Pairing litter decomposition with microbial community ... - SOIL

SOIL 8 163ndash176 2022httpsdoiorg105194soil-8-163-2022copy Author(s) 2022 This work is distributed underthe Creative Commons Attribution 40 License

SOIL

Pairing litter decomposition with microbialcommunity structures using the Tea Bag Index (TBI)

Anne Daebeler12 Eva Petrovaacute1 Elena Kinz3 Susanne Grausenburger4 Helene Berthold5Taru Sandeacuten6 Roey Angel17 and the high-school students of biology project groups I II and

III from 2018ndash2019+1Soil amp Water Research Infrastructure (SoWa) Biology Centre CAS Ceskeacute Budejovice Czechia

2Centre for Microbiology and Environmental Systems ScienceDivision of Microbial Ecology University of Vienna Vienna Austria

3Open Science ndash Life Sciences in Dialogue Vienna Austria4Federal College for Viticulture and Fruit Growing Klosterneuburg Austria

5Institute for Seed and Propagating Material Phytosanitary Service and ApicultureDepartment for Seed Testing Austrian Agency for Health and Food Safety (AGES) Vienna Austria

6Institute for Sustainable Plant Production Department for Soil Health and Plant NutritionAustrian Agency for Health and Food Safety (AGES) Vienna Austria

7Institute of Soil Biology Biology Centre CAS Ceskeacute Budejovice Czechia+A full list of authors appears at the end of the paper

Correspondence Roey Angel (roeyangelbccascz) and Taru Sanden (tarusandenagesat)

Received 23 September 2021 ndash Discussion started 21 October 2021Revised 8 January 2022 ndash Accepted 4 February 2022 ndash Published 11 March 2022

Abstract Including information about soil microbial communities into global decomposition models is criticalfor predicting and understanding how ecosystem functions may shift in response to global change Here wecombined a standardised litter bag method for estimating decomposition rates the Tea Bag Index (TBI) withhigh-throughput sequencing of the microbial communities colonising the plant litter in the bags Together withstudents of the Federal College for Viticulture and Fruit Growing Klosterneuburg Austria acting as citizenscientists we used this approach to investigate the diversity of prokaryotes and fungi-colonising recalcitrant(rooibos) and labile (green tea) plant litter buried in three different soil types and during four seasons with the aimof (i) comparing litter decomposition (decomposition rates (k) and stabilisation factors (S)) between soil typesand seasons (ii) comparing the microbial communities colonising labile and recalcitrant plant litter between soiltypes and seasons and (iii) correlating microbial diversity and taxa relative abundance patterns of coloniserswith litter decomposition rates (k) and stabilisation factors (S) Stabilisation factor (S) but not decompositionrate (k) correlated with the season and was significantly lower in the summer indicating a decomposition of alarger fraction of the organic material during the warm months This finding highlights the necessity to includecolder seasons in the efforts of determining decomposition dynamics in order to quantify nutrient cycling in soilsaccurately With our approach we further showed selective colonisation of plant litter by fungal and prokaryotictaxa sourced from the soil The community structures of these microbial colonisers differed most profoundlybetween summer and winter and selective enrichment of microbial orders on either rooibos or green tea hintedat indicator taxa specialised for the primary degradation of recalcitrant or labile organic matter respectively Ourresults collectively demonstrate the importance of analysing decomposition dynamics over multiple seasons andfurther testify to the potential of the microbiome-resolved TBI to identify the active component of the microbialcommunity associated with litter decomposition

This work demonstrates the power of the microbiome-resolved TBI to give a holistic description of the litterdecomposition process in soils

Published by Copernicus Publications on behalf of the European Geosciences Union

164 A Daebeler et al Pairing litter decomposition with microbial community structures

1 Introduction

Litter decomposition is one of the most important terrestrialecosystem functions Soil microorganisms drive this processby breaking down plant material leading to the release ofcarbon into the atmosphere as CO2 or into the soil whereit either gets sequestered or further degraded (Talbot andTreseder 2011 Allison et al 2013 Cotrufo et al 2013)Other factors governing litter decomposition include themolecular structure of the litter its physical (dis)connectionfrom the decomposer community and its organondashmineralassociations (Schmidt et al 2011) Litter decompositiontherefore directly affects both Earthrsquos atmosphere and soilhealth Understanding this process holds the key to betteragricultural management mitigating greenhouse gas emis-sions and predicting future soil carbon storage levels How-ever because of the processesrsquo complexity and our inade-quate understanding of how biodiversity affects it the ap-plication of the litter decomposition process to agriculturalpractices and global decomposition models is not perform-ing currently at its full potential Seasonality mainly throughtemperature and moisture effects has been demonstrated toaffect litter decomposition directly and indirectly (Prescott2010) Direct effects occur due to the high sensitivity of bio-logical processes to temperature and water availability andindirect effects may be the consequence of phenotypic orcommunity shift responses of the soil biota Soil microbialcommunity structure has been hypothesised to influence pro-cess rates in soils (McGuire and Treseder 2010 Graham etal 2016) In a few models that linked microbial diversityand decomposition (eg MIMICS Wieder et al 2014 andMEMS Cotrufo et al 2013) microbial diversity positivelyaffected nutrient-cycling efficiency and ecosystem processesthrough either greater intensity of microbial exploitation oforganic matter or functional niche complementarity

Litter quality has been shown to be important for microbialcommunity structure since both bacteria and fungi respondto litter physicochemical changes during the decay process(Aneja et al 2006 Purahong et al 2016) Litter biochemi-cal traits such as the CN ratio and the fraction of lignin areconsidered good indicators of litter quality as they are relatedto nutrient availability and decomposition stage (Prescott2010 Talbot and Treseder 2012) Different litter types canthus select microbial taxa that are more specialised in degrad-ing their components Fungi are known to produce a suite ofoxidative enzymes that degrade the recalcitrant biopolymersof litter (Mathieu et al 2013 Hoppe et al 2015) In contrastonly a few types of bacteria degrade all lignocellulosic poly-mers while most typically target simple soluble compounds(de Boer and van der Wal 2008) Therefore the role of bacte-ria in the decomposition of more recalcitrant material is stilldebated (Wilhelm et al 2019) However still little is knownon how microbial communities specialise in litter types with

different physical and chemical traits (Freschet et al 2011Pioli et al 2020)

The community structure and functioning of microbialcommunities are further affected by biotic interactions (Dae-beler et al 2014 Garbeva et al 2014 Ho et al 2016Wilpiszeski et al 2019) In natural communities such in-teractions generally involve competition for space and re-sources (Boddy 2000 and Faust and Raes 2012 and ref-erences therein) but may also be mutualistic For example ithas been suggested that bacteria can facilitate the activity ofdecaying fungi by providing important nutrients such as ni-trogen (N) and phosphorus (P) (Purahong et al 2016) Like-wise fungi have been demonstrated to help improve the ac-cessibility of litter for bacteria (de Boer et al 2005) There-fore it is likely that decomposition dynamics depend notonly on microbial community structure but also on the facil-itative or competitive interactions among microbial species(Liang et al 2017 Aleklett et al 2021) Despite the indica-tions of the high importance of species interactions for soilorganic matter decomposition dynamics they are not wellunderstood and more studies under natural conditions areneeded

Linking microbial diversity to function across different en-vironments represents a key aspect of ecology In this studywe used high-throughput sequencing in combination withthe Tea Bag Index (TBI) ndash a cost-effective method to studylitter decomposition using commercially available teabags(Keuskamp et al 2013) This combination of methods al-lows insight to be gained into the effect of litter traits on thecommunity structure of microbial decomposers and soil or-ganic matter decomposition rates to be linked with microbialpopulation dynamics The microbial diversity in soils is im-mense (Thompson et al 2017) yet large parts of it are dor-mant (Lennon and Jones 2011) or simply do not participatein litter degradation (Falkowski et al 2008) Therefore theburied teabags used in this study serve as ldquotrapsrdquo for micro-bial litter degraders and assure that the diversity we observeis composed only of its active litter degrading and associatedtaxa

Specifically in this citizen-science-aided project we usedbarcoded-amplicon high-throughput sequencing of the SSUrRNA gene (for bacteria and archaea) and the internal tran-scribed spacer region (ITS) for fungi to compare the micro-bial community structure in two different standardised littertypes (rooibos and green tea) with the microbial communityin three different local soil types and during four seasonsover the course of a year We tested the extent to which littertypes with different traits represent selective substrates formicrobial community colonisation Finally we related themicrobial diversity and species relative abundance patternswith two proxies (decomposition rate (k) and stabilisationfactor (S)) that describe the decomposition of labile mate-rial (Keuskamp et al 2013) We generally assumed that lit-

SOIL 8 163ndash176 2022 httpsdoiorg105194soil-8-163-2022

A Daebeler et al Pairing litter decomposition with microbial community structures 165

ter decomposition dynamics and soil microbial communitieswill differ between the seasons Soil types were expected tobe similar to each other given the small geographic scaleand given that management practices are the same Morespecifically we hypothesised that (i) microbial diversity inthe teabags will correlate positively with decomposition ratesand negatively with stabilisation factors and that (ii) differ-ent subsets of local soil microbiota will colonise labile andmore recalcitrant litters and thus each litter type will se-lect a different specialised community To the best of ourknowledge this is the first study utilising the Tea Bag In-dex method in combination with a microbiome analysis toinvestigate how litter decomposition is linked with microbialcommunity structure and diversity

2 Materials and methods

21 Description of the study sites

The study site is located at the Agneshof vineyards ofthe agricultural college for Viticulture and Pomology inKlosterneuburg just north of Vienna Austria (Fig 1) Themean annual temperature was 111 C and the mean totalannual precipitation was 689 mm between 2010 and 2019 atthe Agneshof weather station We selected three study plotswith three different soil types Fluvisol Cambisol and Luvi-sol (Anjos et al 2015) to maximise the difference in soilcharacteristics The Fluvisol and Luvisol sites were culti-vated with wine whereas the Cambisol site was a set-asidegrassland between vineyards At Fluvisol and Luvisol siteswe selected one inter-row sampling area in the middle of thevineyard at least 5 m distance from the vineyard edge In con-trast at the Cambisol we selected a sampling area that wasat least 5 m distance from the neighbouring vineyard

22 Soil sampling and soil chemical analyses

Composite soil samples of 10ndash12 individual soil cores weretaken from 0ndash10 cm depth in two (winter and spring) orthree (summer and autumn) field replicates every 3 monthsbetween March 2018 and December 2018 to characterisethe sites Soils were sieved through a 2 mm stainless sieveand air-dried prior to further analyses Soil pH was mea-sured electrochemically (pHmV Pocket Meter pH 340iWTW) in 001 M CaCl2 at a soil-to-solution ratio of 1 5(OumlNORM L1083) Plant-available phosphorus (P) and potas-sium (K) were determined by calciumndashacetatendashlactate (CAL)extraction (OumlNORM L1087) Total soil organic C concen-trations of the soil samples were analysed by dry combus-tion in a LECO RC-612 TruMac CN at 650 C (OumlNORML1080 LECO Corp) Total N was determined accordingto OumlNORM L1095 with elemental analysis using a CNS(carbon nitrogen sulfur) 2000 SGA-410ndash06 at 1250 CKMnO4 determination of labile carbon was analysed accord-ing to Tatzber et al (2015) Potential nitrogen mineralisation

was measured by the anaerobic incubation method (Keeney1982) as modified according to Kandeler (1993) The soiltexture was determined according to OumlNORM L1061-1 andL1062-2 For testing statistical differences in the chemicalparameters between plots samples from different seasonswere pooled

23 Tea Bag Index (TBI)

The Tea Bag Index (ie the decomposition rate (k)and stabilisation factor (S)) was assessed accordingto Keuskamp et al (2013) between December 2017and December 2018 In short commercially availablegreen tea (EAN 87 22700 05552 5) and rooibos tea(EAN 87 22700 18843 8) in non-woven polypropylene meshbags produced by Lipton (Unilever) were used as standard-ised litter bags Green and rooibos teabags in eight repli-cate pairs (four replicate pairs for calculating the TBI param-eters and four replicate pairs for molecular analysis) wereweighed and buried pairwise at a depth of 8 cm for 3 months(Winter December 2017ndashMarch 2018 spring March 2018ndashJune 2018 summer June 2018ndashSeptember 2018 autumnSeptember 2018ndashDecember 2018) There was a 15 cm dis-tance between the green and rooibos teabags within a repli-cate pair and at least 75 cm between the replicate pairs Sub-sequently the teabags used for calculating the TBI were re-trieved cleaned of adhering soil particles and dried for atleast 3 d on a warm dry location before reweighing Valuesfor k and S were calculated using the mass losses of greenand rooibos teas as described in Keuskamp et al (2013)The teabags used for molecular analysis were frozen on-siteon dry ice after being dug out and cleaned and transportedin a frozen state to the Anaerobic and Molecular Microbiol-ogy lab at SoWa BC CAS Ceskeacute Budejovice Czechia forfurther analysis

3 Molecular analyses of prokaryotic and fungalcommunity structures

31 DNA extraction and amplicon sequencing

DNA was extracted from 025 g of frozen tea material orsoil using the DNeasy PowerSoil DNA Isolation Kit (Qi-agen) according to the manufacturerrsquos instructions Am-plicon sequencing was done using a two-step barcodingapproach (Naqib et al 2019) DNA was also extractedfrom two unburied green tea and rooibos teabags to es-timate the microbial load and diversity already presenton the tea material The DNA extracts were then quanti-fied using a Quant-iTtrade PicoGreentrade dsDNA Assay Kit(Thermo) and diluted to 10 ng microLminus1 for use as templatesfor PCR amplification The V4 region of the 16S rRNAgene (16S hereafter) was amplified using primers 515F-mod-CS1 (aca ctg acg aca tgg ttc tac aGT GYC AGC MGCCGC GGT AA) and 806-mod-CS2 (tacggtagcagagacttg-

httpsdoiorg105194soil-8-163-2022 SOIL 8 163ndash176 2022


Figure 1 Geographical map of the investigated sites in Klosterneuburg Austria The study sites are marked by black squares copy INVEKOSDATA 2019 Agrarmarkt Austria als Geodatenstelle Background httpsbasemapat (last access 9 February 2021)

gtctGGACTACNVGGGTWTCTAAT Walters et al 2016)while ITS was amplified using primers ITS1f-CS1 (acact-gacgacatggttctacaCTTGGTCATTTAGAGGAAGTAA) andITS2-CS2 (tacggtagcagagacttggtctGCTGCGTTCTTCATC-GATGC Gardes and Bruns 1993) Amplification was donein a T100 thermal cycler (Biorad) with amplification cyclesranging from 25 to 28 for 16S and 29 to 32 cycles for ITSdepending on the amount of template The full protocol isavailable online (Angel 2021) Before processing the sam-ples from the experiment we performed a preliminary testusing the chloroplast blocker pPNA (PNA Bio) For this pur-pose DNA from the unburied teabags two buried teabagsand a soil sample was used for amplification with or with-out PNA (025 microM final conc) If PNA was used an ad-ditional PCR step of PNA clamping (75 C for 10 s) wasadded in each cycle between the denaturation and primerannealing step according to the manufacturerrsquos instructionsIn addition no-template PCR control (NTC) and ldquoblank ex-traction controlrdquo (DNA extraction and amplification with-out a sample) were sequenced in each batch (season) Amock community (ZymoBIOMICS Microbial Community

DNA Standard II Zymo Research) was also amplified andsequenced Library construction and sequencing were per-formed at the DNA Services Facility at the University of Illi-nois Chicago using an Illumina MiniSeq sequencer (Illu-mina) in the 2times 250 cycle configuration (V2 reagent kit)

32 Sequence data processing and classification

Primer regions were trimmed off the amplicon sequencedata using cutadapt (V23 Martin 2011) Downstream se-quencing processing steps were done in R (V403 R CoreTeam 2020) Quality trimming and clustering into ampli-con sequence variants (ASVs) were done using the DADA2pipeline (Callahan et al 2016) For 16S the followingquality filtering options were used no truncate maxN= 0maxEE= c(2 2) and truncQ= 2 For ITS the followingoptions were used minLen= 50 maxN= 0 maxEE= c(22) and truncQ= 2 Chimera sequences were removed withremoveBimeraDenovo() using the ldquoconsensusrdquo method ldquoal-lowOneOffrdquo Taxonomic classification of the 16S ASVs wasdone with assignTaxonomy() against the SILVA database(Ref NR 99 V1381 Quast et al 2012) while for ITS it



was done against the UNTIE database (Nilsson et al 2018)Potential contaminant ASVs were removed using decontam(Davis et al 2018) employing the default options Unclassi-fied taxa and those classified as either ldquoeukaryotardquo ldquochloro-plastrdquo or ldquomitochondriardquo (in the 16S dataset) or as ldquobacteriardquoor ldquoarchaeardquo (in the ITS dataset) were removed In additionfor beta-diversity analysis ASVs appearing in lt 5 of thesamples were removed

33 Statistical analysis

All statistical analysis was performed in R (V403 R CoreTeam 2020) Significant differences between the decompo-sition rates (k) stabilisation factors (S) and the chemical pa-rameters were determined using ANOVA followed TukeyrsquosHSD on the estimated marginal means (Lenth 2021) func-tions emmeans() and contrast()) To increase the statisticalpower samples from different seasons were pooled for test-ing the differences in soil parameters Spearman rank (ρ) cor-relations were performed to investigate the relationship be-tween soil properties and the decomposition rates stabilisa-tion factors and microbial amplicon sequence variant (ASV)richness and diversity Sequence data handling and manip-ulation were done using the phyloseq package (McMurdieand Holmes 2013) For alpha-diversity analysis all sampleswere subsampled (rarefied) to the minimum sample size us-ing a bootstrap subsampling with 1000 iterations to accountfor library size differences while for beta-diversity analy-sis library size normalisation was done by converting thedata to relative abundance and multiplying by median se-quencing depth The inverse Simpson index the ShannonH diversity index and the BergerndashParker dominance indexwere calculated using the function EstimateR() in the veganpackage (Oksanen et al 2018) and tested using ANOVAin the stats package followed by a post hoc Tukey HSDtest on the estimated marginal means Variance partition-ing and testing were done using the PERMANOVA (Mcar-dle and Anderson 2001) function adonis() using HornndashMorisita distances (Horn 1966) Differences in order com-position between the soil or litter types were tested similarlyto STAMP (Parks et al 2014) but written in R Briefly therelative abundance of the orders was compared between sam-ples using the KruskalndashWallis rank-sum test (kruskaltest()in the stats package) followed by the MannndashWhitney posthoc test (wilcoxtest() in the stats package) and false dis-covery rate (FDR) corrected using the BenjaminindashHochbergmethod (Benjamini and Hochberg 1995) padjust() in thestats package) Differentially abundant ASVs were detectedusing ALDEx2 (Fernandes et al 2013) using the option de-nom= iqlr for the aldexclr() function Differences in compo-sition between the two teabag types and the soil were testedin a pairwise manner Plots were generated using ggplot2(Wickham 2016)

4 Results and discussion

41 Soil chemical characteristics and Tea Bag Index

The weather during the study was typical for the region withmean seasonal air temperatures ranging from 2 C in the win-ter to 225 C in the summer and most of the rain occurringsummer and autumn (Table S1 in the Supplement)

To be able to draw conclusions about the connections be-tween the community structures of microbial litter decom-posers litter types and the decomposition process we firstdetermined the basic soil characteristics of the three studyplots The Luvisol plot exhibited the lowest contents of P KTOC and sand and the highest pH (Table S2)

Next we determined the decomposition rate k (which re-flects the rate by which the labile fraction of the litter is de-composed) and the stabilisation factor S (which reflects theproportion of non-decomposed hydrolysable labile fractionthat is remaining after the incubation) for all three studyplots and during four seasons Despite decomposition ratesranging from 0008 dminus1 at the Cambisol site to 0025 dminus1

at the Fluvisol site (in the summer) no significant differ-ences could be found between sites or seasons neither whenpooled (P = 015) nor at each site (P = 045 Fig 2) Theresults are well within the range of summer data previouslyreported from Austria (Buchholz et al 2017 Keuskamp etal 2013 Sandeacuten et al 2020 2021) However to the best ofour knowledge we are the first ones to report data during fourseasons There was also no correlation of any measured soilproperties with the decomposition rates (data not shown) asalso reported in Sandeacuten et al (2020) These results were sur-prising given the differences in the determined soil charac-teristics (Table S2) and temperature between the seasons andpoint to the possibility that micro conditions in the soil play alarger role than expected in determining decomposition rates

The stabilisation factors in general were higher than pre-sented in Keuskamp et al (2013) however this is the firsttime that the stabilisation factors are presented for differentseasons and it seems logical that the S would be higher inthe colder seasons than over the summer The summer val-ues are well in agreement with Keuskamp et al (2013) andthe other seasons with Sarneel et al (2020) from tundra envi-ronments and Sandeacuten et al (2021) from urban Austrian sitesthat both reported stabilisation factors up to 04 which givesfurther support for our findings We did however observesignificantly lower stabilisation factors in summer than inall other seasons (padjusted le 0001) Specifically the stabil-isation factors determined at the Cambisol and Luvisol siteswere significantly lower in summer than those determined forthe other three seasons (padjusted le 005) These results indi-cate that the conditions during the summer months favourthe decomposition of a larger fraction of the organic mate-rial but not necessarily in a faster manner This is in apparentcontrast to the well-known relationship between temperatureand decomposition rates (see Kirschbaum 1995 and refer-



Figure 2 Decomposition rate (k) and stabilisation factor (S) determined at three sites with different soil types during four seasons Valuesshown are averages (n= 4 lower in case a teabag broke during incubation or in case the rooibos was not in the first phase of decompositionanymore and the main assumptions for calculation of k were violated thus k could not be calculated) and the interquartile range (violinshapes) Lowercase letters indicate significant differences between samples (seasons and sites together padjusted le 005)

ences therein) However one should consider that (1) tem-perature sensitivity of the decomposition rate is much higherat low temperature and that (2) the decomposition rate is alsoheavily affected by moisture which could counterbalance theeffect of temperature Similarly Elumeeva et al (2018) alsofound a correlation of S but not of k to various edaphic fac-tors Notably soil moisture pH and altitude did not affect k(but were correlated with S) Since S measures ldquohow muchrdquo(rather than ldquohow fastrdquo) of the labile litter fraction is sta-bilised material is remaining after incubation it can be as-sumed that S is more sensitive than k to the composition ofmicrobial communities that are active at the time of decom-position This is because the underlying processes amountingto S involve the concerted (discrete) action of several micro-bial guilds each specialising in decomposing one or moresubstrates sequentially (Zheng et al 2021 Glassman et al2018)

42 Establishment of an amplicon sequencing protocolto be used with the TBI

Before investigating the microbial communities colonisingthe teabags we extracted DNA from unburied teabags serv-ing as negative controls and performed rRNA gene ampli-con sequencing (16S and ITS regions) DNA yields weresimilar between buried and the unburied teabags and aver-aged 61plusmn 129 and 230plusmn 140 mg gminus1 probably becauseas expected most of the DNA originated from the plant cells(data not shown) A preliminary study using the chloroplast

blocker pPNA was carried out to estimate the effect of plantchloroplast on the sequencing output (what proportion of thesequences is dominated by chloroplasts) As Table S3 showsonly the unburied tea material had a significant portion ofchloroplast reads (645 and 154 for green tea and rooi-bos respectively) while after 3 months in the ground nochloroplast reads were detected (most likely due to a combi-nation of microbial colonisation and chloroplast DNA degra-dation) Using the pPNA blocker nearly eliminated the num-ber of chloroplast reads while at the same time not biasing thecommunity composition in the samples (Fig S1 in the Sup-plement) Nevertheless since chloroplasts were undetectedin the buried teabags pPNA was not used in subsequent sam-ple processing Figure S1 also shows that as expected themicrobial load on the unburied tea material was minimal andyielded only about 5 of the reads after quality filteringSuch a low microbial load is expected for above-ground plantparts (eg Chen et al 2020 Knorr et al 2019)

To then study the prokaryotic and fungal communitiespotentially associated with litter degradation we extractedDNA from soil and triplicate rooibos and green tea teabagsburied for 3 months for each season at each site and per-formed rRNA gene amplicon sequencing

43 Richness and diversity of prokaryotic and fungalcommunities in teabags and soils

Sequencing the prokaryotic 16S rRNA gene and the fungalITS region and downstream data processing yielded a total



of 16 100 16S ASVs and 3685 ITS ASVs After removingall ASVs with a prevalence of lt 5 of the samples 421716S ASVs and 402 ITS ASVs remained As expected ar-chaea were rare and comprised only about 09 of the readsAt all three sites the observed prokaryotic and fungal ASVrichness (ie the observed number of prokaryotic and fungalASVs) was roughly twice as high in the soil samples than inteabag samples (Fig S2) Likewise the microbial diversityas estimated by the inverse Simpson Shannon and BergerndashParker indexes was generally significantly higher in the soilsamples than in the teabag samples There was however nosignificant difference between rooibos and green teabags interms of prokaryotic and fungal ASV richness or diversityThese findings indicate selective microbial colonisation ofboth the labile and the more recalcitrant litter sourced fromthe surrounding soil during all seasons

Both the Fluvisol and the Cambisol site teabags buried insummer harboured a significantly richer prokaryotic commu-nity than those buried in other seasons Such seasonal differ-ences were not observed with the fungal communities Sim-ilar to the observed differences in richness the biodiversityof prokaryotic communities detected in the teabags was of-ten significantly higher in summer samples than in all otherseasons This hints that there are more distinct prokaryoticspecies capable of colonising and degrading litter in sum-mer than in the other seasons An explanation for this obser-vation could be the activation of decomposing prokaryoticsoil populations by summer conditions such as elevated tem-perature (Kirschbaum 1995 Allison et al 2010) Finallythe prokaryotic ASV richness and diversity of the commu-nities that colonised the teabags were negatively correlatedwith the stabilisation factor S (p lt 001 correlation factorρ =minus060 and p lt 001 correlation factor ρ =minus054 re-spectively) These findings suggest that less inhibition of lit-ter decomposition occurred in the summer season in whichricher and more diverse prokaryotic communities were asso-ciated with the degradation With these data we can partiallyconfirm our hypothesis of a positive correlation between mi-crobial diversity and litter decomposition Possibly a positiverelationship between microbial diversity and litter decompo-sition is prevalent in many soil types as it has also been ex-perimentally established in a Cambisol (Maron et al 2018)

The richness and diversity of fungal ASVs were generallycomparable across the seasons at all sites except for a higherrichness at the Cambisol site in winter compared to the otherseasons This could be explained by the hyphae growth formof many fungi which allows them to span various micrositesin the heterogeneous soil environment and makes them lesssensitive to chemical gradients and seasonal changes (Yusteet al 2010)

44 Selective colonisation of teabags by prokaryotic andfungal soil populations

After 3 months of burial the community structures ofprokaryotes and fungi detected in teabags were significantlydifferent from those detected in soil (Table S4 Fig 3) More-over the differential abundance analysis clearly showed en-richment of about a third of the 16S ASVs and 14 ndash20 ofthe ITS ASVs in the teabags compared to the soil (Fig S4)Therefore as has been observed before in various soil sys-tems and with a range of different litter types (Aneja et al2006 Bray et al 2012 Albright and Martiny 2018 Yanet al 2018 Wei et al 2020) we conclude that the littermaterial in the teabags selected for colonisation by a size-able portion of the bacterial archaeal and fungal popula-tion acting as the active litter decomposers and associatedpopulations While fungi colonise new patches using hyphalgrowth prokaryotes depend on mechanisms such as adhe-sion to hyphae motility or passive diffusion in water Inter-estingly Albright et al (2019) have shown that unicellularbacteria and fungi do not differ in their dispersal abilitiesSuccessful colonisation of the teabags is related to microbialtraits such as growth rate mobility and adhesive and com-petitive ability (Albright and Martiny 2018) and does notexclusively indicate the degrading ability of a microbe Theresults presented here therefore may be biased towards themore fast-growing strong competitors which disperse easilyand may overlook contributions of slower degraders in soilComparing the green tea with the rooibos samples we coulddetect a subgroup of ca 42 and 25 of the 16S rRNAgene ASVs that preferentially colonised either the green teaor rooibos teabags In contrast only four ITS ASVs differedbetween the green tea and rooibos samples indicating non-preferential colonisation For both prokaryotes and fungisoil or litter type explained the largest fraction of the vari-ance in the community composition (38 and 23 respec-tively Table S4) although not surprisingly this was drivenmainly by the differences between soil and teabag commu-nities However the difference between green tea and rooi-bos was not negligible and explained 13 and 9 of thevariance in a pairwise comparison for prokaryotes and fungirespectively (Table S4 Fig 3) The season was also a ma-jor factor explaining 20 and 22 of the variance respec-tively and driven mostly by the difference between summerand winter (27 and 29 of the variance respectively Ta-ble S4 Fig S4) In contrast soil type had only little effectexplaining only 3 and 5 of the variance respectivelyThis contrasts with the recent findings of Pioli et al (2020)in which a large effect of soil type on the community com-position was reported though this is not surprising consid-ering that this study was performed on a very local scalewith identical climate and similar soil types and managementpractices

Further interested in the identity and possible colonisationpreferences of the microbial decomposers we compared the



Figure 3 Principal coordinate analyses plots of prokaryotic (a) andfungal (b) community compositions based on MorisitandashHorn dis-tances of 16S rRNA gene and ITS region ASVs derived from soilgreen tea and rooibos teabags The models were obtained using theformula distance matrix simfield+ sample typetimes season

relative abundance of dominant orders between the differentteabags and the soil types and between the teabag communi-ties in the summer vs winter in a pairwise manner This anal-ysis allowed us to identify specific bacterial and fungal lin-eages that exhibited distinct colonisation patterns and wereenriched in samples from the more labile green tea or themore recalcitrant rooibos bags In contrast to our hypothesis(different litter types will select different microbial colonis-ers) many of the detected microbes were either equally se-lected or deselected by both tea types

Most strikingly members of the Pseudomonadales andSphingobacteriales were the most enriched in both green teaand rooibos teabags as compared to the soil from the samesites with differences of 114 and 94 for Pseudomon-adales and 82 and 83 for Sphingobacteriales respec-tively (Fig 4a c) Additionally Flavobacteriales Micrococ-cales and Burkholderiales also showed higher relative abun-dances than the surrounding soil (Fig 4a) Similar increasesin relative abundance of Pseudomonadales FlavobacterialesSphingobacteriales and Micrococcales have been observedin straw decomposing soil microcosms (Jimeacutenez et al 2014Guo et al 2020) Members of Pseudomonadales Flavobac-teriales and Rhizobiales are well known to have the capacityto degrade plant lignin (hemi-)cellulose or carboxymethylcellulose (Koga et al 1999 McBride et al 2009 Wanget al 2013 Talia et al 2012 Jackson et al 2017) Fur-thermore Pseudomonadales and Flavobacteriales can be in-volved in the degradation of furanic compounds (Jimeacutenez etal 2013 Loacutepez et al 2004) The increased presence of Sph-ingobacteriales may be explained by their capability to pro-duce β-glucosidases that remove the cello-oligosaccharidesproduced by polymer degraders (Matsuyama et al 2008)and is considered a critical and rate-limiting step in cellu-lose degradation (du Plessis et al 2009) Indeed soils withhigh β-glucosidase activity were also found to be dominatedby members of the Sphingobacteriales (particularly from theChitinophagaceae family Bailey et al 2013) Merely thebacterial orders Cytophagales Enterobacteriales and Rhizo-biales exhibited relative abundance patterns indicative of aselection only by rooibos or green tea as we hypothesised(Fig 4a c)

We further observed significant selective enrichments offungal orders in both teabag types Members of the Hypocre-ales had a 20 higher relative abundance in green tea thanin soil but were not enriched in the rooibos teabags (Fig 4b)In contrast Helotiales were only enriched in rooibos teabagsby 12 (Fig 4d) Hypocreales are common saprophyticfungi and are known to produce cellobiohydrolases and en-doglucanases necessary for the depolymerisation of cellu-loses (Lynd et al 2002 Martinez et al 2008) Helotiales areabundant cellulolytic soil fungi that have been shown to spe-cialise in degrading recalcitrant organic carbon (Newsham etal 2018 MGnify 2020) Our findings could point towardsmembers of the exclusively enriched orders being specialisedprimary degraders of either labile of recalcitrant organic mat-ter and not associated scavengers of degradation productsSince the green tea litter contains a higher hydrolysable frac-tion and a lower C N ratio (Keuskamp et al 2013) we as-sume the Enterobacteriales and Hypocreales which were ex-clusively enriched in green tea samples were involved in thedegradation of more labile compounds such as cellulose Cy-tophagales Rhizobiales and Helotiales which were exclu-sively selected by rooibos tea on the other hand may be im-portant primary degraders of more recalcitrant organic mat-



Figure 4 Comparative analysis of the relative abundance based on bacterial and fungal order profiles between soil or litter types and seasonsSamples were compared using KruskalndashWallis rank-sum test followed by the MannndashWhitney post hoc test and FDR-corrected using theBenjaminindashHochberg method Only orders with a median relative abundance of gt 5 are shown



ter Members of these five orders could therefore possiblyserve as indicator species in future decomposition studies

5 Conclusions

In contrast to our expectations the different seasons did notdisplay different litter decomposition rates but they did dif-fer in the extent to which the hydrolysable litter fraction wasdegraded At the same time microbial richness and diversityof litter-colonising degraders were positively correlated withthe fraction of degraded organic carbon indicating a pos-itive relationship between litter degradation and microbialbiodiversity in soil Microbial colonisation of the tea litterwas substrate-selective and season-dependent for prokary-otes and fungi On average about 28 ndash30 of the prokary-otic ASVs and 14 ndash19 of the fungal ASVs preferentiallycolonised the rooibos and green tea respectively Howevertheir exact identity varied between seasons especially be-tween the summer and winter A total of five microbial orderswere identified as possible indicator species through their ex-clusive colonisation of either recalcitrant (rooibos) or labile(green tea) litter This work demonstrates the power of themicrobiome-resolved TBI to give a holistic description of thelitter decomposition process in soils

Code availability The scripts to reproduce this and all down-stream statistical analysis steps are available online underhttpsgithubcomroey-angelTeaTime4Schools (last access9 March 2022) and httpsdoiorg105281zenodo6340082(Angel 2022)

Data availability Raw sequence reads have been depositedin the Sequence Read Archive under BioProject accession noPRJNA765214 (httpswwwncbinlmnihgovbioprojectterm=PRJNA765214 NCBI 2022)

Supplement The supplement related to this article is availableonline at httpsdoiorg105194soil-8-163-2022-supplement

Team list Group I Fabian Bauer Lorenz Baumgartner LisaBrandl Michael Buchmayer Karl Daschl Rene Dietrich Ste-fan Ebner Margarete Jaumlger Michaela Kiss Lea Kneissl KatjaLangmann Elias Mauerhofer Jakob Paschinger Tobias Rabl So-phie Schachenhuber Josef Schmid Marlene Steinbatz Group IIMathias Ettenauer Lorenz Hauck Mathias Herl Johannes HonsigThomas Gangl Anja Glaser Gabriel Huumlmer Georg Lenzatti StefanLichtscheidl Lena Mayer Valentin Oppenauer Jacob RennhoferKatharina Schoumlnner Ciara Seywald Group III Jan GamzunovAnna Hess Christoph Schurm Klaus Stacher Lena UngersboumlckFlorian Valachovich Mirjam Weissmann Marius Wittek JenniferWosak Valentin Zahel Lucas Zuumlger

Author contributions TS RA and AD designed the study TSHB SG and EK performed sampling with support from students ofbiology projects groups I II and III EK developed lab protocolsand EP optimised the lab protocols and performed the DNA extrac-tion and PCR amplification TS RA and AD analysed the data Themanuscript was written by AD TS and RA with contributions fromall co-authors

Competing interests The contact author has declared that nei-ther they nor their co-authors have any competing interests

Disclaimer Publisherrsquos note Copernicus Publications remainsneutral with regard to jurisdictional claims in published maps andinstitutional affiliations

Acknowledgements We thank Michael Schwarz for preparingthe map in Fig 1

Financial support This work was supported by the Aus-trian Ministry of Education Science and Research (BMBWF)through a Sparkling Science Grant to Taru Sandeacuten He-lene Berthold Elena Kinz Anne Daebeler Roey Angel and Su-sanne Grausenburger (SPA 06044 TeaTime4Schools) In additionEva Petrovaacute and Roey Angel were supported by the Czech MEYS(EF16_0130001782 ndash SoWa Ecosystems Research) Anne Dae-beler was supported by the FWF grant T938

Review statement This paper was edited by Ingrid Lubbers andreviewed by Thomas Reitz and one anonymous referee

References

Albright M B N and Martiny J B H Dispersal alters bacterialdiversity and composition in a natural community ISME J 12296ndash299 httpsdoiorg101038ismej2017161 2018

Albright M B N Chase A B and Martiny J B H Experimen-tal Evidence that Stochasticity Contributes to Bacterial Compo-sition and Functioning in a Decomposer Community mBio 10e00568-19 httpsdoiorg101128mBio00568-19 2019

Aleklett K Ohlsson P Bengtsson M and Hammer EC Fungal foraging behaviour and hyphal space explorationin micro-structured Soil Chips ISME J 15 1782ndash1793httpsdoiorg101038s41396-020-00886-7 2021

Allison S D Wallenstein M D and Bradford M A Soil-carbonresponse to warming dependent on microbial physiology NatGeosci 3 336ndash340 httpsdoiorg101038ngeo846 2010

Allison S D Lu Y Weihe C Goulden M L Mar-tiny A C Treseder K K and Martiny J B H Micro-bial abundance and composition influence litter decomposi-tion response to environmental change Ecology 94 714ndash725httpsdoiorg10189012-12431 2013

Aneja M K Sharma S Fleischmann F Stich S HellerW Bahnweg G Munch J C and Schloter M Micro-



bial Colonization of Beech and Spruce LitterInfluence ofDecomposition Site and Plant Litter Species on the Diver-sity of Microbial Community Microb Ecol 52 127ndash135httpsdoiorg101007s00248-006-9006-3 2006

Angel R General Bacteria and Archaea 16S-rRNA (515Fmod-806Rmod) for Illumina Amplicon Sequencing V4 protocolsio[data set] httpsdoiorg1017504protocolsiobsxanfie 2021

Angel R roey-angelTeaTime4Schools First release ndash EGU SOIL(V10) Zenodo [code] httpsdoiorg105281zenodo63400822022

Bailey V L Fansler S J Stegen J C and McCue L A Linkingmicrobial community structure to β-glucosidic function in soilaggregates ISME J 7 2044ndash2053 2013

Benjamini Y and Hochberg Y Controlling the False DiscoveryRate A Practical and Powerful Approach to Multiple Testing JRoy Stat Soc B 57 289ndash300 httpsdoiorg101111j2517-61611995tb02031x 1995

Boddy L Interspecific combative interactions between wood-decaying basidiomycetes FEMS Microbiol Ecol 31 185ndash194httpsdoiorg101111j1574-69412000tb00683x 2000

Bray S R Kitajima K and Mack M C Temporal dynamicsof microbial communities on decomposing leaf litter of 10 plantspecies in relation to decomposition rate Soil Biol Biochem49 30ndash37 httpsdoiorg101016jsoilbio201202009 2012

Buchholz J Querner P Paredes D Bauer T Strauss PGuernion M Scimia J Cluzeau D Burel F Kratschmer SWinter S Potthoff M and Zaller J G Soil biota in vineyardsare more influenced by plants and soil quality than by tillage in-tensity or the surrounding landscape Sci Rep-UK 7 17445httpsdoiorg101038s41598-017-17601-w 2017

Callahan B J McMurdie P J Rosen M J Han A W JohnsonA J A and Holmes S P DADA2 High-resolution sampleinference from Illumina amplicon data Nat Meth 13 581ndash583httpsdoiorg101038nmeth3869 2016

Chen T Nomura K Wang X Sohrabi R Xu J Yao LPaasch B C Ma L Kremer J Cheng Y Zhang L WangN Wang E Xin X-F and He S Y A plant genetic networkfor preventing dysbiosis in the phyllosphere Nature 580 653ndash657 httpsdoiorg101038s41586-020-2185-0 2020

Cotrufo M F Wallenstein M D Boot C M Denef Kand Paul E The Microbial Efficiency-Matrix Stabilization(MEMS) framework integrates plant litter decomposition withsoil organic matter stabilization do labile plant inputs formstable soil organic matter Glob Change Biol 19 988ndash995httpsdoiorg101111gcb12113 2013

Daebeler A Bodelier P L E Yan Z Hefting M MJia Z and Laanbroek H J Interactions between Thaumar-chaea Nitrospira and methanotrophs modulate autotrophic ni-trification in volcanic grassland soil ISME J 8 2397ndash2410httpsdoiorg101038ismej201481 2014

Davis N M Proctor D M Holmes S P Relman D A andCallahan B J Simple statistical identification and removal ofcontaminant sequences in marker-gene and metagenomics dataMicrobiome 6 226 httpsdoiorg101186s40168-018-0605-2 2018

de Boer W and van der Wal A Chapter 8 Interactions betweensaprotrophic basidiomycetes and bacteria in British Mycolog-ical Society Symposia Series Vol 28 edited by Boddy L

Frankland J C and van West P Academic Press 143ndash153httpsdoiorg101016S0275-0287(08)80010-0 2008

de Boer W Folman L B Summerbell R C and BoddyL Living in a fungal world impact of fungi on soil bacte-rial niche development FEMS Microbiol Rev 29 795ndash811httpsdoiorg101016jfemsre200411005 2005

du Plessis L Rose S H and van Zyl W H Explor-ing improved endoglucanase expression in Saccharomycescerevisiae strains Appl Microbiol Biot 86 1503ndash1511httpsdoiorg101007s00253-009-2403-z 2009

Elumeeva T G Onipchenko V G Akhmetzhanova A AMakarov M I and Keuskamp J A Stabilization versus de-composition in alpine ecosystems of the Northwestern CaucasusThe results of a tea bag burial experiment J Mt Sci 15 1633ndash1641 httpsdoiorg101007s11629-018-4960-z 2018

Falkowski P G Fenchel T and Delong E F The Microbial En-gines That Drive Earths Biogeochemical Cycles Science 3201034ndash1039 httpsdoiorg101126science1153213 2008

Faust K and Raes J Microbial interactions from net-works to models Nat Rev Microbiol 10 538ndash550httpsdoiorg101038nrmicro2832 2012

Fernandes A D Macklaim J M Linn T G Reid G andGloor G B ANOVA-like differential expression (ALDEx)analysis for mixed population RNA-Seq PLoS One 8 e67019httpsdoiorg101371journalpone0067019 2013

Freschet G T Weedon J T Aerts R van Hal J Rand Cornelissen J H C Interspecific differences in wooddecay rates insights from a new short-term method tostudy long-term wood decomposition J Ecol 100 161ndash170httpsdoiorg101111j1365-2745201101896x 2011

Garbeva P Hordijk C Gerards S and de Boer WVolatile-mediated interactions between phylogeneti-cally different soil bacteria Front Microbiol 5 289httpsdoiorg103389fmicb201400289 2014

Gardes M and Bruns T D ITS primers with enhanced specificityfor basidiomycetesndashapplication to the identification of mycor-rhizae and rusts Mol Ecol 2 113ndash118 1993

Glassman S I Weihe C Li J Albright M B N Looby C IMartiny A C Treseder K K Allison S D and Martiny JB H Decomposition responses to climate depend on microbialcommunity composition P Natl Acad Sci USA 115 11994ndash11999 httpsdoiorg101073pnas1811269115 2018

Graham E B Knelman J E Schindlbacher A Siciliano SBreulmann M Yannarell A Beman J M Abell G Philip-pot L Prosser J Foulquier A Yuste J C Glanville H CJones D L Angel R Salminen J Newton R J BuumlrgmannH Ingram L J Hamer U Siljanen H M P PeltoniemiK Potthast K Bantildeeras L Hartmann M Banerjee S YuR-Q Nogaro G Richter A Koranda M Castle S CGoberna M Song B Chatterjee A Nunes O C LopesA R Cao Y Kaisermann A Hallin S Strickland M SGarcia-Pausas J Barba J Kang H Isobe K PapaspyrouS Pastorelli R Lagomarsino A Lindstroumlm E S BasilikoN and Nemergut D R Microbes as Engines of Ecosys-tem Function When Does Community Structure Enhance Pre-dictions of Ecosystem Processes Front Microbiol 7 214httpsdoiorg103389fmicb201600214 2016

Guo T Zhang Q Ai C Liang G He P Lei Q and ZhouW Analysis of microbial utilization of rice straw in paddy soil



using a DNA-SIP approach Soil Sci Soc Am J 84 99ndash114httpsdoiorg101002saj220019 2020

Ho A Angel R Veraart A J Daebeler A Jia Z Kim S YKerckhof F-M Boon N and Bodelier P L E Biotic Interac-tions in Microbial Communities as Modulators of Biogeochemi-cal Processes Methanotrophy as a Model System Front Micro-biol 7 1285 httpsdoiorg103389fmicb201601285 2016

Hoppe B Purahong W Wubet T Kahl T Bauhus JArnstadt T Hofrichter M Buscot F and Kruumlger DLinking molecular deadwood-inhabiting fungal diversity andcommunity dynamics to ecosystem functions and processesin Central European forests Fungal Divers 77 367ndash379httpsdoiorg101007s13225-015-0341-x 2015

Horn H S Measurement of Overlap in Compara-tive Ecological Studies Am Nat 100 419ndash424httpsdoiorg101086282436 1966

Jackson C A Couger M B Prabhakaran M Ramachan-driya K D Canaan P and Fathepure B Z Isolationand characterization ofRhizobiumsp strain YS-1r that degradeslignin in plant biomass J Appl Microbiol 122 940ndash952httpsdoiorg101111jam13401 2017

Jimeacutenez D J Korenblum E and van Elsas J D Novelmultispecies microbial consortia involved in lignocelluloseand 5-hydroxymethylfurfural bioconversion Appl MicrobiolBiot 98 2789ndash2803 httpsdoiorg101007s00253-013-5253-7 2013

Jimeacutenez D J Dini-Andreote F and van Elsas J Metataxo-nomic profiling and prediction of functional behaviour of wheatstraw degrading microbial consortia Biotechnol Biofuels 7 92httpsdoiorg1011861754-6834-7-92 2014

Kandeler E Bestimmung der N-Mineralisation im anaerobenBrutversuch in Bodenbiologische Arbeitsmethoden 2nd Edn160ndash161 edited by Schinner F Oumlhlinger R Kandeler E andMargesin R Springer Verlag Berling Heidelberg Germany1993

Keeney D R Nitrogen availability indices in Methods of SoilAnalysis Part II 2nd Edn edited by Page A L Millet RH and Keeney D R Agronomy Monograph No 9 711ndash730American Society of Agronomy Madison USA 1982

Keuskamp J A Dingemans B J J Lehtinen T Sarneel J Mand Hefting M M Tea Bag Index a novel approach to col-lect uniform decomposition data across ecosystems Meth EcolEvol 4 1070ndash1075 httpsdoiorg1011112041-210x120972013

Kirschbaum M U F The temperature dependence of soil or-ganic matter decomposition and the effect of global warmingon soil organic C storage Soil Biol Biochem 27 753ndash760httpsdoiorg1010160038-0717(94)00242-s 1995

Knorr K Joslashrgensen L N and Nicolaisen M Fungi-cides have complex effects on the wheat phyl-losphere mycobiome PLOS ONE 14 e0213176httpsdoiorg101371journalpone0213176 2019

Koga S Ogawa J Choi Y-M and Shimizu S Novel bac-terial peroxidase without catalase activity from Flavobacteriummeningosepticum purification and characterization BiochimBiophys Acta 1435 117ndash126 httpsdoiorg101016s0167-4838(99)00190-9 1999

Lennon J T and Jones S E Microbial seed banks the ecologicaland evolutionary implications of dormancy Nat Rev Microbiol9 119ndash130 2011

Lenth R V emmeans Estimated Marginal Means aka Least-Squares Means httpsCRANR-projectorgpackage=emmeans(last access 9 March 2022) 2021

Liang C Schimel J P and Jastrow J D The importance of an-abolism in microbial control over soil carbon storage Nat Mi-crobiol 2 17105 httpsdoiorg101038nmicrobiol20171052017

Loacutepez M J Nichols N N Dien B S Moreno J and BothastR J Isolation of microorganisms for biological detoxification oflignocellulosic hydrolysates Appl Microbiol Biotechnol 64125ndash131 httpsdoiorg101007s00253-003-1401-9 2004

Lynd L R Weimer P J van Z W H and Pretorius I S Micro-bial cellulose utilization fundamentals and biotechnology Mi-crobiol Mol Biol Rev 66 506ndash577 2002

Maron P-A Sarr A Kaisermann A Leacutevecircque J Mathieu OGuigue J Karimi B Bernard L Dequiedt S Terrat SChabbi A and Ranjard L High microbial diversity promotessoil ecosystem functioning Appl Environ Microb 84 e02738-17 httpsdoiorg101128AEM02738-17 2018

Martin M Cutadapt removes adapter sequences from high-throughput sequencing reads EMBnetjournal 17 10ndash12httpsdoiorg1014806ej171200 2011

Martinez D Berka R M Henrissat B Saloheimo M ArvasM Baker S E Chapman J Chertkov O Coutinho P MCullen D Danchin E G J Grigoriev I V Harris P Jack-son M Kubicek C P Han C S Ho I Larrondo L F deLeon A L Magnuson J K Merino S Misra M Nelson BPutnam N Robbertse B Salamov A A Schmoll M TerryA Thayer N Westerholm-Parvinen A Schoch C L YaoJ Barabote R Barbote R Nelson M A Detter C BruceD Kuske C R Xie G Richardson P Rokhsar D S Lu-cas S M Rubin E M Dunn-Coleman N Ward M andBrettin T S Genome sequencing and analysis of the biomass-degrading fungus Trichoderma reesei (syn Hypocrea jecorina)Nat Biotechnol 26 553ndash560 httpsdoiorg101038nbt14032008

Mathieu Y Gelhaye E Dumarccedilay S Geacuterardin P Har-vengt L and Bueacutee M Selection and validation of enzy-matic activities as functional markers in wood biotechnol-ogy and fungal ecology J Microbiol Meth 92 157ndash163httpsdoiorg101016jmimet201211017 2013

Matsuyama H Katoh H Ohkushi T Satoh A Kawahara Kand Yumoto I Sphingobacterium kitahiroshimense sp novisolated from soil Int J Syst Evol Micr 58 1576ndash1579httpsdoiorg101099ijs065791-0 2008

McBride M J Xie G Martens E C Lapidus A HenrissatB Rhodes R G Goltsman E Wang W Xu J HunnicuttD W Staroscik A M Hoover T R Cheng Y-Q and SteinJ L Novel Features of the Polysaccharide-Digesting GlidingBacterium Flavobacterium johnsoniae as Revealed by GenomeSequence Analysis Appl Environ Microbiol 75 6864ndash6875httpsdoiorg101128aem01495-09 2009

McGuire K L and Treseder K K Microbial communi-ties and their relevance for ecosystem models Decompo-sition as a case study Soil Biol Biochem 42 529ndash535httpsdoiorg101016jsoilbio200911016 2010



McMurdie P J and Holmes S phyloseq An R Pack-age for Reproducible Interactive Analysis and Graph-ics of Microbiome Census Data PLoS ONE 8 e61217httpsdoiorg101371journalpone0061217 2013

Mcardle B and Anderson M Fitting Multivariate Models toCommunity Data A Comment on Distance-Based RedundancyAnalysis Ecology 82 290ndash297 httpsdoiorg1018900012-9658(2001)082[0290fmmtcd]20co2 2001

Naqib A Poggi S and Green S J Deconstructing the poly-merase chain reaction II an improved workflow and effectson artifact formation and primer degeneracy PeerJ 7 e7121httpsdoiorg107717peerj7121 2019

NCBI Teabag microbiome ndash Klosterneuburg Targeted loci environ-mental NCBI [data set sample] httpswwwncbinlmnihgovbioprojectterm=PRJNA765214 last access 9 March 2022

Newsham K K Garnett M H Robinson C H and CoxF Discrete taxa of saprotrophic fungi respire different agesof carbon from Antarctic soils Sci Rep-UK 8 7866httpsdoiorg101038s41598-018-25877-9 2018

Nilsson R H Larsson K-H Taylor A F S Bengtsson-PalmeJ Jeppesen T S Schigel D Kennedy P Picard K GloumlcknerF O Tedersoo L Saar I Kotildeljalg U and Abarenkov K TheUNITE database for molecular identification of fungi handlingdark taxa and parallel taxonomic classifications Nucleic AcidsResearch 47 D259ndashD264 httpsdoiorg101093nargky10222018

Oksanen J Blanchet F G Friendly M Kindt R Legendre PMcGlinn D Minchin P R OrsquoHara R B Simpson G LSolymos P Stevens M H H Szoecs E and Wagner H Ve-gan Community Ecology Package httpsCRANR-projectorgpackage=vegan (last access 9 March 2022) 2018

Parks D H Tyson G W Hugenholtz P and BeikoR G STAMP statistical analysis of taxonomic andfunctional profiles Bioinformatics 30 3123ndash3124httpsdoiorg101093bioinformaticsbtu494 2014

Pioli S Sarneel J Thomas H J D Domene X AndreacutesP Hefting M Reitz T Laudon H Sandeacuten T PiscovaacuteV Aurela M and Brusetti L Linking plant litter micro-bial diversity to microhabitat conditions environmental gradi-ents and litter mass loss Insights from a European study us-ing standard litter bags Soil Biol Biochem 144 107778httpsdoiorg101016jsoilbio2020107778 2020

Prescott C E Litter decomposition what controls it and how canwe alter it to sequester more carbon in forest soils Biogeochem-istry 101 133ndash149 httpsdoiorg101007s10533-010-9439-02010

Purahong W Wubet T Lentendu G Schloter M Pecyna M JKapturska D Hofrichter M Kruumlger D and Buscot F Lifein leaf litter novel insights into community dynamics of bacte-ria and fungi during litter decomposition Mol Ecol 25 4059ndash4074 httpsdoiorg101111mec13739 2016

Quast C Pruesse E Yilmaz P Gerken J Schweer TYarza P Peplies J and Gloumlckner F O The SILVA ri-bosomal RNA gene database project improved data process-ing and web-based tools Nucleic Acids Res 41 D590ndashD596httpsdoiorg101093nargks1219 2012

R Core Team R A Language and Environment for Statistical Com-puting httpwwwR-projectorg (last access 28 June 2021)2020

Sandeacuten T Spiegel H Wenng H Schwarz M and SarneelJ M Learning Science during Teatime Using a Citi-zen Science Approach to Collect Data on Litter Decom-position in Sweden and Austria Sustainability 12 7745httpsdoiorg103390su12187745 2020

Sandeacuten T Wawra A Berthold H Miloczki J Schweinzer AGschmeidler B Spiegel H Debeljak M and Trajanov ATeaTime4Schools Using Data Mining Techniques to Model Lit-ter Decomposition in Austrian Urban School Soils Front EcolEvol 9 703794 httpsdoiorg103389fevo20217037942021

Sarneel J M Sundqvist M K Molau U BjoumlrkmanM P and Alatalo J M Decomposition rate and stabi-lization across six tundra vegetation types exposed to gt

20 years of warming Sci Total Environ 724 138304httpsdoiorg101016jscitotenv2020138304 2020

Schmidt M W I Torn M S Abiven S Dittmar T Guggen-berger G Janssens I A Kleber M Kogel-Knabner ILehmann J Manning D A C Nannipieri P Rasse D PWeiner S and Trumbore S E Persistence of soil organic mat-ter as an ecosystem property Nature 478 49ndash56 2011

Talbot J M and Treseder K K Dishing the dirton carbon cycling Nat Clim Change 1 144ndash146httpsdoiorg101038nclimate1125 2011

Talbot J M and Treseder K K Interactions among lignin cellu-lose and nitrogen drive litter chemistrydecay relationships Ecol-ogy 93 345ndash354 httpsdoiorg10189011-08431 2012

Talia P Sede S M Campos E Rorig M Principi D TostoD Hopp H E Grasso D and Cataldi A Biodiversity char-acterization of cellulolytic bacteria present on native Chaco soilby comparison of ribosomal RNA genes Res Microbiol 163221ndash232 httpsdoiorg101016jresmic201112001 2012

MGnify Who eats the tough stuff DNA stable isotope probing (SIP)of bacteria and fungi degrading 13C-labelled lignin and cellu-lose in forest soils Sampling event dataset GBIForg [data set]httpsdoiorg1015468cewp3n 2020

Tatzber M Schlatter N Baumgarten A Dersch G KoumlrnerR Lehtinen T Spiegel H KMnO4 determinationof active carbon for laboratory routines three long-termfield experiments in Austria Soil Res 53 190ndash204httpsdoiorg101071SR14200 2015

Thompson L R Sanders J G McDonald D Amir A LadauJ Locey K J Prill R J Tripathi A Gibbons S M Ack-ermann G Navas-Molina J A Janssen S Kopylova EVaacutezquez-Baeza Y Gonzaacutelez A Morton J T Mirarab SZech X Z Jiang L Haroon M F Kanbar J Zhu Q JinS S Kosciolek T Bokulich N A Lefler J Brislawn CJ Humphrey G Owens S M Hampton-Marcell J Berg-Lyons D McKenzie V Fierer N Fuhrman J A ClausetA Stevens R L Shade A Pollard K S Goodwin K DJansson J K Gilbert J A and Knight R A communal cata-logue reveals Earthrsquos multiscale microbial diversity Nature 551457ndash463 2017

Walters W Hyde E R Berg-Lyons D Ackermann GHumphrey G Parada A Gilbert J A Jansson J KCaporaso J G Fuhrman J A Apprill A and KnightR Improved bacterial 16S rRNA gene (v4 and v4-5)and fungal internal transcribed spacer marker gene primers



for microbial community surveys mSystems 1 e00009-15httpsdoiorg101128mSystems00009-15 2016

Wang Y Liu Q Yan L Gao Y Wang Y and WangW A novel lignin degradation bacterial consortium forefficient pulping Bioresource Technol 139 113ndash119httpsdoiorg101016jbiortech201304033 2013

Wei H Ma R Zhang J Saleem M Liu Z Shan XYang J and Xiang H Crop-litter type determines the struc-ture and function of litter-decomposing microbial communitiesunder acid rain conditions Sci Total Environ 713 136600httpsdoiorg101016jscitotenv2020136600 2020

Wickham H ggplot2 Elegant Graphics for Data AnalysisSpringer-Verlag New York USA ISBN 978-3-319-24277-42016

Wieder W R Grandy A S Kallenbach C M and BonanG B Integrating microbial physiology and physio-chemicalprinciples in soils with the MIcrobial-MIneral Carbon Sta-bilization (MIMICS) model Biogeosciences 11 3899ndash3917httpsdoiorg105194bg-11-3899-2014 2014

Wilhelm R C Singh R Eltis L D and Mohn W W Bacterialcontributions to delignification and lignocellulose degradationin forest soils with metagenomic and quantitative stable isotopeprobing ISME J 13 413ndash429 httpsdoiorg101038s41396-018-0279-6 2019

Wilpiszeski R L Aufrecht J A Retterer S T Sullivan MB Graham D E Pierce E M Zablocki O D PalumboA V and Elias D A Soil aggregate microbial communi-ties towards understanding microbiome interactions at biolog-ically relevant scales Appl Environ Microb 85 e00324-19httpsdoiorg101128AEM00324-19 2019

Anjos L Gaistardo C Deckers J Dondeyne S Eberhardt EGerasimova M Harms B Jones A Krasilnikov P Rein-sch T Vargas R and Zhang G World reference base forsoil resources 2014 International soil classification system fornaming soils and creating legends for soil maps 3rd ed editedby Schad P Van Huyssteen C and Micheli E FAO Rome(Italy) ISBN 978-92-5-108369-7 2015

Yan J Wang L Hu Y Tsang Y F Zhang Y Wu J Fu X andSun Y Plant litter composition selects different soil microbialstructures and in turn drives different litter decomposition patternand soil carbon sequestration capability Geoderma 319 194ndash203 httpsdoiorg101016jgeoderma201801009 2018

Yuste J C Pentildeuelas J Estiarte M Garcia-Mas J MattanaS Ogaya R Pujol M and Sardans J Drought-resistantfungi control soil organic matter decomposition and its re-sponse to temperature Glob Change Biol 17 1475ndash1486httpsdoiorg101111j1365-2486201002300x 2010

Zheng H Yang T Bao Y He P Yang K Mei X Wei ZXu Y Shen Q and Banerjee S Network analysis and sub-sequent culturing reveal keystone taxa involved in microbial lit-ter decomposition dynamics Soil Biol Biochem 157 108230httpsdoiorg101016jsoilbio2021108230 2021


Abstract

Introduction

Materials and methods

Description of the study sites

Soil sampling and soil chemical analyses

Tea Bag Index (TBI)

Molecular analyses of prokaryotic and fungal community structures

DNA extraction and amplicon sequencing

Sequence data processing and classification

Statistical analysis

Results and discussion

Soil chemical characteristics and Tea Bag Index

Establishment of an amplicon sequencing protocol to be used with the TBI

Richness and diversity of prokaryotic and fungal communities in teabags and soils

Selective colonisation of teabags by prokaryotic and fungal soil populations

Conclusions

Code availability

Data availability

Supplement

Team list

Author contributions

Competing interests

Disclaimer

Acknowledgements

Financial support

Review statement

References


1 Introduction

Litter decomposition is one of the most important terrestrialecosystem functions Soil microorganisms drive this processby breaking down plant material leading to the release ofcarbon into the atmosphere as CO2 or into the soil whereit either gets sequestered or further degraded (Talbot andTreseder 2011 Allison et al 2013 Cotrufo et al 2013)Other factors governing litter decomposition include themolecular structure of the litter its physical (dis)connectionfrom the decomposer community and its organondashmineralassociations (Schmidt et al 2011) Litter decompositiontherefore directly affects both Earthrsquos atmosphere and soilhealth Understanding this process holds the key to betteragricultural management mitigating greenhouse gas emis-sions and predicting future soil carbon storage levels How-ever because of the processesrsquo complexity and our inade-quate understanding of how biodiversity affects it the ap-plication of the litter decomposition process to agriculturalpractices and global decomposition models is not perform-ing currently at its full potential Seasonality mainly throughtemperature and moisture effects has been demonstrated toaffect litter decomposition directly and indirectly (Prescott2010) Direct effects occur due to the high sensitivity of bio-logical processes to temperature and water availability andindirect effects may be the consequence of phenotypic orcommunity shift responses of the soil biota Soil microbialcommunity structure has been hypothesised to influence pro-cess rates in soils (McGuire and Treseder 2010 Graham etal 2016) In a few models that linked microbial diversityand decomposition (eg MIMICS Wieder et al 2014 andMEMS Cotrufo et al 2013) microbial diversity positivelyaffected nutrient-cycling efficiency and ecosystem processesthrough either greater intensity of microbial exploitation oforganic matter or functional niche complementarity

Litter quality has been shown to be important for microbialcommunity structure since both bacteria and fungi respondto litter physicochemical changes during the decay process(Aneja et al 2006 Purahong et al 2016) Litter biochemi-cal traits such as the CN ratio and the fraction of lignin areconsidered good indicators of litter quality as they are relatedto nutrient availability and decomposition stage (Prescott2010 Talbot and Treseder 2012) Different litter types canthus select microbial taxa that are more specialised in degrad-ing their components Fungi are known to produce a suite ofoxidative enzymes that degrade the recalcitrant biopolymersof litter (Mathieu et al 2013 Hoppe et al 2015) In contrastonly a few types of bacteria degrade all lignocellulosic poly-mers while most typically target simple soluble compounds(de Boer and van der Wal 2008) Therefore the role of bacte-ria in the decomposition of more recalcitrant material is stilldebated (Wilhelm et al 2019) However still little is knownon how microbial communities specialise in litter types with

different physical and chemical traits (Freschet et al 2011Pioli et al 2020)

The community structure and functioning of microbialcommunities are further affected by biotic interactions (Dae-beler et al 2014 Garbeva et al 2014 Ho et al 2016Wilpiszeski et al 2019) In natural communities such in-teractions generally involve competition for space and re-sources (Boddy 2000 and Faust and Raes 2012 and ref-erences therein) but may also be mutualistic For example ithas been suggested that bacteria can facilitate the activity ofdecaying fungi by providing important nutrients such as ni-trogen (N) and phosphorus (P) (Purahong et al 2016) Like-wise fungi have been demonstrated to help improve the ac-cessibility of litter for bacteria (de Boer et al 2005) There-fore it is likely that decomposition dynamics depend notonly on microbial community structure but also on the facil-itative or competitive interactions among microbial species(Liang et al 2017 Aleklett et al 2021) Despite the indica-tions of the high importance of species interactions for soilorganic matter decomposition dynamics they are not wellunderstood and more studies under natural conditions areneeded

Linking microbial diversity to function across different en-vironments represents a key aspect of ecology In this studywe used high-throughput sequencing in combination withthe Tea Bag Index (TBI) ndash a cost-effective method to studylitter decomposition using commercially available teabags(Keuskamp et al 2013) This combination of methods al-lows insight to be gained into the effect of litter traits on thecommunity structure of microbial decomposers and soil or-ganic matter decomposition rates to be linked with microbialpopulation dynamics The microbial diversity in soils is im-mense (Thompson et al 2017) yet large parts of it are dor-mant (Lennon and Jones 2011) or simply do not participatein litter degradation (Falkowski et al 2008) Therefore theburied teabags used in this study serve as ldquotrapsrdquo for micro-bial litter degraders and assure that the diversity we observeis composed only of its active litter degrading and associatedtaxa

Specifically in this citizen-science-aided project we usedbarcoded-amplicon high-throughput sequencing of the SSUrRNA gene (for bacteria and archaea) and the internal tran-scribed spacer region (ITS) for fungi to compare the micro-bial community structure in two different standardised littertypes (rooibos and green tea) with the microbial communityin three different local soil types and during four seasonsover the course of a year We tested the extent to which littertypes with different traits represent selective substrates formicrobial community colonisation Finally we related themicrobial diversity and species relative abundance patternswith two proxies (decomposition rate (k) and stabilisationfactor (S)) that describe the decomposition of labile mate-rial (Keuskamp et al 2013) We generally assumed that lit-
































































5 Conclusions












References








































































































Abstract

Introduction




Tea Bag Index (TBI)










Conclusions

Code availability

Data availability

Supplement

Team list


Competing interests

Disclaimer

Acknowledgements

Financial support

Review statement

References































































5 Conclusions












References








































































































Abstract

Introduction




Tea Bag Index (TBI)










Conclusions

Code availability

Data availability

Supplement

Team list


Competing interests

Disclaimer

Acknowledgements

Financial support

Review statement

References

































































































Abstract

Introduction




Tea Bag Index (TBI)










Conclusions

Code availability

Data availability

Supplement

Team list


Competing interests

Disclaimer

Acknowledgements

Financial support

Review statement

References

Pairing litter decomposition with microbial community ... - SOIL

Documents

Transcript of Pairing litter decomposition with microbial community ... - SOIL