2). Temperature effects included reduced weight % of Gram neg-ative bacterial PLFA with temperature and a near-significantincrease with warming in weight % of protozoan PLFA (Fig. 2;Appendix 2). As opposed to the Gram negative PLFA, for whichabsolute increases with temperature were detected but relativeabundance as weight % decreased, the weight % of Gram positivePLFA increased with temperature, congruent with the absolutequantities of Gram positive bacteria PLFA. This suggests that theincrease in biomass of Gram positive bacteria occurred withwarming to a greater degree than for other microbial groups. Fur-thermore, the fungal and eukaryotic PLFA did not vary in weight %with temperature.

Congruent with the Oi þ Oea results, total soil PLFA content washigher overall in Oea only, GC soils (the warmer site). Site effectswere also observed for the protozoan and fungal PLFA, such that allwere found in greater quantities in GC soils (Fig. 2b, d). Site effectswere also detected as decreased weight % of actinobacteria andGram negative bacterial PLFA in the warmer site (GC) relative tocooler site (SR) soils. However, weight % of protozoan PLFA wasgreater in the warmer GC soils. Contrary to this, fungal PLFA did notvary in weight % with site likely because they varied with site tosimilar degrees as the total PLFA pool.

The quantities of fungal PLFA as well as the ratios of fungal tobacterial PLFA were higher in the Oi þ Oea soils, representing theonly PLFAwhose quantity or ratio varied between the Oea only andOi þ Oea soils. Although variation between the Oea only andOiþ Oea soils in absolute PLFA content was primarily limited to thefungal PLFA, relative quantities varied with litterfall treatmentmore ubiquitously across PLFA groups (Appendix 2). Weight % offungal PLFA was greater in Oi þ Oea soils whereas all othergroupings, with the exception of protozoan PLFA, were lower inweight % in Oi þ Oea soils as compared with Oea only soils.

We evaluated the relationship among the quantities of 18:2u6,18:1u9t and 18:1u9c to assess our original approach of solely using18:2u6 as fungal PLFA and treating 18:1u9t and 18:1u9c as non-specific groups due to their ubiquitous presence across microbialgroups. Neither weight % 18:1u9c (p ¼ 0.4072; r2 ¼ 0.0314) nor18:1u9t (p ¼ 0.0949; r2 ¼ 0.0817) varied with the weight % of18:2u6. Weight % 18:1u9c, however, varied positively, thoughweakly, with the weight percent of Gram negative bacterial PLFA(p ¼ 0.0475; r2 ¼ 0.1668) whereas weight percent 18:1u9t did notvary with Gram negative bacterial PLFA (p ¼ 0.5370; r2 ¼ 0.0176).Weight % 18:2u6 varied negatively with weight % of Gram negativePLFA as expected (p ¼ 0.0215; r2 ¼ 0.2178). This suggests that18:1u9c was likely derived from both fungi and Gram negativebacteria while 18:1u9t may be indicative of another population offungi not containing significant quantities of 18:2u6.

Fig. 2. The concentration of phospholipid fatty acids (PLFA) by PLFA grouping from the Salmon River (SR) and Grand Codroy (GC) soils incubated at 15 �C or 20 �C and collected after120 days. Light bars refer to results for the Oea only (or control) soils while the dark bars represent PLFA concentrations from the Oi þ Oea soils. Data are presented in units of mgPLFA per g soil organic carbon (mg PLFA g SOC�1) and as the mean (n ¼ 3) � the standard error.

S.E. Ziegler et al. / Soil Biology & Biochemistry 60 (2013) 23e3228

4.2. Stable carbon isotopic composition of phospholipid fatty acids

To compare effects of temperature and site in Oi-C incorporation,we limited the dataset to the 11 PLFA resolved across both sets ofincubations (Oea only and Oi þ Oea soils): i15:0, a15:0, i16:0, 16:0,16:1u7, 17:1u9, cy17:0, cy19:0, 18:1u9t, 18:1u9c, and 18:2u6 (seeAppendix 3 for d13C-PLFA data). The difference in d13C of each in-dividual PLFA measured in Oi þ Oea soils versus Oea only soils wassignificant in all cases with the exception of i16:0 in SR soils at 15 �C,16:1u7and18:1u9t in SR soils at 20 �C, and16:1u7and17:1u9 inGCsoils at 20 �C. The significantly lower PLFA signatures in all but thesefive individual cases reflects incorporation of C derived from Oimaterial and was used to estimate incorporation of Oi-C into soilPLFA. In the five exceptions, where a difference in d13C between thetwo treatments was not significant, we assumed no microbialincorporation of Oi-C into individual PLFA before estimating Oi-Cincorporation into the PLFA groupings to which they contribute.The average difference between d13C of PLFA in Oi þ Oea soil minusthe PLFA from Oea only soils (DOi-C) for all combined PLFA resolvedranged from �3.6 � 0.9& to �2.5 � 0.8 across both sites and tem-peratures. The DOi-C for individual PLFA across the entire datasetranged from �6.4 � 1.2 to �1.0 � 0.4& (Appendix 3).

4.3. Oi-C incorporation into phospholipid fatty acids

The d13C PLFA of Oi þ Oea and Oea-only soils were used tocalculate the absolute Oi-derived C in individual or groups of PLFAper g SOC, the fraction of total Oi-derived C in PLFA incorporatedinto individual and groups of PLFA, and the proportion of PLFA-Cderived from Oi-C within individual or groups of PLFA. Thesethree metrics enabled us to investigate the influence of site andtemperature on the (1) net incorporation of Oi-C into microbialbiomarkers and therefore, its absolute allocation among PLFAgroups, (2) distribution of the Oi-derived C incorporation among

PLFA groups, and (3) potential preference between Oi and Oea-derived Cwithin amicrobial group as defined by PLFA in these soils.

The quantity of Oi-derived C incorporated into total PLFA,measured as mg Oi-C g SOC�1, ranged from 445 � 77 in the SR soilsincubated at 15 �C to 816 � 253 in the GC soils incubated at 20 �C.The quantity of Oi-derived C incorporated into a given PLFA groupwas lowest for the Gram positive bacterial PLFA and highest amongthe fungal and Gram negative bacterial PLFA, for which it rangedfrom 109 � 5 to 133 � 14 and 81 �12 to 161 � 29 mg Oi-C g SOC�1,respectively. The quantity of Oi-derived C incorporated into PLFAafter 120 days varied only with temperature in the case of theGram-positive bacterial PLFA (Appendix 4), such that greater Oi-derived C was incorporated into these PLFA at the higher temper-ature (Fig. 3a). The ratio of the incorporation of Oi-derived C intofungal relative to bacterial PLFA, however, was greater in the coolerSR site soil and exhibited an almost significant interaction of siteand temperature (p ¼ 0.0753), suggesting a decrease of Oi-derivedC incorporation into fungi relative to bacteria with temperature inthe SR soils (Fig. 3b).

The fraction of total Oi-derived C incorporated into PLFAdetected in total bacterial and Gram positive bacterial PLFAincreased significantly with temperature (Fig. 4a). An opposite,near-significant trend was observed for the fraction of Oi-derivedC in fungal PLFA, which tended to decrease with temperature(p ¼ 0.0756; Fig. 4b). Further, the fraction of Oi-derived C incor-porated into Gram positive PLFA exhibited a site by temperatureeffect such that that the increase with temperature was greater forcooler SR site soils. Fungal PLFA also exhibited a significant siteeffect indicating a greater fraction of the total Oi-derived C incor-poration into these PLFA occurred in the cooler site (SR) relative tothe warmer site (GC) soils, congruent with the decreasing trend inthe fraction of Oi-C in this PLFA with temperature (Fig. 4b).

We used proportions of Oi versus Oea-derived C in PLFA groupsin these soils to estimate microbial C source and how it may have

Fig. 3. The quantity of Oi-carbon in the Gram positive bacterial (a) phospholipid fattyacids (PLFA), and the ratio of the quantity of Oi-C in the fungal to bacterial PLFA (b) at120 days in the Salmon River (SR) and Grand Codroy (GC) soils incubated at 15 �C and20 �C. Values were determined from the stable carbon isotopic values of the added Oimaterial, and the individual PLFA within the Oi-Oea and Oea only soils and are pro-vided as the mean � one standard error (n ¼ 3) and in units of mg Oi-C incorporationinto the PLFA per g soil organic carbon (SOC).

Fig. 4. The fraction of the total Oi carbon incorporated into soil phospholipid fattyacids (PLFA) by PLFA groups as Gram positive bacterial (a), and fungal (b) PLFA in theSalmon River (SR) and Grand Codroy (GC) soils incubated at both 15 �C and 20 �C. Thismeasure provides a means to assess the difference in where Oi-C ended up within thePLFA group at 120 day by site and temperature. Values are reported as the mean(n ¼ 3) � 1 standard error.

S.E. Ziegler et al. / Soil Biology & Biochemistry 60 (2013) 23e32 29

varied across microbial groups. Gram positive bacterial PLFA, whichexhibited increases in net Oi-C incorporationwith warming, did notexhibit any change in the proportion of PLFA-C derived from Oi-C(Fig. 5a; Appendix 4). This suggests that increased Oi-C in Grampositive bacterial PLFA was likely due to biomass increases and notnecessarily a change in C-source use with warming. The marginallysignificant site by temperature effect, however, suggests that SRsoils may have experienced ameaningful increase in the proportionof Oi-C incorporated into Gram positive PLFAwith warming. FungalPLFA exhibited temperature and marginally significant site effects,indicating that warming decreased the proportion of Oi-C incor-poration into these PLFA and that the proportion of Oi-C tended tobe lower in the warmer GC site soils (Fig. 5b). No other PLFA groupsexhibited any significant site or temperature effects in terms of theproportion of Oi-C incorporated into PLFA-C.

5. Discussion

Climate warming can change how litterfall is routed through themicrobial community in ways relevant for both soil organic matterformation and loss. The greater microbial biomass at the warmersite and increased biomass with temperature in soils from bothsites offer complementary evidence for increased microbial pro-cessing of O-horizon soil, and associated potential for altered for-mation of microbial byproducts, in a warmer climate. This increasein soil organic matter processing was linked to greater, preferentialmineralization of slower-turnover SOC pools (Li et al., 2012). Thecurrent work expands on those results by identifying the microbial

players likely driving those temperature effects, and highlightsconsistencies between SOC responses to increased temperature inthe laboratory and differences in microbial activities between siteswith distinct temperature regimes.

The potential significance of warming effects observed in thisstudy is underscored by the recognition of the potentially largemicrobial contributions to soil organicmatter pools (Amelung et al.,2001; Simpson et al., 2007; Liang et al., 2010). Warming-inducedchanges in how C inputs are incorporated and passed through themicrobial community have the potential to greatly alter soil for-mation processes through changes in the proportions of majormicrobial byproducts produced. For example, results here implythat climate warming may reduce the relative amount of Oi con-verted into fungal chitin, while potentially increasing its conversioninto bacterial peptidoglycan. These products appear to have dif-ferent residence times in soil (Amelung et al., 2001; Liang et al.,2007; Spence et al., 2011). Changes in the relative production ofthese two microbial products from labile source inputs could,therefore, have ramifications for soil formation, potentially influ-encing climate feedbacks.

5.1. Warming alters the routing of SOC pools through soil bacteriaand fungi

Changes with temperature in Oi-C and Oea-C flow throughdistinct microbial groups were more evident compared to changesin absolute or relative microbial biomass. Across all treatments, and

Fig. 5. The proportion of Gram positive bacterial (a), and fungal (b) phospholipid fattyacid carbon (PLFA-C) derived from the added isotopically unique Oi material at 120days in the Salmon River (SR) and Grand Codroy (GC) soils incubated at either the 15 �Cor 20 �C temperatures. Values are given as the mean � one standard error (n ¼ 3) ofthe proportion of PLFA-C in each PLFA group as mg Oi-C per mg PLFA-C.

S.E. Ziegler et al. / Soil Biology & Biochemistry 60 (2013) 23e3230

regardless of temperature, fungi primarily incorporated added Oi-C(60e98% of PLFA-C) while Gram positive and Gram negative bac-teria derived a smaller fraction of biomass from added Oi-C (23e44% and 40e54% of PLFA-C, respectively). Further, when Oi mate-rial was available, PLFA quantities suggest a relative decrease in theabundance of bacteria relative to fungi. This is congruent withenhancement of fungal activity on higher C:N substrates as hasbeen reported in other studies (Hu et al., 2001; Carney et al., 2007).Combined, these results indicate that fungi in these soils appear toprimarily access relatively labile, higher C:N Oi-C while Grampositive bacteria more readily incorporate lower C:N, slowerturnover Oea-C. The experimental results, however, suggestwarming can significantly alter those preferences. The data suggestthat the greater increase in mineralization of relatively slow-turnover SOC with warming compared to more labile materialduring these incubations (Li et al., 2012) was likely the result of anincrease in fungal use of Oea-C with experimental warming (Figs. 3and 4). Further, the proportion of Gram positive PLFA-C derivedfrom Oi-C doubled with increased temperature in soils from thecooler site. Consequently, Oi-C incorporated into fungal relative tobacterial PLFA decreased when soils from the cooler site werewarmed, such that these F:B ratios were similar to those from thewarmer site soils. Though Gram positive PLFA increased in quantitywith temperature, there was no temperature effect on the pro-portion of PLFA-C from the Oi-C versus Oea-C sources in these PLFA.In combination these data suggest the preferential increase inmineralization of Oea-C with laboratory warming likely did notresult from Gram positive bacterial activity, but rather fungi.

Congruent with these incubation temperature effects, the pro-portion of Gram positive PLFA-C derived from Oi-C was greater andproportions of Oi-C routed through the fungal relative to the bac-terial community were lower in soils from the warmer site. Sim-ilarly, we observed a near-significant trend of lower Oi-Cincorporation by fungi and, by difference, a generally greater Oea-C incorporation by fungi within soils from the warmer site(p ¼ 0.0527; Fig. 5b). Such site effects suggest that soils from thewarmer climate experience a greater proportion of labile Oi-Crouted through bacteria relative to fungi, or a greater routing ofmore slow-turnover Oea-C through fungi relative to bacteria. Lowerproportions of fungal PLFA derived from Oi-C, whether due tolaboratory warming or a warmer native climate, suggest that theincreased PLFA-derived fungal biomass with laboratory warmingand in soils from the warmer site was derived from greater fungalincorporation of Oea-C.

Enhancement of fungal exploitation of the low C:N Oea-C poolswith temperature observed here may signify a warming inducedmechanism relevant to microbial activity and growth in these soils.Organic N is an important source of microbial N in high latitudeforest soils where its accessibility can be limited by temperature(Hobbie et al., 2002; Mack et al., 2004; Allison et al., 2008). In thisexperiment warming increased both microbial respiration (Li et al.,2012) and biomass, likely resulting in an increase in microbialnutrient demand. The proportion of CO2 respired from Oea wassignificantly greater with warming, increasing from 63e84% in theSR soils and 48e92% in GC soils in the 15 and 20 �C incubations,respectively (Li et al., 2012). This increased mineralization ofslower-turnover, lower C:N soil organic matter pools (Oea) withwarming, particularly in the SR soils and likely by fungi, may havealtered the source and/or availability of N in these soils. Though wehave no direct evidence for changes in N availability, the increasedmineralization of Oea may have supported the N required for thewarming induced increases in bacterial biomass and Gram positivebacterial incorporation of the high C:N Oi material. Demands forN may be higher for bacteria relative to fungi given evidence fortheir lower C:N biomass relative to fungi (McGill, 1981; Guesewelland Gessner, 2009). Though Gram positive bacteria are known tohave thicker cell walls, we do not know if these organisms exhibitgreater N demand relative to Gram negative bacteria, nor can weknow the influence of warming on thesemicrobial groups’ resourcedemands. The greater incorporation of the high C:N Oi material byGram positive bacteria observed with warming, however, likely didinduce increased N demand by this group. Therefore, the warminginduced increases in use of lower C:N, slow turnover soil organicmatter pools by fungi may be an important mechanism alteringN cycling in these forest soils in a warmer climate.

5.2. Soil warming may stimulate increased microbial biomass andchanges in its composition

The increased total PLFA quantities with experimental warming,and in the soils native to thewarmer GC sitemay signify an increasein overall microbial biomass on both seasonal and longer timescales. If robust and sustained, an increase in microbial biomasswith warming suggests that warming is likely to have a longerlasting effect of increasing the contributions of microbial biomassto SOC pools in these forests, as noted in other ecosystems (Bellet al., 2010). More specifically, the current study suggests thatwarming of boreal O horizons may induce an increase in Grampositive bacterial biomass that could signify enhanced generationof peptidoglycan under warmer conditions. Gram positive bacterialPLFA represented the only group that exhibited both an absoluteconcentration and weight % increase with temperature, while thefungal to bacterial PLFA ratio decreased with warming. This is

S.E. Ziegler et al. / Soil Biology & Biochemistry 60 (2013) 23e32 31

congruent with bacterial and fungal biomass changes observed aspart of a 12 year soil warming experiment in the temperate forest ofHubbard Brook (Frey et al., 2008), suggesting these patterns may bemore universal and indicative of longer-term warming effects.

Although microbial community composition measured via PLFAdid not exhibit significant site effects, results suggest potential forgreater eukaryote microheterotrophy in soils from the warmerclimate. The protozoan PLFA exhibited greater weight % in the GCrelative to SR soils. Though slight in terms of absolute weight %change, the site effect observed for the protozoan PLFA suggestsincreased microbial food web activities in the warm versus coolersite soils. Assuming this PLFA is proportional to the biomass of soilprotozoans, protozoan biomass may have been as much as 2e3times greater in the GC relative to SR soils, a difference presum-ably reflecting protozoan consumption of other microbial groups.Indeed, protozoans are known to respond rapidly to increases inmicrobial biomass (Bonkowski, 2004). In contrast to the protozoanPLFA, weight% of Gram negative and actinobacteria PLFAwas lowerin GC relative to SR soils e a difference that was small, but statis-tically significant. The lack of larger differences between the twosoils may be a result of both the adaption to the common incuba-tion warming (Luo et al., 2001; Bradford et al., 2008; Bradford andWatts, 2010) and potential protozoan grazing stimulated by theincreased temperature and resulting increase in microbial biomassand its potential contribution to soil organic matter pools.

5.3. Conclusions and implications

Experimental warming in these boreal forest soils inducedchanges in substrate routing through microbial groups relevant topatterns of soil organic matter decay and formation. Preferentialincreases in decay of low C:N material with warming implya change in release of assimilable resources for microbial uptake(Billings and Ballantyne, 2012; Lehmeier et al., 2013), and representa warming-induced shift in decay patterns with potentiallyimportant implications for future CO2 release. The effects observedhere further suggest longer-term warming may alter contributionsof biomass remnants to SOM (Amelung et al., 2001; Simpson et al.,2007) and their influence on soil organic matter geochemistry andturnover. The consistency among datasets, whether obtained fromthe laboratory incubations with varied temperature or intact soilprofiles exposed to distinct MAT, suggests these mechanisms maybe relevant for longer-term climate change scenarios. Temperatureeffects on substrate routing and microbial biomass were consistentwith the patterns observed across sites with distinct MAT. The siteeffects on substrate routing and microbial biomass were also con-gruent with processes (i.e. increased phenol oxidase activity) andgeochemical indicators (e.g. decreased C:N ratio) observed in soilsnative to the warmer forest relative to cooler forest sites (Li et al.,2012). These findings highlight the need for investigations of howaltered routing of labile and slower-turnover C substrates throughbacterial and fungal groups influences soil organic matter decayand formation, and incorporate those results into projections ofrespiratory feedbacks to climate warming.

Acknowledgments

We would like to thank Darrell Harris and Martin Moroni fortheir input regarding site selection and information as well asRoxane Bowden, Nameer Baker and Alison King for their laboratoryassistance. This research was supported by funding to S.E.Z. andS.A.B. from the Humber River Basin Project, a research initiativeadministered through the Grenfell Campus of Memorial University,and the Centre for Forest Sustainability and Innovation bothestablished by the province of Newfoundland and Labrador.

Support was also provided through grants from the Natural Sci-ences and Engineering Research Council, Canada Research ChairsProgram, and Canadian Foundation for Innovation to S.E.Z and theW.M. Keck Foundation to M.L.F. The authors would also like tothank the handling editor and two anonymous reviewers for theircomments and suggestions that greatly improved this manuscript.

Appendix A. Supplementary data

Supplementary data related to this article can be found at http://dx.doi.org/10.1016/j.soilbio.2013.01.001.

References

Allison, S.D., Czimczik, C.I., Treseder, K.K., 2008. Microbial activity and soil respi-ration under nitrogen addition in Alaskan boreal forest. Global Change Biology14, 1156e1168.

Amelung, W., Miltner, A., Zhang, X., Zech, W., 2001. Fate of microbial residues duringlitter decomposition as affected by minerals. Soil Science 166, 598.

Andrews, J.A., Matamala, R., Westover, K.M., Schlesinger, W.H., 2000. Temperatureeffects on the diversity of soil heterotrophs and the delta C-13 of soil-respiredCO2. Soil Biology and Biochemistry 32, 699e706.

Bell, T.H., Klironomos, J.N., Henry, H.A.L., 2010. Seasonal responses of extracellularenzyme activity and microbial biomass to warming and nitrogen addition. SoilScience Society of America Journal 74, 820.

Biasi, C., Rusalimova, O., Meyer, H., Kaiser, C., Wanek, W., Barsukov, P., Junger, H.,Richter, A., 2005. Temperature-dependent shift from labile to recalcitrant car-bon sources of arctic heterotrophs. Rapid Communications in Mass Spectrom-etry 19, 1401e1408.

Billings, S.A., Ballantyne IV, F., 2012. How interactions between microbial resourcedemands, soil organic matter stoichiometry and substrate reactivity determinethe direction and magnitude of soil respiratory responses to warming. GlobalChange Biology. http://dx.doi.org/10.1111/gcb.12029.

Bonkowski, M., 2004. Protozoa and plant growth: the microbial loop in soil revis-ited. New Phytologist 162, 617e631.

Bradford, M., Davies, C., Frey, S., 2008. Thermal adaptation of soil microbial respi-ration to elevated temperature. Ecology Letters 11, 1316e1327.

Bradford, M., Watts, B., 2010. Thermal adaptation of heterotrophic soil respiration inlaboratory microcosms. Global Change Biology 16, 1576e1588.

Butler, E., Whelan, M.J., Ritz, K., Sakrabani, R., van Egmond, R., 2011. Solvent-basedwashing removes lipophilic contaminant interference with phospholipid fattyacid analysis of soil communities. Soil Biology and Biochemistry 43, 2208e2212.

Cannone, N., Wagner, D., Hubberten, H.W., Guglielmin, M., 2008. Biotic and abioticfactors influencing soil properties across a latitudinal gradient in Victoria Land,Antarctica. Geoderma 144, 50e65.

Carney, K.M., Hungate, B.A., Drake, B.G., Megonigal, J.P., 2007. Altered soil microbialcommunity at elevated CO2 leads to loss of soil carbon. Proceeding of the Na-tional Academy of Science USA 104, 4990e4995.

Claus, H., 2003. Laccases and their occurrence in prokaryotes. Archives of Micro-biology 179, 145e150.

Conant, R.T., Ryan, M.G., Ågren, G.I., Birge, H.E., Davidson, E.A., Eliasson, P.E.,Evans, S.E., Frey, S.D., Giardina, C.P., Hopkins, F.M., Hyvönen, R.,Kirschbaum, M.U.F., Lavallee, J.M., Leifeld, J., Parton, W.J., Megan Steinweg, J.,Wallenstein, M.D., Martin Wetterstedt, J.Å., Bradford, M.A., 2011. Temperatureand soil organic matter decomposition rates e synthesis of current knowledgeand a way forward. Global Change Biology 17, 3392e3404.

Craine, J.M., Fierer, N., McLauchlan, K.K., 2010. Widespread coupling between therate and temperature sensitivity of organic matter decay. Nature Geoscience 3,854e857.

Dalenberg, J., Jager, G., 1989. Priming effect of some organic additions to C-14-labeled soil. Soil Biology and Biochemistry 21, 443e448.

Davidson, E.A., Janssens, I.A., 2006. Temperature sensitivity of soil carbon decom-position and feedbacks to climate change. Nature 440, 165e173.

Duran, N., Rosa, M., D’Annibale, A., Gianfreda, L., 2002. Applications of laccases andtyrosinases (phenoloxidases) immobilized on different supports: a review.Enzyme and Microbial Technology 31, 907e931.

Federle, T., Dobbins, D., Thornton-Manning, J., Jones, D., 1986. Microbial biomass,activity, and community structure in subsurface soils. Ground Water 24,365e374.

Fernandez, C., 2011. The role of chitin in the decomposition of ectomycorrhizalfungal litter. Ecology 93, 24e28.

Findlay, R., 2004. Determination of microbial community structure using phos-pholipid fatty acid profiles. In: Kowalchuk, G.A. (Ed.), Molecular MicrobialEcology Manual.

Fissore, C., Giardina, C., Swanston, C., 2009. Variable temperature sensitivity of soilorganic carbon in North American forests. Global Change Biology 15, 2295e2310.

Frey, S.D., Drijber, R., Smith, H., Melillo, J., 2008. Microbial biomass, functional ca-pacity, and community structure after 12 years of soil warming. Soil Biology andBiochemistry 40, 2904e2907.

S.E. Ziegler et al. / Soil Biology & Biochemistry 60 (2013) 23e3232

Frostegard, A., Baath, E., 1996. The use of phospholipid fatty acid analysis to estimatebacterial and fungal biomass in soil. Biology and Fertility of Soils 22, 59e65.

Frostegard, A., Baath, E., Tunlio, A., 1993. Shifts in the structure of soil microbialcommunities in limed forests as revealed by phospholipid fatty acid analysis.Soil Biology and Biochemistry 25, 723e730.

Frostegard, A., Tunlid, A., Bååth, E., 2011. Use and misuse of PLFA measurements insoils. Soil Biology and Biochemistry 43, 1621e1625.

Gu, L., Post, W., King, A., 2004. Fast labile carbon turnover obscures sensitivity ofheterotrophic respiration from soil to temperature: a model analysis. GlobalBiogeochemical Cycles 18, GB1022.

Guesewell, S., Gessner, M.O., 2009. N: P ratios influence litter decomposition andcolonization by fungi and bacteria in microcosms. Functional Ecology 23,211e219.

Hartley, I.P., Ineson, P., 2008. Substrate quality and the temperature sensitivity of soilorganic matter decomposition. Soil Biology and Biochemistry 40, 1567e1574.

Hendrey, G., Ellsworth, D., Lewin, K., 1999. A free-air enrichment system forexposing tall forest vegetation to elevated atmospheric CO2. Global ChangeBiology 5, 293e309.

Hobbie, S., Nadelhoffer, K., Hogberg, P., 2002. A synthesis: the role of nutrients asconstraints on carbon balances in boreal and arctic regions. Plant Soil 242,163e170.

Hoffmann, J.P., 2004. Generalized Linear Models. Allyn & Bacon.Horvath, R.S., 1972. Microbial co-metabolism and the degradation of organic

compounds in nature. Bacteriology Reviews 36, 146e155.Hu, S., Chapin, F., Firestone, M., Field, C., Chiariello, N., 2001. Nitrogen limitation of

microbial decomposition in a grasslandunderelevatedCO2.Nature 409,188e191.IPCC, 2007. IPCC Fourth Assessment Report: Climate Change 2007, vol. 4. Inter-

governmental Panel on Climate Change, pp. 213e252.Janssens, I.A., Vicca, S., 2010. Biogeochemistry: soil carbon breakdown. Nature

Geoscience 3, 823e824.Janzen, H., 2004. Carbon cycling in earth systems e a soil science perspective.

Agriculture, Ecosystems and the Environment 104, 399e417.Kirby, R., 2005. Actinomycetes and lignin degradation. Advances in Applied

Microbiology 58, 125e168.Kirschbaum, M., 1995. The temperature-dependence of soil organic-matter

decomposition, and the effect of global warming on soil organic-C storage.Soil Biology and Biochemistry 27, 753e760.

Kirschbaum, M.U.F., 2000. Will changes in soil organic carbon act as a positive ornegative feedback on global warming. Biogeochemistry 48, 21e51.

Kuzyakov, Y., Friedel, J., Stahr, K., 2000. Review of mechanisms and quantification ofpriming effects. Soil Biology and Biochemistry 32, 1485e1498.

Lehmeier, C.A., Min, K., Niehues, N.D., Ballantyne IV, F., Billings, S.A., 2013. Tem-perature-mediated changes of exoenzyme-substrate reaction rates and theirconsequences for the carbon to nitrogen flow ratio of liberated resources. SoilBiology and Biochemistry 57, 374e382.

Leifeld, J., Zimmermann, M., Fuhrer, J., Conen, F., 2009. Storage and turnover ofcarbon in grassland soils along an elevation gradient in the Swiss Alps. GlobalChange Biology 15, 668e679.

Li, J., Ziegler, S., Lane, C.S., Billings, S.A., 2012. Warming-enhanced preferentialmicrobial mineralization of humified boreal forest soil organic matter: Inter-pretation of soil profiles along a climate transect using laboratory incubations.Journal of Geophysical Research-biogeochemistry 117, G02008.

Liang, C., Cheng, G., Wixon, D.L., Balser, T.C., 2010. An absorbing Markov Chainapproach to understanding the microbial role in soil carbon stabilization. Bio-geochemistry 106 (3), 303e309.

Liang, C., Fujinuma, R., Balser, T.C., 2008. Comparing PLFA and amino sugars formicrobial analysis in an Upper Michigan old growth forest. Soil Biology andBiochemistry 40, 2063e2065.

Liang, C., Zhang, X., Balser, T.C., 2007. Net microbial amino sugar accumulationprocess in soil as influenced by different plant material inputs. Biology andFertility of Soils 44, 1e7.

Lichter, J., Billings, S.A., Ziegler, S.E., Gaindh, D., Ryals, R., Finzi, A.C., Jackson, R.B.,Stemmler, E.A., Schlesinger, W.H., 2008. Soil carbon sequestration in a pineforest after 9 years of atmospheric CO2 enrichment. Global Change Biology 14,2910e2922.

Lindsey, J.K., 1997. Applying Generalized Linear Models (Springer Texts in Statistics).Springer.

Lloyd, J., Taylor, J., 1994. On the temperature dependence of soil respiration. Func-tional Ecology, 315e323.

Luo, Y., Wan, S., Hui, D., Wallace, L., 2001. Acclimatization of soil respiration towarming in a tall grass prairie. Nature 413, 622e625.

Mack, M., Schuur, E., Bret-Harte, M., Shaver, G., Chapin, F., 2004. Ecosystem carbonstorage in arctic tundra reduced by long-term nutrient fertilization. Nature 431,440e443.

McGill, W., 1981. Comparative aspects of cycling of organic C, N, S and P through soilorganic matter. Geoderma 26, 267e286.

Melillo, J., Steudler, P., Aber, J., Newkirk, K., Lux, H., Bowles, F., Catricala, C., Magill, A.,Ahrens, T., Morrisseau, S., 2002. Soil warming and carbon-cycle feedbacks to theclimate system. Science 298, 2173e2176.

Nakatsubo, T., Uchida, M., Horikoshi, T., Nakane, K., 1997. Comparative study of themass loss rate of moss litter in boreal and subalpine forests in relation totemperature. Ecological Research 12, 47e54.

Oechel, W.C., Vourlitis, G.L., Hastings, S.J., Zulueta, R.C., Hinzman, L., Kane, D., 2000.Acclimation of ecosystem CO2 exchange in the Alaskan Arctic in response todecadal climate warming. Nature 406, 978e981.

Ogawa, H., Amagai, Y., Koike, I., Kaiser, K., Benner, R., 2001. Production of refractorydissolved organic matter by bacteria. Science 292, 917e920.

Olsson, P.A., 1999. Signature fatty acids provide tools for determination of the dis-tribution and interactions of mycorrhizal fungi in soil. FEMS MicrobiologyEcology 29, 303e310.

Potthoff, M., Dyckmans, J., Flessa, H., Beese, F., Joergensen, R.G., 2008. Decom-position of maize residues after manipulation of colonization and its con-tribution to the soil microbial biomass. Biology and Fertility of Soils 44,891e895.

Rabinovich, M., Bolobova, A., Vasil’chenko, L., 2004. Fungal decomposition of nat-ural aromatic structures and xenobiotics: a review. Applied Biochemistry andMicrobiology 40, 1e17.

Ringelberg, D.B., Davis, J.D., Smith, G.A., Pfiffner, S.M., Nichols, P.D., Nickels, J.S.,Henson, J.M., Wilson, J.T., Yates, M., Kampbell, D.H., 1989. Validation of signaturepolarlipid fatty acid biomarkers for alkane-utilizing bacteria in soils and sub-surface aquifer materials. FEMS Microbiol Letters 62, 39e50.

Rustad, L., Campbell, J., Marion, G., Norby, R., Mitchell, M., Hartley, A., Cornelissen, J.,Gurevitch, J., 2001. A meta-analysis of the response of soil respiration, net ni-trogen mineralization, and aboveground plant growth to experimental eco-system warming. Oecologia 126, 543e562.

Schindlbacher, A., de Gonzalo, C., Díaz-Pinés, E., Gorría, P., Matthews, B., Inclán, R.,Zechmeister-Boltenstern, S., Rubio, A., Jandl, R., 2010. Temperature sensitivity offorest soil organic matter decomposition along two elevation gradients. Journalof Geophysical Research 115, G03018.

Simpson, A.J., Simpson, M.J., Smith, E., Kelleher, B.P., 2007. Microbially derived in-puts to soil organic matter: are current estimates too low? EnvironmentalScience and Technology 41, 8070e8076.

Sinsabaugh, R.L., 2010. Phenol oxidase, peroxidase and organic matter dynamics ofsoil. Soil Biology and Biochemistry 42, 391e404.

Spence, A., Simpson, A.J., McNally, D.J., Moran, B.W., McCaul, M.V., Hart, K., Paull, B.,Kelleher, B.P., 2011. The degradation characteristics of microbial biomass in soil.Geochimica et Cosmochimica Acta 75, 2571e2581.

Steinberger, Y., Zelles, L., Bai, Q., Lutzow, von, M., Munch, J., 1999. Phospholipidfatty acid profiles as indicators for the microbial community structure in soilsalong a climatic transect in the Judean Desert. Biology and Fertility of Soils 28,292e300.

Strickland, M.S., Rousk, J., 2010. Considering fungal: bacterial dominance in soilsemethods, controls, and ecosystem implications. Soil Biology and Biochemistry42, 1385e1395.

Sierra, C.A., Trumbore, S.E., Davidson, E.A., Frey, S.D., Savage, K.E., Hopkins, F.M.,2012. Predicting decadal trends and transient responses of radiocarbon storageand fluxes in a temperate forest soil. Biogeosciences 9, 3013e3028.

Throckmorton, H.M., Bird, J.A., Dane, L., Firestone, M.K., Horwath, W.R., 2012. Thesource of microbial C has little impact on soil organic matter stabilisation inforest ecosystems. Ecology Letters 15, 1257e1265.

Tremblay, L., Benner, R., 2006. Microbial contributions to N-immobilization andorganic matter preservation in decaying plant detritus. Geochimica et Cosmo-chimica Acta 70, 133e146.

Trumbore, S.E., Chadwick, O.A., Amundson, R., 1996. Rapid exchange between soilcarbon and atmospheric carbon dioxide driven by temperature change. Science272, 393e396.

Wallenstein, M., 2011. Ecology of extracellular enzyme activities and organic matterdegradation in soil: a complex community-driven process. In: Dick, R.P. (Ed.),Methods of Soil Enzymology. SSSA Book Ser..

White, D., Davis, W., Nickels, J., King, J., Bobbie, R., 1979. Determination of thesedimentary microbial biomass by extractible lipid phosphate. Oecologia 40,51e62.

White, D., Ringelberg, D., 1998. Signature lipid biomarker analysis. In: Burlage, R.S.(Ed.), Techniques in Microbial Ecology, pp. 255e272.

White, D., Stair, J., Ringelberg, D., 1996. Quantitative comparisons of in situ micro-bial biodiversity by signature biomarker analysis. Journal of Industrial Micro-biology and Biotechnology 17, 185e196.

Wright, S.F., Jawson, L., 2001. A pressure cooker method to extract glomalin fromsoils. Soil Science Society of America Journal 65, 1734.

Wu, Y., Yu, X., Wang, H., Ding, N., Xu, J., 2009. Does history matter? Temperatureeffects on soil microbial biomass and community structure based on thephospholipid fatty acid (PLFA) analysis. Journal of Soils and Sediments 10,223e230.

Zelles, L., Bai, Q.Y., Beck, T., Beese, F., 1992. Signature fatty-acids in phospholipidsand lipopolysaccharides as indicators of microbial biomass and communitystructure in agricultural soils. Soil Biology and Biochemistry 24, 317e323.

Zhang, W., Parker, K.M., Luo, Y., Wan, S., Wallace, L.L., Hu, S., 2005. Soil microbialresponses to experimental warming and clipping in a tallgrass prairie. GlobalChange Biology 11, 266e277.

Ziegler, S.E., White, P.M., Wolf, D.C., Thoma, G.J., 2005. Tracking the fate and recy-cling of 13C-labeled glucose in soil. Soil Science 170, 767e778.

Zogg, G.P., Zak, D.R., Ringelberg, D.B., MacDonald, N.W., Pregitzer, K.S., White, D.C.,1997. Compositional and functional shifts in microbial communities due to soilwarming. Soil Science Society of America Journal 61, 475e481.