Nematodes associated with phytodetritus in salt marshes

Nematodes associated with phytodetritus in salt marshes. Colonization patterns and species- and diversity related impacts on

decomposition processes

Nematoden geassocieerd met fytodetritus in schorren. Kolonisatiepatronen en soorten diversiteitsafhankelijke invloed op

afbraakprocessen

Ilse De Mesel

Proefschrift voorgelegd tot het behalen van de graad van Doctor in de Wetenschappen (Biologie)

Promotor: Prof. Dr. Magda Vincx Co-promotor: Prof. Dr. Ir. Jean Swings

Dr. Tom Moens

Table of contents

Acknowledgements – Dankwoord … … … … … … … … ...… … … … … … … … … … … i

Summary… … … … … … … … … … … … … … ..… … … … … … … … … … … ...… … … .iii

Samenvatting … … … … … … … … … … … … … … … .… … … … … … … … .… … … ...vii

Chapter 1: General Introduction and Aims… … … ...… … … … … … … … … … … … 1

Introduction… … … … … … … … … … … … … … … … … … … … … … … … … … … … … 3 Topics and outline of the thesis… … … … … … … … … … … … … … … … … … … … .… .6

Chapter 2: Influence of bacterivorous nematodes on the decomposition of cordgrass … … … … … … … … … … … … … … … … … … … … … … … … … … … … … .9

De Mesel I, Derycke S, Swings J, Vincx M, Moens T (2003) Influence of bacterivorous nematodes on the decomposition of cordgrass. Journal of

Experimental Marine Biology and Ecology 296: 227-242

Abstract… … … … … … … … … … … … … … … … … … … … … … … … … … … … … … 11 Introduction… … … … … … … … … … … … … … … … … … … … … … … … … … … … ..13 Materials and Methods… … … … … … … … … … … … … … … … … … … … … ...… … ..14

Experimental set-up… … … … … … … … … … … … … … … … … … … … … ...14 Sampling… … … … … … … … … … … … … … … … … … … … … … .… … … ...15 Weight loss… … … … … … … … … … … … … … … … … … … … … … … … ....16 C/N-analysis… … … … … … … … … … … … … … … … … … … … … … ..… … 16 Nematodes… … … … … … … … … … … … … … … … … … … … … … … ..… ...17 Microbial activity… … … … … … … … … … … … … … … … … … … … … … ..17 Statistical analysis… … … … … … … … … … … … … ...… … … … … … … ......17

Results… … … … … … … … … … … … … … … … … … … … … … … … … … … … … … ..18 Decomposition… … … … … … … … … … ...… … … … … … … … … … … … … 18 Nematodes… … … … … … … … … … … … … … … … … … … ...… ...… … … ...21 Microbial activity… … … … … … … … … … … … … … … … … … … … ...… ...22

Discussion… … … … … … … … … … … … … … … … … … … … … … … … … … … … … 24 Decomposition process… … … … … … … … … … … … … … … … … ..… … … 24 Microbial activity… … … … … … … … … … … … … … … … … … … ..… … … 25

Nematodes and their influence on the microbial activity and decomposition process… … … … … … … … … … … … … ...… … … … … … … … … … … … ....25 Conclusions… … … … … … … … … … … … … … … … … … … … … … … … ...28

Acknowledgements… … … … … … … … … … … … … … … … … … … … … … … ...… ...29

Chapter 3: Top-down impact of bacterivorous nematodes on the bacterial

community structure: a microcosm experiment… … … … … … … ...… … … … … ...31

De Mesel I, Derycke S, Moens T, Van der Gucht K, Vincx M, Swings J (in press) Top-down impact of bacterivorous nematodes on the bacterial community

structure: a microcosm experiment. Environmental Microbiology

Abstract… … … … … … … … … … … ..… … … … … … … … … … … … … … … … … … ..33 Introduction… … … … … … … … … … … … … … … … … … … … … … … … … … ..… … 35 Materials and Methods… … … … … … … … … … … … … … … … … … … … ...… … … ..37

Experimental set-up and sampling… … … … … … … … … … … … … .… … ...37 DNA-extraction and – purification… … … … … … … … … … … … … … … … .38 PCR and DGGE… … … … … … … … … … … … … … … … … … … … … … … 38 Sequencing of excised DGGE bands… … … … … … … … … … … … … … … .39 Bacterial diversity… … … … … … … … … … … … … … … … … … … … ...… ..39 Statistical analysis of the DGGE fingerprints… ...… … … … … … … … … … 40 Nucleotide sequence accession numbers… … … … … … … … … … … … … ...40

Results and Results and Discussion… … … … … … … … .… … … … … … … … … … … 41 Effects on bacterial community composition… … … … … … … … … … … … .41 Effects on bacterial diversity… … … … … … … … … … … … … … … … … … .48 Conclusions… … … … … … … … … … … … … … … … … … … … … .… … … ..51

Acknowledgements… … … … … … … … … … … … … … … … … … … … … … … … … ..52

Chapter 4: Effect of nematode diversity on organic matter decomposition and the associated microbial community: a microcosm study… … … ..… … … … … … 53

De Mesel I, Derycke S, Swings J, Vincx M, Moens T

Abstract… … … … … … … … … … … … … … … … … … … … … … … … … … … … … ....55 Introduction… … … … … … … … … … … … … … … … … … … … … … … … … … … … ..57 Materials and Methods… … … … … … … … … … … … … … … … … … … … … … ...… ..58

Experimental set-up… … … … … … … … … … … … … … … … … … … … … ...58 Decomposition rate… … … … … … … … … … … … … … … … … … … … .… ..59 Nematodes… … … … … … … … … … … … … … … … … … … … … … … ...… ..60 Microbial activity… … … … … … … … … … … … … … … … … … … … … … ..60 Bacterial diversity… … … … … … … … … … … … … … … … … … ...… … … ..60 Statistical analysis… … … ...… … … … … … … … … … … … … … … … … ......61

Results… … … … … … … … … … … … … … … … … … … … … … … … … … … … … ......62 Nematodes… … … … … … … … … … … … … … … … … … … … … … … … .....62 Microbial activity… … … … … … … … … … … … … … … … … … … … … ......63 Bacterial diversity… … … … … … … … … … … … … … … … … … ...… … … ..67 Decomposition rate… … … … … … … … … … … … … … … … … … … … … ...69

Discussion… … … … … … … … … … … … … … … … … … … … … … … … … … … … … 73 Development of nematode populations… … … … … … … … … … … … … … .73 Effects of nematodes on microbial activity and bacterial diversity… … … ..75 Impacts on the decomposition rates… … … … … … … … … … … … … … … ..76 Conclusions… … … … … … … … … … … … … … … … … … … … … … … … ...78

Acknowledgements………………………………………………………………..…..79

Chapter 5: Development of a nematode community on decaying organic matter (Spartina anglica) under contrasting conditions.....................................................81

De Mesel I, Vanaverbeke J, Vincx M, Moens T (submitted) Development of a

nematode community on decomposing organic matter (Spartina anglica) under contrasting conditions

Abstract… … … … … … … … … … … … … … … … … … … … … … … … … … … … … … 83 Introduction… … … … … … … … … … … … … … … … … … … … … … … … … … … … ..85 Materials and Methods… … … … … … … … … … … … … … … … … … … … … … … … .86

Experimental set-up… … … … … … … … … … … … … … … … … … … … … ..86 Sampling… … … … … … … … … … … … … … … … … … … … … … … … ..… .87 Nematode community… … … … … … … … … … … … … … … … … … … … ... 87 Statistical analysis… … … … … … … … … … … ...… … … … … … … … … .… 88

Results… … … … … … … … … … … … … … … … … … … … … … … … … … … … … … ..89 Nematode densities… … … … … … … … … … … … … … … … … … … … ...… 89 Nematode community… … … … … … … … … … … … … … … … … … … … ... 91

Discussion… … … … … … … … … … … … … … … … … … … … … … … … … … … … … 95 Acknowledgements… … … … … … … … … … … … … … … … … … … … … … … … … 100

Conclusions… … … … … … … … … … … … ..… … … … … … … … … … … … … … … .101 Species manipulation experiment… … … … … … … … … … … … … … … ...103 Colonization experiment… … … … … … … … … … … … … … … … … … … .108

Appendix A...… … … … … … … … … … … … … … … … … … … … … … … … … … ....111

Cited literature… … … … … … … … … … ..… … … … … … … … … … … ..… … … … .123

Dankwoord

i

Dankwoord

Het moment is gekomen om een aantal mensen te bedanken die me de voorbije jaren

met raad en daad hebben bijgestaan. Graag wil ik beginnen bij mijn promotor Prof.

Dr. Magda Vincx en mijn co-promotor Prof. Dr. Ir. Jean Swings. Zonder jullie steun

zou van dit doctoraat geen sprake geweest zijn! Jullie hebben mij indertijd de kans

gegeven om een doctoraatsbeurs aan te vragen en zijn altijd achter mij blijven staan.

Magda, bedankt om me de vrijheid te laten ‘mijn ding’ te doen! Jean, ik ben enorm

blij dat ik op het Labo voor Microbiologie heb mogen kennis maken met de

microbiële onderzoekswereld. Jouw enthousiasme heeft me vaak deugd gedaan!

Dr. Tom Moens heeft me indertijd warm gemaakt voor dit onderzoeksproject, en daar

ben ik nog steeds heel erg blij om. De discussies rond mijn manuscripten hebben

zeker de kwaliteit van dit proefschrift omhoog gehaald.

Bedankt Prof. Dr. Ann Vanreusel voor de hulp bij mijn statistische problemen en voor

het nalezen van een aantal van mijn manuscripten, maar ook voor de babbels

tussendoor en voor het aangename gezelschap op de staalnamecampagne met de

Polarstern. Ik vond het enorm fijn dat ik van jou en Magda de kans heb gekregen

ervaring op te doen tijdens de campagnes met de Discovery en Polarstern.

Dirk Vangansbeke en Annick Van Kenhove wil ik graag bedanken voor de vele

chemische analyses, Bernard Timmerman voor de hulp bij de staalnames. Annick,

Bernard en Bart Beuselinck hebben ook heel wat staaltjes voor mij opgespoeld en

nematoden in preparaat gebracht, waarvoor mijn dank! Bedankt Gunther Van

Ryckegem voor de ergosterolanalyses.

De mensen van het oranje labo op de microbiologie, Dr. Katleen Van der Gucht, Nele

Vloemans en Sylvie Cousin verdienen hier zeker een plaatsje! Eerst en vooral omdat ik

altijd van jullie labo heb mogen gebruik maken. Nele, jij hebt me indertijd de DGGE

techniek en alles daarrond aangeleerd. Ik kon altijd op je hulp rekenen als ik weer

eens een overklaarbaar probleem had. Katleen, bedankt voor de hulp bij de

verwerking van de DGGE analyses en de suggesties bij het manuscript!

Zonder Margo Cnockaert was ik waarschijnlijk nog altijd op zoek naar die ene pipet,

dat bepaald product, dat specifieke protocol.... Margo, heel erg bedankt voor alle hulp,

tips, je luisterend oor! Het was niet altijd evident om mijn draai te vinden op het labo,

maar ik heb altijd op jou kunnen rekenen!

Verder wil ik ook iedereen van de microbiologie bedanken die me op de één of andere

manier heeft geholpen of gesteund. Ik denk hierbij speciaal aan Renata Coopman voor

Dankwoord

ii

de hulp bij de sequenties, Anne Willems voor het overleg tussen de labo’s, Evie De

Brandt voor de laatste DGGE, en natuurlijk ook Liesbeth Masco (eten? 12h?).

Ook een woordje van dank voor Prof. Dr. Jacques Vanfleteren omdat ik ettelijke

keren van zijn labo heb mogen gebruik maken. Bij Annemie De Vreese en Andy

Vierstraete kon ik altijd terecht met vragen en problemen. Jullie tips hebben mij

dikwijls een flinke stap vooruit geholpen!

In de beginjaren kon ik rekenen op Kris Hostens om alle computerproblemen op te

lossen, de laatste jaren komt Guy Desmet de wormen van mijn computer verjagen en

andere depannages uitvoeren! Ook Tim Deprez en Tom Gheskiere hebben mij al vaak

met raad en daad bijgestaan.

Ik heb het geluk gehad drie fijne thesisstudentes te mogen begeleiden: bedankt Sofie

Derycke, Annelies Phillips en Phaedra Van Brussel voor jullie enthousiasme!

Bedankt Marleen De Troch voor het lezen van een aantal van mijn manuscripten. Dr.

Peter Herman wil ik graag bedanken voor de hulp met statistische problemen en het

doornemen van één van mijn manuscripten.

Een speciaal woordje van dank ook voor de vele mensen van de mariene die

misschien niet rechtstreeks bij mijn onderzoek betrokken waren, maar die wel altijd

klaar stonden met goeie raad, bemoedigende woordjes, .... Ik denk hierbij vooral aan

Maaike Steyaert, Jan Vanaverbeke, Sofie Derous, Thomas Remerie, Jan Wittoeck,

Nancy Fockedey, Sofie Derycke, Maarten Raes, Henni Hampel, Isolde De Grem,

Annick Verween (ook voor het uitlenen van je wrak... euh fiets natuurlijk!), ....

Eigenlijk verdienen alle ‘Marbiollers’ van de voorbije 4,5 jaar hier een plaatsje als

dank voor de goede sfeer al die tijd!

Graag wil ik nog een aantal mensen buiten ‘den unief’ bedanken om mij af en toe

eens de doctoraatsdruk te laten vergeten. Bedankt Liesbeth Desmyter, Nathalie Hardy,

Tim en Sofie, Ann en Daf, Peter en An, Frederik, kleine Lukja....

Tot slot een speciaal woordje van dank voor mijn ouders voor alle steun de voorbije

jaren.... en natuurlijk ook voor Peter.

De financiële steun voor dit doctoraatsonderzoek werd verschaft door ‘Instituut voor

de aanmoediging van Innovatie door Wetenschap en Technologie in Vlaanderen’

(IWT).

Summary

iii

Summary Decaying phytodetritus deposited on the sediment surface in salt marshes becomes

quickly colonised by nematodes. Both taxonomic and functional diversity within these

communities are generally low; they are often dominated by the bacterivore, deposit

feeding Monhysterida and Rhabditida. Bacterial grazers are assumed to stimulate the

decomposition rate of organic matter through interactions with the bacterial

community. It is however not clear whether the co-occurrence of these functionally

similar species is of importance in the functioning of the ecosystem. One of the most

accepted hypotheses suggests that species within a functional group are redundant and

thus perform the same role within an ecosystem. The general aims of a first

experiment were to test whether bacterivorous nematodes affect (i) the decomposition

of organic matter and (ii) the activity of the associated microbial community, and to

examine whether these effects (iii) differed between functionally similar species or

(iv) were diversity dependent. (v) Species specific effects on the bacterial community

structure were also analysed.

The influence of four bacterivorous nematode species (Diplolaimelloides meyli,

Diplolaimelloides oschei, Diplolaimella dievengatensis, Panagrolaimus paetzoldi) on

the decomposition of cordgrass (Spartina anglica) and its associated microbial

community was investigated using laboratory microcosm experiments. Cordgrass

leaves were incubated on a sediment surface with a natural mixture of bacteria

originating from sediment, cordgrass detritus and habitat water. Nematodes were

applied separately or in combinations of two or three species. This experimental set-

up allowed testing for species-specific and/or diversity-dependent effects.

In contrast with our expectations, no stimulation of the decomposition process nor of

the microbial community was observed in the presence of the nematodes. In absence

of nematodes, a more extensive growth of fungi, important in the early stages of

cordgrass decay, might have caused higher decomposition rates. Fungal growth

seemed to be hampered by the nematodes, but whether this was a direct or indirect

effect, could not be distinguished. Furthermore, obvious differences were found

between nematode treatments. Panagrolaimus paetzoldi reached much higher

densities than the other species, both in the monospecific and in the combination

treatments. Within the communities containing exclusively monhysterid nematode

species, inhibitory inter-specific interactions were observed, probably due to chemical

Summary

iv

interference. Panagrolaimus paetzoldi caused a decrease in microbial activity. This

was most likely a result of overgrazing. Remarkably, this low microbial activity did

not result in a slow-down of the decomposition process compared to the other

treatments with nematodes. We suggest that P. paetzoldi might be able to directly

assimilate detrital compounds. Also within the treatments containing exclusively

monhysterid nematode species, differences in microbial activity could not be directly

linked to decomposition rates. The fact that fungal and bacterial activity could not be

measured completely separately complicates the establishment of any such link, but

differences in the efficiency of the bacterial communities in each treatment should

also be considered.

Not only impact on bacterial activity, but also on bacterial community composition

was analysed in presence of bacterivorous nematodes. The bacterial community

structure in the monospecific treatments and the control was analysed by means of

Denaturing Gradient Gel Electrophoresis of the 16S rRNA genes. Multi Dimensional

Scaling showed grouping of the bacterial communities per treatment. Analysis of

Similarities indicated that the differences between treatments were significantly larger

than differences within treatments. Our results demonstrate that nematodes can have a

significant structuring top-down influence on the ‘pool’ of bacteria growing on the

detritus, even at the low densities as obtained by the monhysterid species, and that

these impacts are species-specific. Differences in bacterial community composition

between the treatments with monhysterids (D. meyli, D. oschei and D. dievengatensis)

can be explained by species-specific food preferences. Panagrolaimus paetzoldi on

the other hand feeds unselectively, and thus affects the bacterial community

differently. Part of the consumed bacteria are not digested and consequently excreted

again into the environment. Not palatability, but rather cell wall properties would then

determine whether bacteria are digested. Bacterial diversity was assessed from the

DGGE-profiles in the monospecific and control treatments, as well as in the

combination treatments. A top-down effect of the nematodes on the diversity of the

bacterial community was mainly evident under high grazing pressure, i.e. in the

presence of P. paetzoldi, but also with increasing nematode diversity a decrease in

bacterial diversity was observed.

Our results do not support the idea that bacterivorous nematodes are functionally

redundant. The response of P. paetzoldi in the system was profoundly different from

that of the monhysterid nematode species (D. meyli, D. oschei and D. dievengatenis),

Summary

v

yielding considerably higher densities, which resulted in clear impacts on the bacterial

community. As for the monhysterid species, differences in microbial activity and

decomposition rate were found in the presence of different species or species

combinations, without any clear species-specific or diversity-dependent link. This

suggests an idiosyncratic relationship between bacterivorous nematode diversity and

ecosystem functioning.

In a second experiment, the development of the nematode community on decaying

cordgrass detritus in litterbags was studied in the laboratory and in the field. The aims

were to test whether (i) detritus availability could impact the distribution of the

nematode community in an undisturbed laboratory environment, (ii) this distribution

could be affected by natural disturbances, mainly related to tidal currents, in the field

and (iii) the contrasting environmental conditions in the field and in the laboratory

would alter the nematode community structure. Nematodes showed a clear preference

for the detritus under controlled conditions in the laboratory, and the nematode

assemblage on the cordgrass differed considerably from that in the surrounding

sediments. By contrast, in the field, nematodes were rather distributed at random,

probably through passive transport with the tides, showing that active habitat

colonization was of minor importance in the hydrodynamically impacted system.

We assumed that the contrasting incubation conditions in the laboratory and in the

field would affect the structural and functional aspects of the nematode community. In

the laboratory, incubations were performed in the dark, preventing growth of

microalgae, which is an important food source for epistrate feeders and deposit

feeders. Epistrate feeders were however equally abundant in the field and in the

laboratory, indicating that other food sources, probably bacteria, can be consumed.

Deposit feeders, dominating field samples, were mainly replaced by microvores in the

laboratory; still some species were able to adapt to a principally bacterial diet.

Nematode diversity was not affected by resource availability, nor by the daily tidal

disturbance.

Samenvatting

vii

Samenvatting In schorren wordt rottend fytodetritus dat op het sedimentoppervlak terecht komt snel

gekoloniseerd door nematoden. Deze nematodengemeenschappen zijn over het

algemeen gekenmerkt door zowel een lage taxonomische als functionele diversiteit.

Ze worden vaak gedomineerd door de bacterivore, ‘deposit-feeding’ Monhysterida en

Rhabditida. Bacteriële grazers worden verondersteld het afbraakproces van detritus te

stimuleren via interacties met de bacteriegemeenschap. Het is echter niet duidelijk of

het samen voorkomen van deze functioneel erg gelijkende soorten van belang is voor

het functioneren van het systeem. Een vaak gestelde hypothese is dat soorten binnen

een functionele groep dezelfde rol zouden vervullen in een ecosysteem. De algemene

doelstellingen van een eerste experiment waren na te gaan of bacterivore nematoden

(i) de afbraak van organisch materiaal en (ii) de activiteit van de geassocieerde

microbiële gemeenschap beïnvloeden, en te onderzoeken of deze impact (iii)

verschilde tussen functioneel gelijkende soorten of (iv) diversiteitsafhankelijk was.

(v) Ook soortspecifieke effecten op de bacteriële gemeenschapsstructuur werden

geanalyseerd.

Met behulp van microcosmosexperimenten werd de invloed onderzocht van vier

bacterivore nematodensoorten (Diplolaimelloides meyli, Diplolaimelloides oschei,

Diplolaimella dievengatensis en Panagrolaimus paetzoldi) op de afbraak van slijkgras

(Spartina anglica) en op de geassocieerde bacteriële gemeenschap. Slijkgrasbladeren

werden geïncubeerd op een sedimentoppervlak en geënt met een natuurlijk bacterieel

mengsel. Nematoden werden ofwel afzonderlijk toegevoegd, ofwel in combinaties

van twee of drie soorten. Deze experimentele opzet liet toe om soortspecifieke en/of

diversiteitsafhankelijke effecten te achterhalen.

In tegenstelling tot de verwachtingen werd er geen stimulatie van het afbraakproces,

noch van de microbiële gemeenschap waargenomen in de aanwezigheid van

nematoden. In afwezigheid van nematoden zou een sterkere groei van fungi

verantwoordelijk kunnen zijn voor de hogere afbraaksnelheden. Fungi zijn belangrijk

in de eerste stadia van het decompositieproces van slijkgras. De groei van fungi leek

gehinderd in aanwezigheid van nematoden, maar of dit een gevolg was van directe of

indirecte interacties kon niet onderscheiden worden. Ook tussen de experimentele

reeksen met verschillende nematodensoorten en -combinaties werden belangrijke

verschillen gevonden. Panagrolaimus paetzoldi haalde veel hogere densiteiten dan de

Samenvatting

viii

andere soorten, zowel in de monospecifieke als in de combinatiereeksen. Binnen de

gemeenschappen met enkel monhysteride nematodensoorten werden negatieve

interspecifieke interacties waargenomen, die vermoedelijk veroorzaakt werden door

chemische interferentie. In aanwezigheid van P. paetzoldi werd de microbiële

activiteit onderdrukt, waarschijnlijk als gevolg van overbegrazing. Vreemd genoeg

resulteerde deze lagere microbiële activiteit niet in een vertraging van het

decompositieproces in vergelijking met andere nematodenreeksen. We

veronderstellen dat P. paetzoldi in staat is om rechtstreeks componenten van het

organische materiaal te assimileren. Ook binnen de reeksen waarin enkel

monhysteriden groeiden kon geen verband gelegd worden tussen microbiële activiteit

en afbraaksnelheid. Dit zou kunnen liggen aan het feit dat de activiteit van fungi en

bacteriën niet volledig afzonderlijk kon bepaald worden, maar ook verschillende

groeiefficiënties van de bacteriële gemeenschap vormen een mogelijke verklaring.

Naast de impact van bacterivore nematoden op de bacteriële activiteit, werd ook de

invloed op de bacteriële gemeenschapssamenstelling onderzocht. De bacteriële

gemeenschapsstructuur in de monospecifieke reeksen en de controle werd

geanalyseerd aan de hand van ‘Denaturing Gradient Gel Electrophoresis’ van de 16S

rRNA genen. ‘Multi Dimensional Scaling’ toonde een groepering van de bacteriële

gemeenschappen per reeks. ‘Analysis of Similarities’ bevestigde dat de verschillen

tussen de reeksen significant groter waren dan de verschillen binnen de reeksen. Onze

resultaten tonen aan dat nematoden zelfs bij lage densiteiten, zoals deze van de

monhysteriden, een aanzienlijke structurerende ‘top-down’ impact kunnen hebben op

de ‘pool’ van bacteriën die groeien op detritus, en dat deze invloed soortspecifiek is.

Verschillen in bacteriële gemeenschapsstructuur in de reeksen met de monhysteride

nematodensoorten (D. meyli, D. oschei en D. dievengatensis) kunnen verklaard

worden door soortspecifieke voedselpreferenties. Panagrolaimus paetzoldi

daarentegen voedt zich niet selectief en scheidt een deel van de opgenomen bacteriën

onverteerd terug uit; deze soort heeft daardoor een erg verschillende impact op de

bacteriële gemeenschap. In dit geval zou niet zozeer ‘smaak’, maar eerder

eigenschappen van de celwand bepalen of bacteriën al dan niet worden verteerd. Aan

de hand van de DGGE-profielen werd ook de bacteriële diversiteit bepaald in de

monospecifieke reeksen, in de controle, en in de combinatiereeksen. Een top-down

effect van de nematoden op de diversiteit van de bacteriële gemeenschappen werd

Samenvatting

ix

duidelijk onder hoge graasdruk, zoals in aanwezigheid van P. paetzoldi, maar ook met

toenemende nematodendiversiteit nam de bacteriële diversiteit af.

De hypothese dat veel bacterivore nematodensoorten redundant zouden zijn wordt

niet ondersteund door onze resultaten. De respons van P. paetzoldi verschilde

fundamenteel van die van de monhysteride nematodensoorten (D. meyli, D. oschei en

D. dievengatensis). Panagrolaimus paetzoldi bereikte veel hogere densiteiten, wat

resulteerde in een duidelijke impact op de bacteriële gemeenschap. Binnen de reeksen

met monhysteride soorten werden verschillen gevonden in microbiële activiteit en

afbraaksnelheden in aanwezigheid van verschillende soorten of soortencombinaties,

maar zonder duidelijk soort- of diversiteitsafhankelijk verband. Dit suggereert een

idiosyncratisch verband tussen diversiteit van bacterivore nematoden en het

functioneren van het systeem.

In een tweede experiment werd de ontwikkeling van nematodengemeenschappen op

rottend slijkgras bestudeerd met behulp van ‘litterbag’- incubaties, enerzijds onder

gecontroleerde omstandigheden in het laboratorium, en anderzijds in het veld, waar

incubaties onderhevig waren aan fluctuaties, vooral geassocieerd met de

getijdenwerking. Het doel van het experiment was na te gaan of (i) de aanwezigheid

van detritus een invloed heeft op de verspreiding van nematoden in een onverstoord

milieu, (ii) deze verspreiding kon beïnvloed worden door de natuurlijke verstoringen

in het veld, en of (iii) de verschillende omgevingsfactoren in het veld en het labo een

impact hebben op de structuur van de nematodengemeenschap.

Onder gecontroleerde omstandigheden in het labo toonden nematoden een duidelijke

voorkeur voor het detritus, en de nematodengemeenschap op het slijkgras verschilde

fundamenteel van die in het omringende sediment. In het veld daarentegen waren de

nematoden eerder ad random verspreid, waarschijnlijk ten gevolge van passief

transport met de getijden, wat aantoont dat actieve kolonisatie hier van minder belang

was.

We verwachtten dat de contrasterende incubatie-omstandigheden (laboratorium-veld)

de structurele en functionele aspecten van de nematodengemeenschap zouden

beïnvloeden. In het labo werden de incubaties uitgevoerd in het donker, hetgeen de

groei van microalgen, een belangrijke voedselbron voor ‘epistrate feeders’ en

‘deposit-feeders’, verhinderde. ‘Epistrate feeders’ waren echter in gelijke mate

aanwezig in het veld en in het laboratorium, wat aantoont dat ze ook andere

voedselbronnen kunnen benutten, vermoedelijk bacteriën. ‘Deposit-feeders’ waren

Samenvatting

x

dominant aanwezig in de veldstalen, en werden in het laboratorium voornamelijk

vervangen door microvoren; toch konden enkele soorten zich aanpassen aan een

hoofdzakelijk bacterieel dieet. De nematodendiversiteit werd niet beïnvloed door

voedselbeschikbaarheid, noch door de (voornamelijk) tidale verstoringen.

Chapter 1

General Introduction and Aims


3

Introduction Salt marshes are among the most productive ecosystems of the world (Odum, 1971).

They occur along many coasts and estuaries in temperate regions. The annual primary

production can be up to 4 kg C per m², most of it produced by grasses (Spartina spp.)

which generally dominate these ecosystems (Howarth, 1993). The net export of

organic matter to coastal waters is usually limited, especially in the Westerscheldt

Estuary (Hemminga et al. 1992, 1993), and only a very small fraction of the

production is permanently accreted in marsh sediments. Most of the organic matter

produced in marshes is decomposed in situ, fueling intense rates of microbial

processes (Howarth, 1993). Fungi and bacteria are the principal competitors for

organic substrates. Fungi are mostly aerobic decomposers and dominate the

aboveground decomposition processes of plant detritus (Newell et al., 1989). Once the

detritus is deposited on the sediment surface, or becomes mixed with the sediment,

bacteria take over their function (Howarth, 1993). The benthic bacterial community

can be affected by the activity of their grazers, mainly protozoa and nematodes.

Free-living nematodes are generally the dominant metazoan meiobenthic taxon in

marine and estuarine sediments. Their densities vary between 105 and 107 individuals

per m² (Heip et al., 1985). Despite their numerical dominance, the function and

trophic position of nematodes in benthic food webs has long been poorly understood.

Meanwhile it has been shown that they play an important role in energy fluxes in the

benthos. On the one hand they are food sources for macrofauna and epi- and

hyperbenthic organisms (Gee, 1989; Coull, 1990; Hamerlynck and Vanreusel, 1993).

On the other hand, a wide spectrum of particulate organic food sources, e.g. bacteria,

diatoms, ciliates, flagellates or other meiofaunal organisms, can be consumed by

nematodes. Based on morphological aspects of their buccal cavity, an assessment was

made of the feeding preferences of free-living marine nematodes, leading to the

subdivision in trophic groups (Wieser, 1953). This classification has been refined by

incorporating observations on the gut contents (e.g. Perkins, 1958; Hopper and

Meyers, 1967; vanThun, 1968; Deutsch, 1978) and on the feeding habits of a range of

nematode species (Romeyn and Bouwman, 1983; Jensen, 1987; Moens and Vincx,

1997).

Not only the densities of nematodes are high in marine sediments, their communities

are also characterised by high diversities (Heip et al., 1985). The question can be


4

raised why so many (functionally) similar species occur in marine and estuarine

benthic sites. In general, several, often contrasting, hypotheses about the biodiversity-

ecosystem functioning relationship have been formulated (Schläpfer et al., 1999)1.

The extremes range from the (modified) null hypothesis, predicting no effect of

changing species diversity on the functioning of ecosystems, to the rivet hypothesis

(Ehrlich and Ehrlich, 1981), describing an enhancement of processes each time a

species is added to the system. Over the last few years, the redundancy hypothesis has

received a lot of attention. According to this model, the effect of altering species

diversity will decrease at higher diversity levels (Walker, 1992). The underlying idea

is that with increasing diversity, the function of a disappearing species is more likely

to be taken over by other species of the same functional level. In contrast with the

other theories, the idiosyncratic model predicts that altering species diversity will

affect ecosystem functioning, but in an unpredictable manner (Lawton, 1994).

Empirical studies on the biodiversity-function link generally focus on vegetations in

terrestrial systems in which production and stability of the ecosystems are studied

(a.o. Ewel et al., 1991; Givnish, 1994; Silvertown et al., 1994; Tilman and Downing,

1994, Collins, 1995; Tilman et al., 1996). The lack of straightforward results in these

studies demonstrates that a generally applicable hypothesis probably does not exist

and emphasises the necessity to incorporate other ecosystems and other functions into

this discussion.

In many ecosystems, primary production enters the decomposition cycle as plant litter

(Wardle et al., 1997). As mentioned before, this also goes for detritus in salt marshes

along the Westerscheldt (Hemminga 1992, 1993). When deposited on the sediment

surface, decaying phytodetritus becomes colonised by nematodes. These communities

are generally characterised by a lower taxonomic and functional diversity compared to

those of the surrounding sediments, with typically the bacterivore deposit-feeding

Monhysterida and Rhabditida occurring in these spots (Bouwman, 1984; Warwick,

1987; Alkemade et al., 1993). The main reason for this microhabitat preference may

be related to food preferences (Trotter and Webster, 1984; Moens et al., 1999).

However, an evaluation of the relative importance of abiotic disturbances, mainly

1 see also introduction Chapter 4


5

related to tidal action, in structuring intertidal nematode communities in salt marshes

has not yet been made2.

Bacterivorous nematodes are assumed to affect decomposition of organic matter

through interactions with heterotrophic bacteria3. A number of mechanisms have been

proposed by which they could stimulate a bacterial community. Grazing for instance

would enhance bacterial turn-over (e.g. Tenore et al., 1977; Abrams and Mitchell,

1980; Tietjen, 1980; Findlay and Tenore, 1982; Anderson et al., 1983; Rieper-

Kirchner, 1989; Alkemade et al., 1992a, b; Beare et al., 1992). Excretion of N-rich

compounds has been found to be important in stimulating bacterial activity in N-

limited systems (Anderson et al., 1983). Bacterial growth might be facilitated by the

Figure 1: Overview of the fate of primary production in salt marshes along the Westerscheldt.

Dotted lines represent the interactions studied in this thesis.

secretion of mucus trails (Riemann and Schrage, 1978), and through their movements,

nematodes are thought to enhance diffusion of nutrients and oxygen in the sediment

(Cullen, 1973; Alkemade et al., 1992a; Aller and Aller, 1992). Whether nematodes

can also affect bacterial community structure, as observed under protist grazing 2 see also introduction Chapter 5 3 see also introduction Chapter 2


6

pressure (Van Hannen et al., 1999), needs to be investigated4. Experimental data

indicate that nematodes are able to distinguish between, and preferentially consume

different bacterial strains when offered in pure cultures (Moens et al., 1999), but

whether they maintain the ability of selecting bacteria from mixed communities is

unknown.

Bacterivorous nematode communities associated with decaying phytodetritus in salt

marshes are very well suited to test for diversity-related impacts on ecosystem

functioning. Many of the species typically associated with phytodetritus can easily be

cultivated and manipulated under laboratory conditions (Moens and Vincx, 1998).

Additionally, their small body sizes (<1mm) and limited dispersal capacities allows

studying their impact on ecosystem processes at a small spatial scale, in the order of

centimetres (Coleman, 1983, Moore et al., 1996; Verhoef, 1996; Mikola and Setälä,

1998). In specific, it is easily feasible to compose artificial nematode communities

with different species compositions and diversity levels under controlled conditions,

in order to evaluate the importance of diversity related or species-specific impacts on

bacterial communities and decomposition rates.

Topics and outline of this thesis

The first topic of this thesis concentrates on the influence of free-living nematodes

on the decomposition of salt marsh phyto-detritus and its associated bacterial

community. This was studied using microcosm experiments in which cordgrass

detritus was incubated on a sediment surface. The microcosms were inoculated with a

natural mixture of bacteria, originating from the sediment, fresh and decaying

cordgrass, habitat water and nematode cultures. Three monhysterid nematode species

(Diplolaimelloides meyli, Diplolaimelloides oschei, Diplolaimella dievengatensis) and

one rhabditid (Panagrolaimus paetzoldi) commonly found on decaying cordgrass

detritus in the Paulina salt marsh (Westerscheldt Estuary) (Moens, pers. comm.), were

added to the microcosms, either separately, or in combination with one or two of the

other species. The general aims of this experiment were to test whether nematodes

affect (i) the decomposition process and (ii) the activity of the microbial community,

4 see also introduction Chapter 3


7

and to examine whether these influences are (iii) species-specific or (v) rather

diversity-dependent. (iv) Also the bacterial community structure in the presence of the

four nematodes species was studied. These results are discussed in chapter 2, chapter

3 and chapter 4 of this thesis.

In Chapter 2 species-specific influences on the decomposition process and on the

activity of the associated microbial community are described, using the results of the

monospecific treatments and the controls. It was expected that the nematodes would

enhance the decomposition process through stimulation of the bacterial community

and, taking the redundancy hypothesis as basic assumption, that this effect would not

be species-specific. This chapter has been published as De Mesel, I, Derycke, S,

Swings, J, Vincx, M, and Moens, T (2003) Influence of bacterivorous nematodes on

the decomposition of cordgrass. J. Exp. Mar. Biol. Ecol. 296: 227-242.

In Chapter 3 species-specific effects of the different nematode species on the

composition and diversity of the bacterial community developing on the detritus was

examined using Denaturing Gradient Gel Electrophoresis. Such impacts have already

been demonstrated under protist grazing activities (Van Hannen et al., 1999; Rønn et

al., 2002). This chapter has been published as De Mesel, I, Derycke, S, Moens, T, Van

der Gucht, K, Vincx, M, and Swings, J (2004) Top-down impact of bacterivorous

nematodes on the bacterial community structure: a microcosm study. Env. Microbiol.

6(7): 733-744.

In Chapter 4 we examined whether the species-specific effects on the decomposition

process and the associated microbial community observed in the monospecific

treatments (Chapter 2) persisted in the combination treatments, and/or whether the

impact was diversity related. To do so, information on microbial activity, bacterial

diversity, and on the decomposition rate in the monospecific treatments and the

combination treatments with two and three species were integrated.

A second topic of this thesis concerns the structuring force of active dispersion

and food availability on a nematode community and the impact of abiotic

disturbances on this distribution in a salt marsh habitat. Chapter 5 describes an

experiment in which cordgrass and controls (plastic strips) in litterbags were

incubated on salt marsh sediment in the laboratory under controlled conditions and in

the field where they were flooded twice daily. Microhabitat preference for phyto-

detritus in a salt marsh habitat has been described by several authors (Bouwman et al.,

1984; Warwick, 1987). Others however have described a strong passive dispersion of


8

meiofauna in general and nematodes in specific in intertidal habitats through tidal

currents (e.g. Sherman and Coull, 1980). We tested whether (i) detritus availability

could impact the distribution of the nematode community in an undisturbed laboratory

environment, (ii) this distribution could be affected by natural disturbances in the

field, and (iii) the contrasting environmental conditions in the field and in the

laboratory would alter the nematode community structure.

This chapter has been submitted for publication as De Mesel, I, Vanaverbeke, J,

Vincx, M, and Moens, T. (submitted) Development of a nematode community on

decaying organic matter under contrasting conditions.

Published as:

De Mesel, I, Derycke, S, Swings, J, Vincx, M, Moens, T (2003) Influence of

bacterivorous nematodes on the decomposition of cordgrass. Journal of Experimental

Marine Biology and Ecology 296: 227-242

Chapter 2

Influence of bacterivorous nematodes on the decomposition of cordgrass

Chapter 2: Species-specific effects on decomposition 11

Abstract

The influence of bacterivorous nematodes (Diplolaimelloides meyli,


the decomposition of a macrophyte (Spartina anglica) in an aquatic environment was

investigated by using laboratory microcosm experiments. Several earlier studies have

shown enhancement of the decomposition process in the presence of nematodes, but

nematode species-specific effects were never tested. In this study four bacterivorous

nematode species were applied separately to microcosms to investigate such species-

specific influences.

No stimulation of the decomposition process nor of the microbial community was

observed in the presence of the nematodes, both were highest in the absence of

nematodes. However, clear differences were found between nematode treatments.

Panagrolaimus paetzoldi reached much higher numbers than the other species,

causing a decrease in microbial activity, probably due to (over)grazing. Remarkably

this low microbial activity did not result in a slow-down of the decomposition process

compared to the other nematode treatments, raising the question whether P. paetzoldi

might be able to directly assimilate detrital compounds. Other nematode species

reached much lower densities, but nevertheless an influence on the decomposition

process was observed. However, this experiment does not support the view that

bacterivorous nematodes enhance decomposition rates.

The experimental results show that in nematode communities the use of functional

groups is inadequate for biodiversity studies. The four nematode species used in this

study belong to the same functional group, but are clearly not functionally redundant

since they all have a different influence on the cordgrass decomposition. This suggests

that the relationship between nematode species diversity and ecosystem functioning

may be idiosyncratic.


Introduction

Nematodes occur in all types of soils and sediments and are often very abundant,

especially in marine and brackish sediments (Heip et al., 1985). They are classified in

trophic or functional groups based on the morphology of their buccal cavity (Wieser,

1953). Nematodes within a functional group are thought to perform the same or very

similar function(s) in ecosystems. This view has recently been questioned by Mikola

and Setälä (1998) based on experimental studies in a soil food web in which species-

specific effects, rather than functional group effects, were observed.

Bacterivorous nematodes are deposit feeders sensu Jensen (1987) which mainly

appear in spots with high microbial activity. They are thought to enhance

decomposition processes through stimulation of the microbial community (Riemann

and Schrage, 1978; Abrams and Mitchell, 1980; Anderson et al., 1981; Findlay and

Tenore, 1982). Three main ways of stimulation have been proposed: (i) bioturbation

resulting in a higher diffusion of oxygen and nutrients (Abrams and Mitchell, 1980;

Aller and Aller, 1992; Alkemade et al., 1992b), (ii) secretion of nutrient rich

compounds such as mucus (Riemann and Schrage, 1978; Riemann and Helmke, 2002)

and (iii) grazing, which implies keeping the bacterial community active and

remineralising nutrients (a.o. Abrams and Mitchell, 1980; Ingham et al., 1985).

The role of nematodes in decomposition processes has mainly been studied in

terrestrial ecosystems (e.g. Abrams and Mitchell, 1980; Anderson et al., 1981; Ingham

et al., 1985). In marine environments similar studies have been scarce.

The lower parts of salt marshes are often dominated by a cordgrass vegetation. Most

of the cordgrass litter is decomposed in situ on the sediment surface, still attached to

the plant (Groenendijk, 1984; Newell et al., 1985; Buth and Voesenek, 1988). As a

consequence laboratory microcosm experiments studying this decomposition process

tend to be artificial. However, manipulative lab experiments are a necessary tool to

study (aspects of) specific processes, such as litter decomposition. In some studies

litter was oven dried, crushed and mixed with sediments (Findlay and Tenore, 1982;

Alkemade et al., 1992b). Alkemade et al. (1992a) incubated cordgrass fragments on

agar which tends to stimulate an explosive growth of bacteria and bacterivorous

nematodes (pers. observation). In a study of Lillebø et al. (1999) leaves were

submerged during the whole experiment. All have shown a stimulation of the

Chapter 2: Species-specific effects on decomposition

14

decomposition of the cordgrass leaves by bacterivorous nematodes, and these results

are at the basis of the general idea that - much as in many terrestrial systems -

nematodes enhance the decomposition of phytodetritus in aquatic systems.

Our study differed from previous studies in that leaves were incubated on water

saturated sediment obtained from the habitat. The aim of this study was to investigate

whether bacterivorous nematodes commonly found on cordgrass detritus indeed

stimulate the decomposition process. Furthermore species-specific influences were

investigated.

The effect of four bacterivorous nematode species was studied (Diplolaimelloides

meyli, Diplolaimelloides oschei, Diplolaimella dievengatensis and Panagrolaimus

paetzoldi). Based on the above mentioned evidence for a stimulatory effect we

assumed that (1) all four nematode species used would stimulate this decomposition

process and that (2) this effect would not be species-specific since they all belong to

the same functional group.

Materials and Methods Experimental set-up

Decomposition of cordgrass litter was studied in laboratory microcosms. These

microcosms consisted of cordgrass fragments on sediment, inoculated with a

microbial suspension and – except in the control treatment – nematodes in sterile petri

dishes. The microcosms were incubated in the dark at 20 ± 2°C.

Sediment, cordgrass, microbiota and nematodes were all collected in the Paulina

saltmarsh (ca. 60 ha) (Westerscheldt Estuary, The Netherlands) in September, 2000.

The sediment was washed with tap water over a 1 mm sieve and the fraction that was

retrieved on a 38 µm sieve was dried overnight at 180°C. 15 ± 0.5 g sediment was

added to sterile petri dishes (∅ 90 mm) and rehydrated with 6.5 ± 1 ml sterile

artificial sea water (ASW, Dietrich and Kalle, 1957) with a salinity of 20. Extra

nutrients were added as a 1/1000 (vol/vol) Killian solution (von Thun, 1966) in

artificial sea water.

Green cordgrass leaves (Spartina anglica) were cut from standing green plants. They

were washed successively with ethanol (70%) , sterile destilled water, ethanol, sterile


destilled water, ethanol and (twice) ASW (Alkemade et al., 1992a). Leaves were air-

dried and 1.9 ± 0.2 g leaves were incubated on the surface of the sediment.

A natural microbial inoculum was added to the petri dishes. The inoculum was

obtained by mixing habitat water, sediment, decomposing and fresh cordgrass leaves

and agar from the nematode cultures (see below). This mixture was filtered over

Whatman GF/C filters to remove flagellates and other small eukaryotes. Every petri

dish was inoculated with 2 ml of this suspension. The microcosms were incubated in

the dark for 24h at 20 ± 2°C before nematodes were added.

Four nematode species, which are common on decomposing cordgrass in the Paulina

salt marsh (Moens, unpublished), were used in our experiment: the monhysterids

Diplolaimelloides meyli, Diplolaimella dievengatensis and Diplolaimelloides oschei

and the rhabditid Panagrolaimus paetzoldi. Agnotobiotic cultures of all species

growing together with unidentified bacteria are available in the Biology Department.

Each microcosm, except for the controls, received 36 individuals of a single nematode

species (Table 1). Nematodes were handpicked with the tip of a fine needle.

Treatment code Nematode species

A Diplolaimelloides meyli

B Diplolaimelloides oschei

C Panagrolaimus paetzoldi

D Diplolaimella dievengatensis

S (no nematodes)

Table 1: Overview of the treatments

Sampling

Samples were taken after 10, 20, 30, 40, 50, 65 and 75 days. On each sampling

occasion three microcosms from every treatment were destructively sampled.

Bacteria were removed from the cordgrass leaves through rinsing with sterile ASW.

After sub-sampling the rinse water to measure respiration rate and enzymatic activity

of the microbial community (see below), nematodes which still adhered to the leaves

were removed with ASW and MgCl2 (8%). Nematodes were preserved in a

formaldehyde solution (final concentration: 4%).


16

The same procedure as at the start of the experiment was used to clean the cordgrass

leaves: washing with ethanol and sterile destilled water (2X), again ethanol and finally

twice sterile ASW. They were then preserved at –20°C for C- and N- analyses.

The sediment was preserved on 4% formaldehyde solution for quantification of the

nematodes.

Weight loss

At every sampling event the cordgrass fragments were dried to the air and weighed.

These values were converted to dry weight according to the equation (De Mesel,

unpublished):

dry weight = 0.2084*EXP(0.6945*weightair dried)

Weight is expressed as the percentage of the initial dry weight of the fragment still

present in the microcosm at the moment of sampling.

weight = (dry weightsampling/dry weightstart) *100

C/N-analysis

The C- and N-concentrations, expressed as percentages, of the organic matter were

determined with a Carlo-Erba element analyser type NA-1500. Plant fragments were

crushed manually using liquid nitrogen. 3 to 5 mg of the litter was used for analysis.

Carbon and nitrogen content are expressed as the percentage of the initial value still

present at the time of sampling.

C-content = startstart

samplingsampling

dry weight*/100)(%Cdry weight*)100/(%C * 100

N-content = startstart

samplingsampling

dry weight*/100)(%Ndry weight*)100/(%N * 100


Nematodes

Nematodes were extracted from the sediment by centrifugation with Ludox (Heip et

al., 1985). Nematodes were stained with Rose Bengal and counted under a dissecting

microscope.

Microbial activity

Total aerobic activity was measured using the Strathkelvin 928 6-channel Dissolved

Oxygen System. 1 ml of the suspension that was rinsed off the cordgrass leaves (see

above) was transferred to each measurement unit. The oxygen concentration was

measured at 20°C during approximately 30 minutes. Respiration rates (µmol O2 h-1)

were calculated from the slope of the regression of oxygen concentration versus time.

The EnzChek® Protease Assay Kit E6638 for green fluorescence (Molecular Probes)

was used to measure the proteolytic activity of the microbial community.

Fluorescently labelled casein served as substrate. It is hydrolysed by metallo-, serine,

acid and sulphydryl proteases secreted by the micro-organisms, thus releasing the

otherwise quenched fluorescent labels into the medium. 100 µl of the suspension

obtained by rinsing the cordgrass leaves was mixed with an equal volume of the

casein containing solution. After an incubation of 30 minutes in the dark the

fluorescence was read on a Victor multilabel counter using excitation and emission

wavelengths of 505 and 513 nm respectively. Readings were performed at 25°C.

Statistical analysis

Univariate 2-way analysis of variance (ANOVA) was used to test for differences of

remaining weight, C-content, N-content, respiration rate and enzymatic activity

between treatments, times and treatment x time. Respiration rate was log-transformed

and enzymatic activity square-root transformed to meet the required assumptions; for

the other effects no transformation was needed. When ANOVA indicated significant

differences the Tukey HSD test was used for pairwise post hoc comparisons.

All tests were done with the STATISTICA software.


18

Results Decomposition

Weight loss

A two-way ANOVA showed significant differences for the effects ‘time’, ‘treatment’

but not for the combined effect ‘time’ x ‘treatment’ (Table 2). For the effect

‘treatment’ the Tukey HSD post-hoc test indicated a significantly lower weight loss in

the presence of D. dievengatensis (treatment D) than in the control treatment S

(p<0.01) or in the presence of D. meyli (treatment A) (p<0.001). The graph (fig. 1)

shows that most of the differences between treatments occurred during the first 50

days of the experiment although no significant differences were found for the

combined effect. By the end of the experiment (day 75) weight losses were similar for

all treatments.

Figure 1: Remaining weight of the cordgrass leaves expressed as percentage of the initial

values (average of 3 replicates ± 1 standard error)

0 10 20 30 40 50 65 75

Days

25

50

75

100

% w

eigh

t rem

aini

ng

D. meyliD. oscheiP. paetzoldiD. dievengatensiscontrol

0 10 20 30 40 50 65 75

Days

25

50

75

100

% w

eigh

t rem

aini

ng

25

50

75

100

% w

eigh

t rem

aini

ng




Table 2: Results of the two-way ANOVA for the effect ‘weight’ (ns: not

significant, **:p<0.01, ***:p<0.001)

Carbon loss

Carbon content in the cordgrass detritus showed a similar pattern as for weight (fig. 2)

in every treatment. The two-way ANOVA showed significant differences for the

effects ‘time’ and ‘treatment’ and for the combined effect ‘time’ x ‘treatment’ (table

3).

Figure 2: C-content of the cordgrass leaves expressed as percentage of the initial values

(average of 3 replicates ± 1 standard error)

Effect DF Effect DF Error F p-level

Time 6 70 112.8143 ***

Treatment 4 70 4.5950 **

Time x Treatment 24 70 1.4714 ns

0 10 20 30 40 50 65 75

Days

25

50

75

100

% C

rem

aini

ng

D. meyliD. oscheiP. paetzoldiD. dievengatensisControl

0 10 20 30 40 50 65 75

Days

25

50

75

100

% C

rem

aini

ng

0 10 20 30 40 50 65 75

Days

25

50

75

100

% C

rem

aini

ng

25

50

75

100

% C

rem

aini

ng




20

As for the variable weight, C- content of the cordgrass differed significantly between

treatment D (D. dievengatensis) and on the one hand the control treatment S (p<0.01)

and on the other hand treatment A (D. meyli) (p<0.01). In addition the latter two

differed significantly from treatments B (D. oschei) (both p<0.05) and C (P.

paetzoldi) (p<0.01 and p<0.05 respectively). The post hoc test for the combined effect

showed an unique significant difference between treatment C (P. paetzoldi) and the

control treatment S (p<0.05) after 40 days.

Table 3: Results of the two-way ANOVA for the effect ‘Carbon content’

(*:p<0.05, ***:p<0.001)

Nitrogen loss

The N content of the cordgrass fragments, expressed as a percentage of the original

value, is shown in figure 3. The results of the two-way ANOVA are given in table 4.

As shown by the Tukey HSD post hoc test the remaining nitrogen content of the

detritus was significantly higher in treatment D (D. dievengatensis) than in all other

treatments (p<0.001) and significantly lower in control treatment S than in treatments

Table 4: Results of the two-way ANOVA for the effect ‘N-content‘

(***:p<0.001)

A (D. meyli) and C (P. paetzoldi) (both p<0.001). From day 10 until day 40 the

nitrogen content of the cordgrass leaves differed significantly between treatments D

(D. dievengatensis) and control treatment S (day 10, day 20 and day 30: p<0.001; day

40: p<0.01); significant differences between treatment D and treatment B (D. oschei)


Time 6 70 126.8398 ***

Treatment 4 70 7.6499 ***

Time x Treatment 24 70 1.9895 *


Time 6 70 62.10779 ***

Treatment 4 70 33.94058 ***

Time x Treatment 24 70 3.17894 ***


were found from day 20 until day 40 and again on day 65 (day 20: p<0.01, day 30:

p<0.001, day 40: p<0.01 and day 65: p<0.05).

Figure 3: N-content of the cordgrass leaves expressed as percentage of the initial values

(average of 3 replicates ± 1 standard error)

Nematodes

Figure 4: Number of nematodes per microcosm (log scale) (average of 3 replicates ± 1

standard error)

0

25

50

75

100

0 10 20 30 40 50 65 75

% N

rem

aini

ng


0

25

50

75

100

0

25

50

75

100

0 10 20 30 40 50 65 75

% N

rem

aini

ng


Days

10 20 30 40 50 65 75

Days

1

10

100

1000

10000

100000

Num

ber o

f nem

atod

es (l

og)

t

D. meyliD. OscheiP. paetzoldiD. dievengatensis

10 20 30 40 50 65 75

Days

1

10

100

1000

10000

100000

Num

ber o

f nem

atod

es (l

og)

t

D. meyliD. OscheiP. paetzoldiD. dievengatensis


22

From day 30 onwards, P. paetzoldi (treatment C) reached densities 10 to 103 times

higher than those of other nematodes (fig. 4). Densities up to 125000 individuals per

microcosm were recorded.

The differences between the abundances of D. meyli (treatment A), D. oschei

(treatment B) and D. dievengatensis (treatment D) were smaller, D. meyli reaching the

lowest nematode numbers on the average. High variation between replicates was

observed.

Microbial activity

Respiration rate

Aerobic activity of the microbial community, represented by the respiration rate, was

high at the start of the experiment in all treatments (fig. 5A); it steadily decreased and

stabilised from 40 days onwards at lower respiration rates. Significant differences

were indicated by two-way ANOVA for the effects ‘time’, ‘treatment’ and ‘time’ x

‘treatment’. Tukey HSD post-hoc test showed significantly lower respiration rates in

the presence of P. paetzoldi (treatment C) than in the control (treatment S) (p<0.001).

Figure 5: Respiration rate of the microbial community (average of 3 replicates ± 1 standard

error)

In all treatments, respiration rates on days 10 and 20 differed significantly with those

from 40 days onwards (all p<0.001); within treatment C (P. paetzoldi) differences

00.10.20.30.40.50.60.70.80.9

1

10 20 30 40 50 65 75

Days

Res

pira

tion

rate

(µm

ol O

2h-1

)

D. oscheiP. paetzoldiD. dievengatensiscontrol

D. meyli

00.10.20.30.40.50.60.70.80.9

1

00.10.20.30.40.50.60.70.80.9

1

10 20 30 40 50 65 75

Days

Res

pira

tion

rate

(µm

ol O

2h-1

)

D. oscheiP. paetzoldiD. dievengatensiscontrol


D. meyli


were already significant between day 10 and day 30 (p<0.001), indicating a faster

decrease in respiration rate in this treatment.

DF Effect DF Error F p-level

Time 6 70 139,8768 ***

Treatment 4 70 8,085291 ***

Time x Treatment 24 70 1,849047 *

Table 5: Results of the two-way ANOVA for the effect ‘respiration rate’

(*:p<0.05, ***:p<0.001)

Enzymatic activity

Enzymatic activity, which is a measure of aerobic as well as anaerobic microbial

activity, increased at the beginning of the experiment and decreased at the end in most

treatments except in treatment C (P. paetzoldi). In the latter, enzymatic activity was

low during the whole experiment. The Tukey HSD test following two-way ANOVA

(table 6) confirmed that enzymatic activity in the presence of P. paetzoldi was

significantly lower than in other treatments (all p<0.001).

Figure 6: Enzymatic activity of the microbial community (average of 3 replicates ± 1 standard

error)

10 20 30 40 50 65 75

Days

0

5000

10000

15000

20000

25000

Enzy

mat

ic a

ctiv

ity (c

ount

s min

-1)

10 20 30 40 50 65 75

Days

0

5000

10000

15000

20000

25000

Enzy

mat

ic a

ctiv

ity (c

ount

s min

-1)

D. oschei

P. paetzoldi D. dievengatensis control

D. meyli


24


Time 5 59 7,639149 ***

Treatment 4 59 19,04666 ***

Time x Treatment 20 59 3,603706 ***

Table 6: Results of the two-way ANOVA for the effect ‘enzymatic activity’

(***:p<0.001)

Discussion Decomposition process

The major differences in the decomposition process between treatments were

observed during the first 40 days. At the end of the experiment, after 75 days, the

decomposition process in all treatments was in a similar phase as shown by the weight

loss and the elemental analyses. At that time presumably all easily degradable

compounds were decomposed and mainly high molecular weight compounds, such as

cellulose and lignin, were left over (Valiela, 1995). The differences in the

decomposition process between treatments the first 40 days indicate that the final,

comparable situation is reached at different speeds, depending on the presence or

absence and the identity of the nematode species. This demonstrates the importance of

the duration and the sampling interval in these studies: short time experiments or

unique samplings at the end of a long experiment can cause misleading conclusions.

The high nitrogen loss from the cordgrass leaves in all treatments during the first 10

days is a consequence of the leaching of nutrient rich compounds, mainly proteins

(Valiela, 1995), and was probably enhanced by the cleaning of the leaves prior to the

experiment.

The chemical dynamics in the cordgrass leaves are considered to be best suited to

express the decomposition process. In the absence of macrofauna no detritus

fragments are grazed away and the decomposition of the detritus is a result of the

selective removal of organic compounds by the microbiota. Detailed information

about the selective use of organic compounds would be lost when looking only at

weight loss, especially when they make up only small amounts of the total weight,

like nitrogen (2.4% of total weight at the start of the experiment, data not shown).


Microbial activity

Bacteria and fungi are the principal actors in decomposition processes (Valiela, 1995).

Although microbial activity was not measured in situ, respiration rate and enzymatic

activity give a good indication of the microbial activity in the microcosms and thus

enable us to compare between treatments.

The high initial aerobic activity in each treatment can partly be explained by the

leaching of easily degradable compounds from the detritus which are immediately

broken down by the microbial community. The aerobic activity declined during the

first part of the experiment and stabilised from 40 days onwards at lower respiration

rates. The major part of the respiration measured during the decomposition of

cordgrass is considered to be ‘non-bacterial’, most probably fungal (Padgett et al.,

1985). The increase of the total microbial activity, indicated by the enzymatic activity,

while the aerobic activity, represented by the respiration rate, is decreasing, suggests a

shift from aerobic to anaerobic activity in the microbial community. Since aquatic

fungi in general are aerobic (Rheinheimer, 1992), it can be assumed that the

enzymatic activity represents mainly bacterial activity. These findings suggest that

initially the fungal activity dominates and is then taken over by bacterial activity.

Nematodes and their influence on the microbial activity and

decomposition process

The four nematode species used in this experiment (Diplolaimelloides meyli,

Diplolaimelloides oschei, Diplolaimella dievengatensis and Panagrolaimus paetzoldi)

belong to the same functional group, the deposit feeders (Jensen, 1987) and are

assumed to feed mainly on bacteria (Moens and Vincx, 1997). Panagrolaimidae are

enrichment opportunists (Bongers et al., 1995): they are the first to colonise organic

deposits (Bouwman et al., 1984; Warwick, 1987; Bongers et al., 1991). Monhysterids,

to which the three other nematode species belong, are general opportunists and appear

in an early stage at locations with high organic input (Warwick, 1987; Alkemade et

al., 1993). They are typical ‘Aufwuchs species’ indicating their occurrence in

association with decaying organic matter but not, or to a much lesser extent, in the

surrounding sediments (Bouwman et al., 1984).


26

A faster initial increase of P. paetzoldi densities was expected since their generation

time is shorter than of the monhysterids (Moens and Vincx, 1998). To our surprise the

numbers of the monhysterids remained rather low throughout the experiment while

high densities of P. paetzoldi were reached and sustained. It is not likely that this

discrepancy is due to a lack of suitable food sources for the monhysterids. Cultivation

experiments of bacterivorous nematodes on different strains of bacteria, either

arbitrarily chosen or isolated from the natural habitat, have shown that bacterivorous

nematodes can grow and reproduce on a wide range of bacteria (pers. observations,

unpublished; Sohlenius, 1968; Venette and Ferris, 1998). Since the cordgrass

fragments were inoculated with a mixture of bacteria isolated from their natural

habitat the required diet must have been available for each nematode species.

An alternative explanation for the low monhysterid numbers might be found in the

presence of high ammonium concentrations building up in the microcosms due to

intense microbial activity. Although both panagrolaimids and monhysterids occur in

polluted sediments, monhysterids in general show a more sensitive reaction to

chemical compounds in lab experiments then do Panagrolaimidae (pers.

observations). These low numbers of the monhysterids may have influenced our

results. In situ nematode densities are in general one order of magnitude higher.

Moreover, other experiments, such as those by Alkemade et al. (1992a;b), in which

results contrasted with ours, obtained considerably higher nematode numbers.

However, Freckman (1988) noted that nematode biomass should not necessarily be

high to have an influence on the decomposition process. Since differences in

decomposition between treatments, and especially with the control treatment, were

found (see below) we assume that the numbers reached in all treatments were high

enough to have some influence on the decomposition process.

In analogy to many terrestrial soil systems, benthic bacterivorous nematodes are

thought to enhance decomposition processes by stimulating the activity of the

microbial, especially the bacterial, community (Findlay and Tenore, 1982; Alkemade

et al., 1992a,b). This was not confirmed by our experiments. Microbial activity was

amongst the highest in the absence of nematodes, resulting in a higher decomposition

rate. Observations in the course of the experiments suggested that the fungal growth

in the presence of the nematodes was lower compared to the control treatment. This is

confirmed by the lower respiration rate, mainly originating from fungal activity (see

above), in the presence of nematodes. Growth inhibition of fungal mycelia in the


presence of bacterivorous nematodes has already been reported in commercial

mushroom cultures (Kaufman et al., 1983; Grewal, 1991; Grewal and Richardson,

1991; Grewal and Hand, 1992). Since fungi are known to be important in the early

stages of decomposition of cordgrass (Newell et al., 1989; Valiela, 1995) this could be

a reason for the initially slower decomposition process in the presence of nematodes.

The contrast with the former studies, in which generally enhancement of the

decomposition process by nematodes is reported, may partly be explained by

differences in the experimental set up (Findlay and Tenore, 1982; Alkemade et al.,

1992a,b; Lillebø et al., 1999). When crushed cordgrass leaves were mixed with salt

marsh sediments bioturbation was found to be the most important factor for

stimulating the microbial community and the decomposition process (Alkemade et al.,

1992b). Through the movement of nematodes diffusion of oxygen and nutrients is

stimulated, resulting in a higher microbial activity and higher decomposition rate.

This factor is probably of minor importance on intact cordgrass fragments, as in our

experiment, or in the salt marshes where leaves decompose attached to the plant. Only

in late stages of the decomposition process the detritus becomes mixed with the

sediment and bioturbation is likely to play a role.

Grazing by bacterivorous nematodes on the bacteria has also been reported as a source

of stimulation (a.o. Abrams and Mitchell, 1980). Grazing can have two important

consequences. Firstly the bacterial community is kept in the active growth phase,

implying a higher demand of nutrients by the bacteria and thus resulting in a higher

decomposition rate of the organic matter. Secondly, when feeding on bacteria, an

excess of nutrients is consumed, because of the higher C/N ratio of nematodes

compared to bacteria (Anderson et al., 1983; Ferris et al., 1997). After excretion, this

nitrogen is again available for the bacterial community or can bind to the cordgrass

detritus (White and Howes, 1994; Valiela, 1995). It can be questioned whether the

recycled nitrogen can be of importance for the activity of bacteria in eutrophic

systems such as salt marshes. We assume that this can only enhance microbial activity

and consequently the decomposition process in N-limited systems, such as many

terrestrial soil systems.

Several elements indicate that nematode grazing had an important influence on the

microbial activity in our experiment. Firstly the low bacterial activity in the presence

of P. paetzoldi is most probably due to overgrazing since high feeding rates on

bacteria have been observed for this species (Moens, unpubl) and the decrease in


28

activity occurred when nematode densities reached over 50000 individuals per

microcosm. Secondly the absorption of excreted N to the detritus may partly explain

the high nitrogen content in the cordgrass leaves in the presence of the nematodes.

This could mask the selective removal of nitrogen rich compounds from the detritus

by the microbial community.

Many marine and brackish-water nematodes are known to secrete mucus which is

rapidly colonised by high bacterial densities (Riemann and Schrage, 1978; Nehring et

al., 1990; Jensen, 1996). Recently Riemann and Helmke (2002) suggested a

commensalistic relationship between the mucus-secreting nematodes and the

associated bacteria. Nematodes are believed to discharge hydrolytic enzymes in the

mucus which break down, alone or together with the bacterial enzymes, chemical

compounds. The hydrolysed compounds, such as sugars, could then be directly

consumed by the nematodes, thus yielding extra nutrients directly from the detritus. A

better ability of P. paetzoldi to secrete the necessary enzymes and subsequently

assimilate the organic compounds could explain (i) why the decomposition at the end

of the experiment in their presence is similar to other treatments despite the reduced

microbial activity and (ii) why they can sustain high densities while their expected

food source, bacteria, is being repressed. This trait of P. paetzoldi might facilitate

their growth in sites with high organic input (Bongers et al., 1995).

Bacteria growing on the mucus tracks could be nematode specific (Riemann and

Schrage, 1978; Warwick, 1981; Jensen, 1996), implying a potentially different

influence of the bacterial community on the decomposition process in the presence of

each nematode species. The results of a bacterial community analysis from this

experiment are presented in a next paper1 (De Mesel et al., 2004).

Conclusions

Despite the lack of stimulation by the nematodes of the decomposition process and of

the activity of the associated microbial community, differences were found dependent

on the identity of the nematode species present. Panagrolaimus paetzoldi was able to

reach and sustain higher densities even though the bacterial community was

repressed, suggesting a different life strategy compared to the three monhysterid

1 see Chapter 3


nematode species. Within the closely related monhysterids differences in population

development and in influences on the decomposition process were seen. Our results

show that the use of ‘functional groups’ is rather inadequate when studying the

relationship between nematodes and ecosystem processes. Integrating information on

life strategies might be more reliable. These findings also imply that the idea of

species redundancy, assuming that species within a functional group have the same or

a very similar influence on an ecosystem process, might not be applicable in

nematode communities. It seems that the influence of every species is, or can be,

different, agreeing with the conclusions of Mikola and Setälä (1998), that

relationships between nematode species diversity and ecosystem functioning may be

largely idiosyncratic.

Finally, a comparison of our results with those of previous studies highlights the

influence of experimental conditions on the obtained results. As a consequence

caution is required when extrapolating results to a natural situation. However, the

value of controlled experiments to generate hypotheses about in situ processes should

not be underestimated since many processes can not be studied in the field due to

practical problems.

Acknowledgements

We thank Annick Van Kenhove for the chemical analyses and Wim Bert for

identification of one of the nematode species. Prof. Dr. J. Vanfleteren is

acknowledged for the use of the Victor Multilabel counter. Jan Vanaverbeke and two

anonymous referees provided constructive comments on the manuscript. The first and

second author acknowledge grants from the Flemish Institute for the Promotion of

Scientific-Technological Research (I.W.T.). The last author is a postdoctoral fellow

with the Flemish Fund for Scientific Research. Further financial support was obtained

from Ghent University in BOF- projects 1205398 and 011060002.

Published as:

De Mesel, I, Derycke, S, Moens, T, Van der Gucht, K, Vincx, M, Swings, J (2004)

Top-down impact of bacterivorous nematodes on the bacterial community structure: a

microcosm study. Environmental Microbiology 6(7): 733-744

Chapter 3

Top-down impact of bacterivorous nematodes on the bacterial community structure: a microcosm experiment

Chapter 3: Impact of nematodes on a bacterial community

33

Abstract

The influence of bacterivorous nematodes (Diplolaimelloides meyli,


the development of a bacterial community growing on decaying cordgrass detritus

was studied in laboratory microcosm experiments. Cordgrass leaves were incubated

on a sediment surface with a natural bacterial mixture containing bacteria from

sediment, cordgrass detritus and habitat water. The four nematode species were

applied separately to the microcosms; controls remained without nematodes. Samples

were taken seven times over a 65-day period. The bacterial community structure was

analysed by means of DGGE of the 16S rRNA genes. Multi Dimensional Scaling

showed grouping of the samples per treatment. Analysis of Similarities indicated that

the differences between treatments were significantly larger than differences within

treatments. Our results suggest that nematodes can have a significant structuring top-

down influence on the ‘pool’ of bacteria growing on the detritus, even at low

densities. Dissimilarities were similar between all treatments. Differences in bacterial

community composition within the treatments with monhysterids (D. meyli, D. oschei

and D. dievengatensis) can be explained by species-specific food preferences.

Panagrolaimus paetzoldi on the other hand feeds unselectively, and thus affects the

bacterial community differently. A top-down effect of the nematodes on the diversity

of the bacterial community was only evident under high grazing pressure, i.e. in the

presence of P. paetzoldi.


35

Introduction

Heterotrophic bacteria and fungi are an important link in the functioning of many

ecosystems, particularly for the mineralization of nutrients (Valiela, 1995). An

increase in nutrient mineralization has been reported in the presence of grazers of

bacteria, mainly nematodes and protists (Johannes, 1965; Ferris et al., 1997;

Bonkowski et al., 2000). Grazers can affect bacterial communities at different levels.

They are able to influence bacterial activity, either stimulatory (Findlay and Tenore,

1982; Alkemade et al., 1992a,b) or inhibitory (De Mesel et al., 2003). This can be a

direct effect of grazing (Ingham et al., 1985), but bioturbation (Abrams and Mitchell,

1980; Alkemade et al. 1992b) and secretion of mucus trails (Riemann and Schrage,

1978; Jensen, 1996) by nematodes can also be important. Grazing impacts on the

bacterial community structure have also been reported. Protists can cause a size shift

by grazing the medium-sized part of the community (Hahn and Höfle, 2001). Van

Hannen et al. (1999) were the first to demonstrate changes in the bacterial species

composition under protist grazing pressure, while Rønn et al. (2002) found that these

changes were protist-species specific. The structuring influence of nematodes,

however, has not yet been studied well. Only Griffiths et al. (1999) reported bacterial

shifts under nematode grazing using phospholipid fatty acid analysis, community

level physiological profiling (using BIOLOG plates) and DNA fingerprinting

(Denaturing Gradient Gel Electrophorsis). The shifts in the bacterial community

composition would presumably be the result of selective grazing. For example, large

and filamentous bacterial cells can escape uptake by some nematodes due to their

small buccal cavity (Tietjen, 1980). Additionally, experimental data suggest that some

nematode species can distinguish between and preferentially consume different strains

of bacteria, the preference differing between even closely related nematode species

(Tietjen and Lee, 1977; Moens et al., 1999).

The aims of the present paper were: (i) to investigate the potential effect of

bacterivorous nematodes on the bacterial community composition and diversity and

(ii) to assess whether any such effects are nematode-species specific. Cordgrass

(Spartina alterniflora) detritus was used as a substrate for the bacterial community.

Cordgrass is often the dominant vegetation in the lower parts of salt marshes and

decomposes mainly in situ, on the sediment surface (Groenendijk, 1984; Newell et al.,


36

1985; Buth and Voesenek, 1988). High densities of bacterivorous nematodes occur on

the detritus (Montagna and Ruber, 1980; Reice and Stiven, 1983; Buth and de Wolf,

1985; Hemminga and Buth, 1991).

Four bacterivore nematode species commonly found on decaying cordgrass leaves

(Diplolaimelloides meyli, Diplolaimelloides oschei, Diplolaimella dievengatensis and

Panagrolaimus paetzoldi) were inoculated separately in microcosms containing

cordgrass detritus and a natural bacterial inoculum; controls without nematodes were

incubated in parallel. P. paetzoldi (Rhabditida) is an extreme coloniser, appearing

promptly at sites with high organic input where they can quickly reach high densities.

The three other species belong to the Monhysterida and are also characteristic for

organically enriched sites. Rhabditida and Monhysterida generally dominate the

nematode community on decaying phytodetritus in salt marshes (Warwick, 1987).

The choice of the three species within a single family (Monhysteridae) allows testing

for different effects on the bacterial community between closely related grazer

species, with a similar life strategy.

At seven sampling occasions over a period of 65 days, the bacterial communities were

analysed by means of Denaturing Gradient Gel Electrophoresis (DGGE) of the 16S

rRNA gene. DGGE allows a fast screening of complex bacterial communities, without

going into detail on community assemblage or functional aspects of the bacterial

community, which was beyond the scope of this study. The use of DGGE was first

introduced in ecological studies by Muyzer et al. (1993), and has since proven to be

very useful in analysing bacterial communities originating from a large number of

habitats (Muyzer, 1999). Recently, this technique has been applied successfully in a

number of experimental studies (Degans et al., 2002; Massana and Jürgens, 2003;

Zöllner et al., 2003).


37

Materials and Methods Experimental set-up and sampling

The experimental set-up has been described in detail in De Mesel et al. (2003).

Briefly, green cordgrass leaves (Spartina anglica) collected in the Paulina Salt Marsh

(Westerscheldt, The Netherlands) were cleaned with ethanol (70%) and artificial sea

water (ASW) (Dietrich and Kalle, 1957), put in petri dishes (∅ 90 mm) on cleaned

sediment saturated with ASW with a salinity of 20, and inoculated with a microbial

inoculum. This inoculum consisted of bacteria from fresh and decomposing cordgrass

leaves, sediment and habitat water and from the nematode cultures. This mixture was

filtered over Whatman GF/C filters to remove flagellates and other small eukaryotes.

All petri dishes were inoculated with 2 ml of this suspension. After an incubation of

24h, nematodes were added to the microcosms.

Four nematode species, commonly found on decaying cordgrass in the Paulina Salt

Marsh (Moens, unpublished) were selected: Diplolaimelloides meyli,

Diplolaimelloides oschei, Diplolaimella dievengatensis and Panagrolaimus paetzoldi.

These species are available in monospecific cultures in the Marine Biology Section.

Each microcosm, except for the controls, received 36 individuals of a single nematode

species.

Microcosms were incubated in the dark at 20°C.

Three replicate microcosms per treatment were destructively sampled after each of the

following incubation times: 10, 20, 30, 40, 50, 55 and 65 days. However, for the

following days and treatments, only two replicates were successfully analysed: D.

meyli: day 55; D. oschei: day 30, 50 and 65; P. paetzoldi: day 40; control: day 40 and

55. Detritus fragments were homogenized in sterile ASW with a salinity of 20 to

collect the bacterial community. 4 ml of this suspension was centrifuged at 13000

rpm. After removing the supernatant the pellet was preserved at –80°C.


38

DNA-extraction and -purification

DNA was extracted as described by Zwart et al. (1998) using the bead-beating method

concomitant with phenol extraction and ethanol precipitation. The DNA was purified

on a Wizard column (Promega, Madison, WI, USA) according to the manufacturer’s

recommendations.

PCR and DGGE

The amplification of DNA fragments and the DGGE analysis were performed as

described by Van der Gucht et al. (2001). DNA fragments of about 250 bp were

amplified using primers for the V3-region: F357-GC primer (5’-

CGCCCGCCGCGCCCCGCGCCCGGCCCGCCGCCCCCGCCCCCCTACGGGAG

GCAGCAG-3’) and the R518 primer (5’-ATTACCGCGGCTGCTGG-3’) which are

specific for most bacteria (Muyzer et al., 1993). The temperature cycling started with

a preincubation of 5 minutes at 94°C. 20 cycles were performed consisting of

denaturation at 94°C for 1 minute, annealing at 65°C for 1 minute, with the

temperature decreasing 0.5°C every cycle until a temperature of 56°C was reached,

and primer extension at 72°C for 1 minute. During the next 5 cycles the annealing

temperature stayed at 55°C for 1 minute. Finally temperature was held at 72°C for 10

minutes.

DNA concentrations were checked visually by analysing 5µl of the product on 1%

agarose gel, staining with ethidium bromide, and comparison with a molecular weight

marker (Smart ladder, Eurogenetic). 35µl of the samples were loaded on the DGGE

gels. The linear gradient on the DGGE-gels increased from 40% at the top to 75% at

the bottom. To compare the banding patterns between gels a marker was loaded and

some experimental samples were run on different gels. Additionally the information

on the DNA sequence of a number of bands was used to make a reliable alignment of

the gels.

The software package Bionumerics 5.1 (Applied Maths BVBA, Kortrijk, Belgium)

was used to analyse the banding patterns. The Bionumerics software measures an

optical density profile through each lane (corresponding to a single sample), identifies

band positions, and calculates the percent contribution of the intensity of each band to


39

the total intensity of the lane. This procedure yielded a matrix with the relative

intensity of each band in all samples.

Sequencing of excised DGGE bands

The procedure used was based on Van der Gucht et al. (2001), but with some small

modifications. Individual bands were excised from the DGGE-gel and put in a sterile

recipient to which 50 µl 1X TE was added. DNA could diffuse form the acrylamide

gel to the TE-buffer during incubation overnight at 4°C. 5µl of the buffer was used for

amplification (PCR mix and temperature cycle as described above) and the position of

the band was checked on a DGGE gel. DNA from the bands was then amplified with

the F357GC primer (5’-CGCCCGCCGCGCCCCGCGCCCGGCCCGCCGCCCCCG

CCCCCCTACGGGAGGCAGCAG-3’) and the R518+ primer (5’-GCGTTCTTCAT

CGTTGCGAGATTACCGCGGCTGCTGG-3’) using the temperature cycling as

described by Van der Gucht et al. (2001), followed by a sequencing reaction with a

Big Dye TM terminator ready reaction kit (Applied Biosystems, Foster City, CA,

USA) and the Stef1Tex primer (5’-GCGTTCTTCATCGTTGCGAG -3’). The

temperature cycling was as described by Zwart et al. (1998). Sequencing reactions

were analysed on an Applied Biosystem 3100 genetic analyser.

Bands were identified by comparing the partial 16S rRNA gene sequences with

GenBank and EMBL sequences with BLAST (see table 2) (Altschul et al., 1990).

Bacterial diversity

The Shannon Wiener Index (H’) was used to express bacterial diversity and was

calculated as explained by Magurran (1988), based on the relative intensities of the

bands:

H’ = ∑=

=

−ni

iii pp

1ln

n represents the number of bands in the sample and pi the relative intensity of the ith

band. The index indicates how the intensity of the bands is distributed within a

sample.


40

Statistical analysis of the DGGE fingerprints

The banding profiles of the DGGE gels were analysed by means of Multi

Dimensional Scaling (MDS) using the PRIMER software package and based on the

arcsin squareroot transformations of the relative band intensities. The Bray-Curtis

index was chosen as similarity coefficient. MDS analysis places samples in a multi

dimensional space, based on the similarities between them. This multi-dimensional

space can be represented by a 2D ordination plot, with the stress value indicating how

faithfull the high-dimensional relationships between the samples are represented in

the 2D plot. This plot is very easy to interpret as similar samples are placed close

together, and distance between samples increases as similarity decreases (Clarke,

1993; Clarke and Gorley, 2001; Clarke and Warwick, 2001).

Analysis of similarities (ANOSIM) is a non-parametric permutation procedure, based

on the similarity matrix underlying the ordination of the samples. ANOSIM was used

to test for significant differences between treatments (Clarke, 1993; Clarke and

Gorley, 2001; Clarke and Warwick, 2001).

Similarity percentage analysis (SIMPER) was used to weight the contribution of each

band to the similarity/dissimilarity within/between treatments and to seek for those

bands which were important during the succession in each treatment (Clarke and

Gorley, 2001; Clarke and Warwick, 2001).

Univariate 2-way analysis of variance (ANOVA) and the Tukey HSD post hoc test

were used to test for differences in the Shannon-Wiener Index (H’) between

treatments, times and treatment x time (Statistica software).

Nucleotide sequence accession numbers

All partial sequences have been deposited in the EMBL database under accession nos

AY510428-AY510447


41

Results and Discussion Effects on bacterial community composition

The bacterial community composition was analysed by means of Multi Dimensional

Scaling (MDS) and Analysis of Similarities (ANOSIM), both non-parametric

multivariate tests for changes in community structure (Clarke, 1993; Clarke and

Warwick, 2001). The use of either relative band intensities, or of a reduced data

matrix based on presence/absence of bands, to analyse community structure has been

discussed and questioned before (Schauer et al., 2000). Both are commonly used,

depending on the type of dataset studied. We believe that the use of relative band

intensities in this study was justified because bands could clearly be seen fading away

or appearing and intensifying through the succession. Using a binary dataset would

strongly influence our analysis by overestimating very weak, sometimes uncertain

bands. Moreover, Muylaert et al. (2002) showed that the use of relative band

intensities is better suited for ordination analysis than a binary data matrix.

Ordinations have been found very useful in ecological research in general and for

processing community analyses by means of DGGE in specific (Fromin et al., 2002).

Figure 1: MDS ordination plot based on the relative band intensities of the DGGE profiles,

with the letter referring to the treatments and the number to the sampling day.

D. meyli (A)

D. oschei (B)

P. paetzoldi (C)

D. dievengatensis (D)

control (S)

A10

A10

A10A20

A20

A20

A30

A30

A30

A40 A40A40

A50

A50

A50A55

A55

A65

A65

A65

B10

B10

B10

B20

B20

B20

B30 B30

B40B40

B40B50

B50

B55

B55B55

B65

B65

C10

C10C10 C20

C20 C20

C30 C30

C30

C40

C40

C50

C50

C50C55

C55

C55

C65

C65

C65

D10

D10

D10D20

D20

D20D30

D30D30 D40

D40D40

D50D50

D50D55D55

D55D65

D65

D65

S10

S10

S10

S20S20 S20

S30S30

S30

S40

S40

S50

S50

S50

S55

S65

S65

S65

Stress: 0.25

D. meyli (A)

D. oschei (B)

P. paetzoldi (C)


control (S)

D. meyli (A)

D. oschei (B)

P. paetzoldi (C)


control (S)

A10

A10

A10A20

A20

A20

A30

A30

A30

A40 A40A40

A50

A50

A50A55

A55

A65

A65

A65

B10

B10

B10

B20

B20

B20

B30 B30

B40B40

B40B50

B50

B55

B55B55

B65

B65

C10

C10C10 C20

C20 C20

C30 C30

C30

C40

C40

C50

C50

C50C55

C55

C55

C65

C65

C65

D10

D10

D10D20

D20

D20D30

D30D30 D40

D40D40

D50D50

D50D55D55

D55D65

D65

D65

S10

S10

S10

S20S20 S20

S30S30

S30

S40

S40

S50

S50

S50

S55

S65

S65

S65

Stress: 0.25


42

The MDS analysis on all samples from each sampling event showed a clustering per

treatment (fig. 1). A pronounced time aspect was observed in the presence of

nematodes, but was less clear in the control. Although the stress value, indicating how

faithful the high-dimensional relationship among the samples is represented in the 2D

plot, was rather high, ANOSIM indicated that the variation in DGGE profiles was

higher between samples from different treatments than between samples within one

treatment (table 1). Similar observations have been made under protist grazing

pressure in terrestrial soil systems (Rønn et al., 2002). The pairwise comparisons

further showed that dissimilarities between all treatments were similar.

Table 1: Results of the ANOSIM routine to analyse the differences between treatments.

A first explanation for the differences between treatments may be related to

differences in feeding habits between the nematode species used here.

Panagrolaimidae are typical enrichment opportunists (Bongers et al., 1995). Direct

observations indicate they non-selectively consume food particles at high rates as they

show a constant pumping activity of the oesophagus (Moens, unpublished). Since the

passage through the gut is very fast, it is plausible that not all ingested food particles

are digested, implying that some viable cells are again released. The subset of

bacterial cells that maintains viability after passage through the gut is assumed to be

protected mainly by their cell wall. Within the Monhysteridae however, bacterial

selection would rather occur at the level of food ingestion. Moens et al. (1999) carried

out a number of experiments with the monhysterid species used in our study, showing

they can select between bacterial strains. Chemotaxic responses in which ‘taste’ and

‘smell’ of the prey would be of importance may determine this selection (Moens et

Comparison R Statistic

Significance level (%)

D. meyli vs D. oschei 0.727 0.1 D. meyli vs P. paetzoldi 0.838 0.1 D. meyli vs D. dievengatensis 0.839 0.1 D. meyli vs control 0.769 0.1 D. oschei vs P. paetzoldi 0.689 0.1 D. oschei vs D. dievengatensis 0.637 0.1 D. oschei vs control 0.774 0.1 P. paetzoldi vs D. dievengatensis 0.448 0.1 P. paetzoldi vs control 0.602 0.1 D. dievengatensis vs control 0.652 0.1


43

al., 1999 and references therein). This implies that chemical cues emanating from the

bacteria rather than bacterial cell wall structure influence their palatability and uptake

by monhysterid nematodes. The nematode-species-specific preference for different

bacterial strains, observed by Moens et al. (1999), may have caused the development

of different bacterial communities in the presence of each monhysterid nematode

species.

The structure of the bacterial community may also have been affected by some other

nematode traits that have been found important in stimulating bacterial activity.

Bioturbation by the nematodes enhances diffusion of nutrients and oxygen through

sediments (Abrams and Mitchell, 1980; Alkemade et al., 1992b; Aller and Aller,

1992) and could thus affect the bacterial community structure. However, we assume

that bioturbation is mainly of importance when organic matter is mixed in the

sediment, and not, as in our experiment, when larger organic fragments decompose on

the sediment surface (De Mesel et al., 2003). Nematodes can secrete mucus trails

which are characterised by a dense growth of bacteria (Riemann and Schrage, 1978;

Jensen, 1996). These bacteria could be nematode-species-specific, but research on

their identity is, to our knowledge, non-existent. However, this kind of ‘cultivation’ of

bacteria by the nematodes is presumably of minor importance, since most bands can

be found in two or more treatments, but with other relative intensities and/or different

frequencies.

Sequencing of DGGE-bands provided some taxonomic information on the bacteria.

The DNA fragments we cut out of the DGGE gel were 200bp at most; therefore

information at the species level could not be obtained. The identification is restricted

to the genus or higher taxonomic level (see table 2)

Most of the sequences showed high similarities (>95%) with species commonly found

in marine and brackish environments. Members of the genus Clostridium are often

found in marine sediments and on decaying vegetation and are characteristic for

anaerobic environments. They have the ability to break down a large number of

organic compounds (Hippe et al., 1992). Cytophaga species are marine cellulose

degraders (Reichenbach, 1992), while Neptunomonas species are marine bacteria

breaking down aromatic hydrocarbons (Hedlund et al., 1999). Oceanospirillum

species are often associated with particulate plant matter in marine sites (Pot et al.,

1992). One Arcobacter species (A. nitrofigilis) has been described associated with


44

Table 2: Sequence similarities of the excised DGGE bands; for the position on the DGGE

fingerprints, see fig. 2. Unidentified species 7 to 9 only showed very low similarity with

sequences of the database.

Spartina roots in salt marshes (McClung et al., 1983). Sulfurospirillum and

Bdellovibrio species occur in sediments of marine and freshwater systems (Ruby,

1992; Kersters et al., 2003). Unidentified species 1 belongs to the CFB-group

bacteria. Unidentified species 2 to 6 showed high similarities with other partial

Band no.

Band name Closest relative Percentage similarity

Accession no.

1 Clostridium sp.1 Clostridium algidixylanolyticum

99% AY510428

2 Clostridium sp.2 Clostridium lentocellum

95% AY510429

3 Clostridium sp.3 Clostridium sp. 97% AY510430 4 Unidentified 1 Uncultured CFB-

group bacterium from ocean floor

96% AY510431

5 Unidentified 2 Uncultured bacterium clone s3 form marine sediment

97% AY510432

6 Unidentified 3 Uncultured eubacterium clone F10.2

96% AY510433

7 Unidentified 4 Marine bacterium SCRIPPS_739

100% AY510434

8 Sulfurospirillum sp. Sulfurospirillum sp. 96% AY510435 9 Cytophaga sp. Cytophaga sp. 100% AY510436 10 Unidentified 5 Uncultured gamma

proteobacterium clone CD3B12

97% AY510437

11 Arcobacter sp. Arcobacter sp. 98% AY510438 12 Clostridium related Uncultured bacterium

RSb40 100% AY510439

13 Oceanospirillum sp.

Oceanospirillum pusillum

99% AY510440

14 Unidentified 6 Uncultured bacterium clone ZB131

98% AY510441

15 Bdellovibrio sp. Bdellovibrio sp. 98% AY510442 16 Neptunomonas sp. Neptunomonas

naphthovorans 96% AY510443

17 Anaerofilum sp. Anaerofilum pentosovorans

99% AY510444

18 Unidentified 7 no match AY510445 19 Unidentified 8 no match AY510446 20 Unidentified 9 no match AY510447


45

sequences in the database of which no taxonomic information was available; no match

with available sequences (<<95%) was obtained for Unidentified species 7 to 9 (table

2).

Ā ⎯Si ⎯Si /SD ⎯Si % ∑ ⎯Si % Diplolaimelloides meyli (average similarity: 47.56) Clostridium sp.2 0.23 6.78 1.67 14.25 14.25 Arcobacter sp. 0.19 4.87 1.27 10.24 24.49 Unknown 3 0.11 3.53 2.45 7.42 31.92 Unidentified 1 0.14 3.44 1.28 7.23 39.15 Clostridium sp.1 0.1 2.8 1.22 5.88 45.02 Diplolaimelloides oschei (average similarity: 47.54) Clostridium sp.2 0.25 8.62 2.3 18.13 18.13 Unknown 8 0.19 6.41 2.18 13.49 31.61 Arcobacter sp. 0.19 5.44 1.32 11.45 43.06 Unidentified 1 0.11 3.3 2.1 6.95 50.01 Clostridium sp.1 0.12 3.06 1.23 6.43 56.44 Panagrolaimus paetzoldi (average similarity: 42.48) Clostridium sp.2 0.28 10.57 3.66 24.88 24.88 Unidentified 5 0.12 3.66 1.24 8.62 33.49 Clostridium sp.1 0.12 3.02 1.04 7.12 40.61 Anaerofilum sp. 0.1 2.41 0.92 5.67 46.28 Arcobacter sp. 0.14 2.29 0.51 5.38 51.66 Diplolaimella dievengatensis (average similarity: 52.74) Clostridium sp.2 0.28 10.52 3.29 19.95 19.95 Arcobacter sp. 0.27 9.23 2.45 17.5 37.45 Unidentified 7 0.13 4.35 1.92 8.05 45.7 Oceanospirillum sp. 0.13 3.54 1.2 6.71 52.41 Clostridium sp.1 0.12 3.27 1.44 6.2 58.6 control (average similarity: 48.83) Clostridium sp.2 0.28 9.53 2.57 19.52 19.52 Arcobacter sp. 0.21 6.72 2.36 13.77 33.29 Unknown 5 0.15 4.06 1.23 8.32 41.61 Oceanospirillum sp. 0.14 3.96 1.64 8.11 49.72 Clostridium sp.1 0.11 3.27 1.34 6.7 56.42

Table 3: Results of the Simper analysis giving the similarities within treatments. Five species

which contribute most to the similarity are listed. Āi: average abundance of the i th species

over all samples of the treatment; ⎯Si: contribution of the i th species to the total similartity;

⎯Si/SD: the value of the i th species as a discriminating species; ⎯Si%: percentage contribution

of the i th species to the total similarity; ∑ Si %: cumulative contribution to the total similarity.

Similarity percentages (SIMPER) analysis was used to look for those bacterial species

which contributed most to the average similarity within treatments and average

dissimilarity between treatments (Clarke and Gorley, 2001) (table 3 and 4; see table 2

for sequence information of the excised bands). Clostridium sp.1 and sp.2 and

Arcobacter sp. were prominent members of the bacterial community in each

treatment. Density differences in Arcobacter sp. generally explained a substantial part

of the dissimilarity between treatments. One unknown species occurred only in the


46

presence of D. oschei and explained most of the dissimilarity with other treatments.

For the control, the presence of Anaerofilum sp. accounted for much of the

dissimilarity with the nematode treatments.

Table 4: Results of the Simper analysis giving the dissimilarities between treatments. Three

species which contribute most to the dissimilarity are listed.⎯δi: contribution of the i th species

to the total dissimilartity;⎯δi/SD: the value of the i th species as a discriminating species; ⎯δ%:

percentage contribution of the i th species to the total dissimilarity; ∑⎯δi%: cumulative

contribution to the total dissimilarity.

Ā ⎯δi/SD ⎯δi% ∑⎯δi% D. meyli vs D. oschei (average dissimilarity 68.27) Unknown 8 3.93 2.72 5.75 5.75 Arcobacter sp. 2.74 1.38 4.02 9.77 Unknown 18 2.45 1.12 3.58 13.35 D. meyli vs P. paetzoldi (average dissimilarity: 75.46) Arcobacter sp. 3.54 1.44 4.69 4.69 Unidentified 1 2.82 1.55 3.74 8.43 Unidentified 5 2.59 1.75 3.44 11.87 D. meyli vs D. dievengatensis (average dissimilarity: 68.91) Arcobacter sp. 2.96 1.41 4.29 4.29 Unknown 1 2.37 1.11 3.43 7.72 Unknown 14 2.34 1.64 3.4 11.12 D. meyli vs control (average dissimilarity: 68.96) Anaerofilum sp.2 2.95 1.46 4.27 4.27 Unidentified 7 2.75 1.55 3.99 8.27 Arcobacter sp. 2.56 1.43 3.71 11.97 D. oschei vs P. paetzoldi (average dissimilarity: 70.59) Unknown 8 3.88 2.41 5.5 5.5 Arcobacter sp. 3.53 1.46 5 10.51 Unidentified 5 2.59 1.72 3.67 14.18 D. oschei vs D. dievengatensis (average dissimilarity: 61.80) Unknown 8 3.64 2.27 5.89 5.89 Arcobacter sp. 2.78 1.31 4.5 10.39 Unknown 18 2.4 1.07 3.88 14.28 D. oschei vs control (average dissimilarity; 67.39) Unknown 8 3.91 2.6 5.8 5.8 Anaerofilum sp.2 2.9 1.43 4.3 10.1 Arcobacter sp. 2.39 1.34 3.55 13.65 P. paetzoldi vs D. dievengatensis (average dissimilarity: 61.61) Arcobacter sp. 4.11 1.49 6.67 6.67 Unidentified 7 2.78 2.12 4.51 11.18 Unidentified 2 2.57 1 4.18 15.36 P. paetzoldi vs control (average dissimilarity: 67.37) Arcobacter sp. 3.56 1.67 5.28 5.28 Anaerofilum sp.2 2.67 1.33 3.96 9.24 Unidentified 5 2.64 1.74 3.92 13.17 D. dievengatensis vs control (average dissimilarity: 61.20) Anaerofilum sp.2 2.71 1.36 4.43 4.43 Unidentified 7 2.71 2.12 4.43 8.86 Arcobacter sp. 2.43 1.31 3.97 12.83


47

Figure 2: DGGE fingerprints with indications of the band classes. Numbers refer to table 2.

1

2

4

67

11

13

12

1617

10 days20 days30 days40 days50 days55 days65 days

20

1

2

4

67

11

13

12

1617


20

1

211

9

158

13

17


Diplolaimelloides meyli

Diplolaimelloides oschei

2

1

11

133

14

4 10 days20 days30 days40 days50 days55 days65 days

Panagrolaimus paetzoldi

11

1

2

4

510

6


Diplolaimella dievengatensis

11

1

2

4

610


Control

1918

18

18

1

211

9

158

13

17


1

211

9

158

13

17




2

1

11

133

14



11

1

2

4

510

6



11

1

2

4

610


Control

1918

18

18

1

2

4

67

11

13

12

1617


20

1

2

4

67

11

13

12

1617


20

1

2

4

67

11

13

12

1617


20

1

2

4

67

11

13

12

1617


20

1

211

9

158

13

17


1

211

9

158

13

17




2

1

11

133

14



11

1

2

4

510

6



11

1

2

4

610


Control

1918

18

18

1

211

9

158

13

17


1

211

9

158

13

17




2

1

11

133

14



11

1

2

4

510

6



11

1

2

4

610


Control

1918

18

18


48

In all treatments a clear shift of bands with time could be observed (fig. 2), which, for

the nematode treatments, was also reflected in the MDS plot (fig. 1). Simper analysis

(results not shown) of the effect of time within each treatment was performed to

determine those bands which were most important in determining the succession. The

dynamics of Arcobacter sp. were important in all treatments. In the presence of D.

meyli, the appearance of Sulfurospirillum sp. from day 50 onwards and the dynamics

of four unknown species determined the major differences in time. In the treatment

with D. oschei the decreasing intensity of Clostridium sp.2, the disappearance of

Unidentified sp. 6 and the appearance of two unknown species were of major

importance. In the presence of P. paetzoldi the succession was primarily driven by

dynamics in Clostridium sp.1, Unidentified sp.2 and three unknown species. The

succession in the D. dievengatensis treatment was mainly determined by the

appearance of Oceanospirillum sp. after 20 days and the variations of Unidentified

sp.5 and two unknown species. In the control treatment the decreasing intensity of

Clostridium sp.2 and the variation of Unidentified sp. 4 and two unknown species

through time were the most important dynamics in the succession.

Effects on bacterial diversity

It was assumed that each band on the DGGE gel represents one ‘Operational

Taxonomic Unit’ (OTU). However, some shortcomings of the DGGE technique

should be considered when interpreting the results. Species that make up less than 1%

of the total bacterial community will not be represented on a DGGE-gel (Muyzer et

al., 1993). The presence of multiple heterogeneous rRNA operons might result in

multiple bands from one bacterial species (Nübel et al., 1996) and co-migration of

sequences might also bias the banding profile (Vallaeys et al., 1997). Therefore, the

diversity index calculated from the DGGE banding patterns should not be interpreted

as an absolute measure, but rather as an indication of bacterial diversity (Eichner et

al., 1999). Nübel et al. (1999) showed that the number of bands and the Shannon-

Wiener diversity index can be used as an estimate of the bacterial diversity. Since all

samples in our experiment were treated in the same way, it is expected that bias

operated uniformly on all samples allowing proper comparison of the results (Fromin

et al., 2002; Schauer et al., 2000).


49

Bacterial diversity is represented by the Shannon-Wiener Index (fig. 3), which

integrates information on the number of bands and on the band-intensities. The

number of bands (data not shown) showed a similar pattern through time as the

Shannon-Wiener Index.

Figure 3: Shannon-Wiener Index (H’) during the cause of the experiment (average of 3 values

± 1 standard error).

Figure 3 shows that diversity reached a peak after 40 days in the controls and after 50

days in the presence of the monhysterids. Diversity was rather constant in the P.

paetzoldi treatment. A Tukey HSD pairwise post-hoc test only showed significant

differences within treatments in the presence of D. oschei, in which the index on day

10 was significantly lower than on day 50 (p<0.01) and day 55 (p<0.05), and in the D.

dievengatensis treatment where the index on day 50 was significantly higher than on

day 20 (p<0.01) and day 30 (p<0.05). No significant differences between treatments

could be found at any sampling occasion.

The generally lower bacterial diversity from 30 days onwards in the presence of P.

paetzoldi compared to the treatments with Monhysteridae could be ascribed to the

significantly higher densities of the former species. From that moment onwards, P.

paetzoldi reached densities up to 105 individuals per microcosm, while the

monhysterid nematode species (D. meyli, D. oschei, D. dievengatensis) attained

densities of a few thousand individuals (De Mesel et al., 2003). The high densities of

P. paetzoldi could have led to overgrazing of the bacterial community. This effect will

have been intensified by the unselective feeding habit of this species, enabling it to

eliminate a large proportion of the bacterial community. A high grazing pressure in

10 20 30 40 50 55 65Days

1.7

1.9

2.1

2.3

2.5

2.7

2.9

3.1

Shan

non-

Wie

ner I

ndex

(H')


10 20 30 40 50 55 65Days

1.7

1.9

2.1

2.3

2.5

2.7

2.9

3.1

Shan

non-

Wie

ner I

ndex

(H')




50

the presence of P. paetzoldi was confirmed by measurements of respiration and

proteolytic activity of the microbial community (fig. 4a) (De Mesel et al., 2003). A

reduction of diversity at the bacterial level in the presence of high protist densities has

been observed before (van Hannen et al., 1999). Most information on the relationship

between grazer densities and prey diversity, however, originates from studies using

herbivores and plants as a model (Valiela, 1995 and references therein; Hillebrand,

2003). These studies confirm low diversity under high grazing pressure, and suggest

that highest prey diversity develops under intermediate grazing pressure. Applying

this to our experiment, the highest bacterial diversity would be expected in the

presence of the monhysterid species. This however was not unambiguously confirmed

by our findings. Only on day 50 and day 55 was bacterial diversity higher in all

monhysterid treatments versus the control; with none of these differences being

statistically significant. We assume that the monhysterid densities were not high

enough to reach an ‘intermediate disturbance’ level since the numbers of monhysterid

nematodes obtained in our experiments were situated at the lower end of the range

found in field conditions. Furthermore Herman and Vranken (1988) showed that

monhysterids only affected a small subset of the standing stock of bacteria, even at

higher densities than those reached in this experiment. Additionally we found that in

these treatments diversity was linked to nutritional value (N-content) of the leaves

(fig. 4b), rather than to nematode densities. The Shannon-Wiener Index was lowest at

the beginning and at the end of the experiment, and reached a peak around 50 days.

Initially, few opportunistic bacterial species will have benefited from the availability

of high concentrations of labile compounds. As these easily degradable compounds

became depleted, opportunists became replaced by a more diverse community of

decomposers. Towards the end of the experiment, the detritus mainly consisted of

refractory substances, which can only be broken down by few specialised species.

Similar dynamics in bacterial diversity were seen in the control, however, at the end

of the experiment an unexpected increase of bacterial diversity occurred. This

coincides with a peak of nitrogen content in the organic matter (De Mesel et al.,

2003), which may be a result of the binding of nitrogen-rich compounds to the

cordgrass detritus (White and Howes, 1994).


51

A

B

Figure 4: A. Bacterial activity expressed as enzymatic activity (average of 3 samples ± 1

standard error) B. N-content of the cordgrass leaves, expressed as percentage of the initial

values (average of 3 replicates ± 1 standard error). The high loss the first 10 days is probably

due to leaching of nutrient rich compounds (De Mesel et al., 2003).

Conclusion

Our results suggest that the bacterial community composition in the microcosms was

shaped by a combination of bottom-up and top-down effects: the same bacterial

species appeared in different treatments, indicating that the types of bacteria occurring

were primarily determined by the substrate. However the composition and the relative

importance of different members of this ‘pool’ of bacteria was severely modified by

the grazing activities of the bacterivorous nematodes, even at relatively low densities.

A nematode effect on bacterial diversity was only evident under high grazing

pressure, as in the presence of P. paetzoldi.

0 0 10 20 30 40 50 55 65Days

5000

10000

15000

20000

25000

Enzy

mat

ic a

ctiv

ity (c

ount

s min

-1)

0 0 10 20 30 40 50 55 65Days

5000

10000

15000

20000

25000

Enzy

mat

ic a

ctiv

ity (c

ount

s min

-1)

0 10 20 30 40 50 55 65Days

0

2

4

6

8

10

12

mg

N

0 10 20 30 40 50 55 65Days

0

2

4

6

8

10

12

mg

N

0

2

4

6

8

10

12

mg

N

D.meyliD. oscheiP. paetzoldiD. dievengatensiscontrol

D.meyliD. oscheiP. paetzoldiD. dievengatensiscontrol


52

Acknowledgements

We thank Nele Vloemans for the help with the DGGE analyses. The first and second

author acknowledge grants from the Flemish Institute for the Promotion of Scientific-

Technological Research (I.W.T.). Tom Moens is a postdoctoral fellow with the

Flemish Fund for Scientific Research. Further financial support was obtained from

Ghent University in BOF- projects 1205398 and 011060002.

De Mesel, I, Derycke, S, Swings, J, Vincx, M, Moens, T

Chapter 4

Effect of nematode diversity on organic matter decomposition and the associated microbial community: a

microcosm study

Chapter 4: Diversity related impacts on decomposition

55

Abstract

Diversity related and species-specific impacts of bacterivorous nematodes on the

decomposition rate of cordgrass detritus (Spartina anglica) and on the associated

microbial community were investigated using laboratory microcosm experiments.

Four bacterivorous nematode species (Diplolaimelloides meyli, Diplolaimelloides

oschei, Diplolaimella dievengatensis and Panagrolaimus paetzoldi) were added either

separately or in combinations of two or three species to the microcosms.

In contrast with previous reports, no stimulation of the decomposition process was

observed in the presence of nematodes. In absence of nematodes, a more extensive

growth of fungi, important in the early stages of cordgrass decay, might have caused

the higher decomposition rates. Fungal growth seemed to be hampered in the

presence of nematodes, but whether this was a consequence of direct or indirect

interactions (through facilitation of bacterial growth) could not be distinguished.

Still, clear differences in process rates were found between nematode treatments.

Panagrolaimus paetzoldi, reaching considerably higher densities than the other

nematode species, suppressed bacterial activity and diversity, probably due to

overgrazing. This was however not translated into a slower decomposition process.

Within treatments exclusively containing monhysterid nematode species (D. meyli, D.

oschei and D. dievengatensis), differences in microbial activity and decomposition

rates were found, but again no link was observed between both. The lack of

completely separate measurements of bacterial and fungal activity, different C-use

efficiencies of bacterial communities or direct assimilation of detrital compounds by

(some) nematode species may have complicated the establishment of any such link.

Our data did not support any hypothesis predicting enhancement of process rates with

an (initial) increase in numbers of nematode species, nor for redundancy among the

studied species. We rather obtained support for an idiosyncratic diversity model as

differences in the effect of bacterivore nematode species and species combinations

could not be predicted at the start of the experiment. This could be explained by the

unpredictable interactions between nematode species. Within the treatments

containing only monhysterid species, densities were much lower in the combination

treatments compared to the monospecific treatments, indicating interspecific

inhibitory interactions, probably of a chemical nature.


57

Introduction

The role of biodiversity in the functioning of ecosystems has been discussed

extensively over the past few years. Many studies have concentrated on terrestrial

systems and emphasised on the link between plant diversity and productivity. These

results have led to a number of hypotheses, which generally describe a predictable

consequence of altering biodiversity (Schläpfer et al., 1999). The extremes range from

no effect of diversity as predicted by the (modified) null hypothesis, to an

improvement of the ecosystem functioning each time a species is added to the system,

referred to as the rivet hypothesis (Ehrlich and Ehrlich, 1981). Of particular interest is

the redundancy hypothesis (Walker, 1992), which states that it is of major importance

that all functions within an ecosystem are fulfilled; species within a functional group

are thought to be redundant. This does not exclude however, that ecosystem stability

could be better assured when each functional group comprises several equivalent

species (Walker, 1995) as their different responses to environmental factors can allow

persistence of the ecosystem function after disturbance. In contrast to the previous

hypotheses, the idiosyncratic model suggests that altering biodiversity generally

affects ecosystem functioning, but in an unpredictable manner (Lawton, 1994). This

results from a lack of proper understanding of the complex and often subtle

interactions between species within and across functional groups.

Fewer studies up to now have investigated the importance of diversity in other

ecosystem processes, such as decomposition. Decomposition is mainly a microbial

process and can be affected by higher trophic levels. Bacterial grazers, such as

protozoa and nematodes, have often been found to stimulate bacterial activity and

subsequently decomposition processes (e.g. Riemann and Schrage, 1978; Findlay and

Tenore, 1982; Alkemade 1992a, b, 1993).

Here we present the results of a microcosm study where four estuarine bacterivorous

nematode species from the same functional group have been used as a model to study

within-functional-group redundancy. The decomposition of cordgrass, often the

dominant vegetation in salt marshes (Howarth, 1993), was the ecosystem function of

interest. The effect of Diplolaimelloides meyli, Diplolaimelloides oschei,

Diplolaimella dievengatensis and Panagrolaimus paetzoldi, all commonly found on

decaying cordgrass in the Paulina Salt Marsh (Westerscheldt) (Moens, pers. comm.),


58

was investigated. Analysis of the monospecific treatments has already shown some

species-specific effects, mainly on the bacterial community structure (De Mesel et al.,

2004) and activity (De Mesel et al., 2003), and to a lesser extent also on the

decomposition rate (De Mesel et al., 2003). Here we focus on the effects of pooling

two or three nematode species. We tested (1) whether species-specific effects

observed in the monospecific treatments were maintained in treatments with species

combinations and (2) whether there was any diversity-dependent effect on (i) the

decomposition rate, (ii) microbial activity, and (iii) bacterial diversity.

Materials and Methods Experimental set-up

The experimental set-up has been described in detail in De Mesel et al. (2003).

Briefly, green cordgrass leaves (Spartina anglica) were collected in the Paulina Salt

Marsh (Westerscheldt, The Netherlands). In the laboratory, they were cleaned with

ethanol (70%) and artificial sea water (ASW, Dietrich and Kalle, 1957). The leaves

were cut in fragments, and 1.9 g ± 0.2 g (wet weight) was put in petri dishes (∅ 90

mm) on sterile sediment saturated with ASW with a salinity of 20. A microbial

inoculum consisting of bacteria from fresh and decomposing cordgrass leaves,

sediment and habitat water and from the nematode cultures was then added. This

microbial inoculum was first filtered over Whatman GF/C filters to remove flagellates

and other small eukaryotes. All petri dishes were inoculated with 2 ml of this

suspension. After a 24 h incubation, nematodes were added to the microcosms.

Four nematode species, commonly found on decaying cordgrass in the Paulina Salt

Marsh (Moens, pers. comm.) were selected for this experiment: Diplolaimelloides

meyli, Diplolaimelloides oschei, Diplolaimella dievengatensis and Panagrolaimus

paetzoldi. The former three belong to the Monhysteridae, the latter one to the

Panagrolaimidae. Together with the Rhabditidae, these nematode families are

typically associated with decomposing plant detritus in aquatic sediments (Warwick,

1987). All species were applied separately to the microcosms (De Mesel et al., 2003),

and in two- and three- species combinations. Each time, a total of 36 individuals were


59

inoculated in the microcosms (2 x 18 in treatments with 2 species; 3 x 12 in

treatments with 3 species). An overview of the treatments is given in table 1.

Microcosms were incubated in the dark at 20 ± 2°C.

Three replicate microcosms per treatment were destructively sampled after each of the

following incubation times: 10, 20, 30, 40, 50, 65 and 75 days. Microbial activity

(respiration rate and enzymatic activity) and decomposition parameters (weight,

carbon- and nitrogen- content) were measured at each sampling; nematode densities

and bacterial diversity were analysed on day 30, 50 and 65.

Table 1: Overview of the treatments

Decomposition rate

Percentage C and N in the cordgrass leaves were determined with a Carlo-Erba

element analyser type NA-1500 and converted to absolute amounts; C- and N-

contents of the detritus were expressed in mg. Similar patterns were obtained when

decomposition parameters are expressed as percentages remaining of the initial values

(cfr. De Mesel et al., 2003), however using absolute weight units instead of

percentages, allows proper transformation in order to fulfil the requirements for the

statistical analysis (see below).

Treatment code Nematode species

A Diplolaimelloides meyli

B Diplolaimelloides oschei

C Panagrolaimus paetzoldi

D Diplolaimella dievengatensis

E D. meyli, D. oschei

G D. meyli, D. dievengatensis

I D. oschei, D. dievengatensis

J P. paetzoldi, D. dievengatensis

L D. meyli, D. oschei, D. dievengatensis

N D. meyli, P. paetzoldi, D. dievengatensis

S (no nematodes)


60

Nematodes


al., 1985). After staining with Rose Bengal they were counted and, in the combination

treatments, identified under a dissecting microscope.

Microbial activity

The microbial community was rinsed off the cordgrass leaves with ASW in order to

measure their activity. Respiration rate of the microbial community was measured

with a Strathkelvin 928 6-channel Dissolved Oxygen System on one millilitre sub-

samples. The EnzCheck® Protease Assay Kit E6638 for green fluorescence

(Molecular Probes) was used to measure the proteolytic activity of the microbial

community. Both protocols are described in detail in De Mesel et al. (2003).

Bacterial diversity

An assessment of bacterial diversity was made based on DGGE (Denaturing Gradient

Gel Electrophoresis) banding patterns of the V3 region of the 16S rDNA genes. The

protocols used for DNA-extraction and -purification, amplification of the fragments

and the DGGE analysis have been described in detail in De Mesel et al. (2004). The

banding patterns were analysed with the software package Bionumerics 5.1 (Applied

Maths BVBA, Kortrijk, Belgium). An optical density profile through each lane

(corresponding to a single sample) identifies band positions, and calculates the

percent contribution of the intensity of each band to the total intensity of the lane. In

analogy with the results presented in De Mesel et al. (2004), bacterial diversity is

expressed as the Shannon Wiener Index (H’), based on the relative band intensities

(Magurran, 1988):

H’=∑=

n

iii pp

1ln

n represents the number of bands in the sample and pi the relative intensity of the ith

band.


61


To evaluate the importance of nematode diversity on the decomposition rate (N- and

C-content and weight), microbial activity (respiration rate), and bacterial diversity, a

nested analysis of variance (ANOVA), in which treatments were nested within

diversity levels (i.e. level 0, 1, 2, 3 referring to 0, 1, 2, 3 nematode species

respectively), was performed by means of the GLM module in the Statistica software

package. The time factor was also integrated into the analysis. The output of the

nested ANOVA provided information on the effect of diversity, on the effect of time

and indicated if differences were found between treatments nested within a diversity

level. Due to the unbalanced design of the experiment, the combined effect ‘diversity

x time’ could not be tested with the Statistica software. When ANOVA indicated

significant differences, the Tukey HSD test was used for pairwise post hoc

comparison. Weight and respiration rate were fourth root transformed, N-content

square root transformed and C-content log-transformed to meet the required

assumptions; no transformation was required for bacterial diversity. The data on

enzymatic activity did not fulfil the assumptions for nested ANOVA, even after

transformation. Due to the unbalanced design of the experiment, no non-parametric

alternative could be applied for a similar analysis.

Species- or combination-specific effects were tested by means of a two-way ANOVA

in which the effect treatment, time and the combined effect of treatment and time was

tested. When the overall test was significant, Tukey HSD post hoc test was used for

pairwise comparison. C-content, weight, and respiration rate were log transformed

and proteolytic activity square root transformed to meet the required assumption. A

transformation of N-content was not necessary.

While counting nematodes, it became obvious that monhysterid densities reached in

combination treatments were lower than in monospecific treatments (see results).

Whether these differences were significant was tested by means of two-way ANOVA,

including the effects time and treatment in the analysis. This test was run on total

nematode densities and species-specific densities in treatments containing exclusively

monhysterid species. Again, Tukey HSD post hoc test was used for pairwise

comparison. Densities were log-transformed to meet the required assumptions.


62

Results Nematodes

Nematode densities in all treatments with Panagrolaimus paetzoldi were 10 to 103

times higher than in other treatments (fig. 1). In the combination treatments this was

solely due to the high numbers of P. paetzoldi; the monhysterids made up only a

minor part of the total nematode numbers. In general, high variation between

replicates was present.

Nematode densities clearly differed between treatments containing only monhysterid

nematode species. Due to the considerably higher densities of P. paetzoldi, these

differences were not revealed in any statistical test when all treatments were included.

Therefore, a two-way ANOVA on species-specific densities in treatments containing

only monhysterid nematodes was carried out (fig. 2). It showed that for both D. oschei

and D. dievengatensis densities were significantly higher in the monospecific

treatments compared to the combination treatments (for all pairwise comparisons

p<0.01); within combination treatments, densities of both species were similar. For D.

1 nematode species

1

10

100

1000

10000

100000

30 50 65

ABCD

2 nematode species

1

10

100

1000

10000

100000

30 50 65

EGIJ

3 nematode species

1

10

100

1000

10000

100000

30 50 65days

LN

daysdays

1 nematode species

1

10

100

1000

10000

100000

30 50 65

ABCD

ABCD

ABCD

2 nematode species

1

10

100

1000

10000

100000

30 50 65

EGIJ

EGIJ

EGIJ

3 nematode species

1

10

100

1000

10000

100000

30 50 65days

LNLN

daysdays

Figure 1: Total nematode densities per microcosm in the monospecific treatments and the combination treatments (log scale) (average of 3 replicates ± 1 standard error)


63

Figure 2: Nematode densities per microcosm in treatments containing exclusively

monhysterid species (average of 3 replicates ± 1 standard error)

meyli, densities in the monospecific treatments were only significantly higher

compared to the 3-species combination (p<0.01). Also the total numbers of nematodes

were generally lower in the combination treatments, but differences with the

monospecific treatments were not always significant due to high standard errors (table

2).

Table 2: Results of the Tukey HSD post hoc test

following two-way ANOVA for total nematode densities

in treatments containing exclusively monhysterid species

(ns: not significant; *:p<0.05; **:p<0.01)

Microbial activity

Respiration rate

For respiration rate, the Tukey HSD post hoc test following nested ANOVA (table 3)

showed that the difference between the control treatment (diversity level = 0) and all

dversity levels was significant (p<0.05; p<0.01; p<0.05) (fig. 3A). Within diversity

A B D E G I L A * ns ns ns ns ns B * ns ** ** ns ns D ns ns * ns ns ns E ns ** * ns ns ns G ns ** ns ns ns ns I ns ns ns ns ns ns L ns ns ns ns ns ns

D. meyliD. oscheiD. dievengatensis

0

200

400

600

800

1000

1200

A B D E G I L30 days 50 days 65 days

A B D E G I L A B D E G I L

Num

ber o

f nem

atod

esD. meyliD. oscheiD. dievengatensis

0

200

400

600

800

1000

1200

A B D E G I L30 days 50 days 65 days

A B D E G I L A B D E G I L



0

200

400

600

800

1000

1200

0

200

400

600

800

1000

1200

A B D E G I LA B D E G I L30 days 50 days 65 days

A B D E G I LA B D E G I L A B D E G I LA B D E G I L

Num

ber o

f nem

atod

es


64

A

B

Figure 3: (A) Average respiration rate (± standard error) per diversity level at each sampling

occasion; (B) Average respiration rate (± standard error) per treatment over the whole

experiment

levels, for the monospecific treatments a significant difference between treatments A

and C (p<0.01), and for the 2-species combinations between treatment G and I

(p<0.05) were found. Only the significant differences between the control and the

two-species level were largely confirmed (significant differences for 3 combinations

out of 4) by Tukey HSD test (table 5) following two-way ANOVA (table 4).

DF MS F p-level

Diversity 3 0.016483127 5.863176 ***

Treat(Div)a 7 0.012415832 4.416408 ***

Time 6 0.894397227 318.144 ***

Error 214 0.002811297

Table 3: Results of the nested ANOVA for respiration rate (***:p<0.001)

(aTreatment nested within diversity)

Days

0

0.2

0.4

0.6

0.8

1

10 20 30 40 50 65 75

Res

pira

tion

rate

(µm

ol O

2h-1

)

0 species1 species2 species3 species

Days

0

0.2

0.4

0.6

0.8

1

10 20 30 40 50 65 7510 20 30 40 50 65 75

Res

pira

tion

rate

(µm

ol O

2h-1

)


0.1

0.2

0.3

0.4

S A B C D E G I J L N

Res

pira

tion

rate

(µm

ol O

2h-1

)

0.1

0.2

0.3

0.4

S A B C D E G I J L NS A B C D E G I J L N

Res

pira

tion

rate

(µm

ol O

2h-1

)


65

Figure 3B gives an indication on the overall respiration rate in each treatment over the

whole experiment. Details on temporal changes per treatment are represented in the

appendix.

DF Effect DF error F p –level

Time 6 154 428.8421 ***

Treatment 10 154 7.632465 ***

Time x Treatment 60 154 2.828222 ***

Table 4: Results of the two-way ANOVA for respiration rate (***:p<0.001)

Table 5: Results of the Tukey HSD post hoc test following the two-way

ANOVA for respiration rate (ns: not significant; *:p<0.05; **:p<0.01;

***:p<0.001)

Proteolytic activity

Within the nematode treatments, the overall trend is a slight increase in enzymatic

activity with increasing nematode diversity, while the activity in the control is situated

in between the other treatments (fig. 4A). Nested ANOVA, to test for the significance

Table 6: Results of the two-way ANOVA for proteolytic activity

(***:p<0.001)

A B C D S E G I J L N A ns *** ns ns ns 1 *** *** *** ns B ns ns ns ns ns ns ns ns ns ns C *** ns ns *** ns ** ns ns ns * D ns ns ns ns ns ns * ns ns ns S ns ns *** ns ** ns *** *** *** ns E ns ns ns ns ** ns ns ns ns ns G ns ns *** ns ns ns *** ns ns ns I *** ns ns * *** ns *** ns ns ** J ** ns ns ns *** ns ns ns ns ns L ** ns ns ns *** ns ns ns ns ns N ns ns * ns ns ns ns ** ns ns


Time 5 132 22.63869 ***

Treatment 10 132 12.27109 ***



66

of the differences at the diversity level, could not be run as the assumptions were not

fulfilled.

Within each diversity level, enzymatic activity is reduced when P. paetzoldi is present

(treatments C, J and N)(fig. 4B). This is largely confirmed by the post hoc test (table

7) following two-way ANOVA (table 6).

In the appendix, a detailed representation of the proteolytic activity in each treatment

throughout the experiment is given.

A

B

Figure 4: (A) Average proteolytic activity (± standard error) per diversity level at each

sampling occasion; (B) Average proteolytic activity (± standard error) per treatment over the

whole experiment

0

5000

10000

15000

20000

25000

30000

Days

Enzy

mat

ic a

ctiv

ity (c

ount

s min

-1)

10 20 30 40 50 65 750

5000

10000

15000

20000

25000

30000

Days

Enzy

mat

ic a

ctiv

ity (c

ount

s min

-1)

10 20 30 40 50 65 75

5000

70009000

11000

1300015000

1700019000

S A B C D E G I J L NEnzy

mat

ic a

ctiv

ity (c

ount

s min

-1)

5000

70009000

11000

1300015000

1700019000

5000

70009000

11000

1300015000

1700019000

S A B C D E G I J L NS A B C D E G I J L NEnzy

mat

ic a

ctiv

ity (c

ount

s min

-1)




67

A B C D S E G I J L N A ns *** ns ns ns ns ** ns * ns B ns *** ns ns ns ns ns * ns ns C *** *** *** *** *** *** *** ns *** *** D ns ns *** ns ns ns ns ** ns ns S ns ns *** ns ns ns ns *** ns ns E 1 ns *** ns ns ns * ns * ns G ns ns *** ns ns ns ns ns ns ns I ** ns *** ns ns * ns *** ns ** J ns * ns ** *** ns ns *** *** ns L * ns *** ns ns ns ns ns *** * N ns ns *** ns ns ns ns ** ns *

Table 7: Results of the Tukey HSD post hoc test following the two-

way ANOVA for proteolytic activity (ns: not significant; *:p<0.05;

**:p<0.01; ***:p<0.001)

Bacterial diversity

A

B

Figure 5: (A) Average bacterial diversity (± standard error) per nematode diversity level at

each sampling occasion; (B) Average bacterial diversity (± standard error) in each treatment

over the whole experiment

1,5

1,7

1,9

2,1

2,3

2,5

2,7

2,9

30 50 65days

Shan

non-

Wie

ner I

ndex

(H’)

1,5

1,7

1,9

2,1

2,3

2,5

2,7

2,9

1,5

1,7

1,9

2,1

2,3

2,5

2,7

2,9

30 50 65days

Shan

non-

Wie

ner I

ndex

(H’)

1,5

1,7

1,9

2,12,3

2,5

2,7

2,9

S A B C D E I J L N

Shan

non-

Wie

ner I

ndex

(H’)

1,5

1,7

1,9

2,12,3

2,5

2,7

2,9

1,5

1,7

1,9

2,12,3

2,5

2,7

2,9

S A B C D E I J L NS A B C D E I J L N

Shan

non-

Wie

ner I

ndex

(H’)




68

A generally lower bacterial diversity in the combination treatments than in the

monospecific treatments and, to a lesser extent, the control was found (fig. 5A).

Tukey HSD post-hoc test following nested ANOVA indicated only the differences

between diversity level 1 and the 2- and 3- species combinations as statistically

significant (respectively p<0.01 and p<0.001). Although the overall nested ANOVA

showed a significant difference between treatments within diversity levels (table 8),

the Tukey HSD post hoc test did not reveal any pairwise differences. Within diversity

levels, however, the treatments containing P. paetzoldi (treatments C, J and N)

generally have a lower bacterial diversity (fig. 5B). For the 3-species combination,

and to a lesser extent the 2-species combination, the lower bacterial diversity in

presence of P. paetzoldi compared to other treatments is confirmed by the post hoc

comparison following two-way ANOVA (tables 9 and 10). For details on bacterial

diversity within each treatment and an overview of the DGGE-banding patterns, see

appendix. Information on bacterial diversity in treatment G (combination treatment

with D. meyli and D. dievengatensis) is not available.

DF MS F p

Diversity 3 0.716504474 10.31198 ***

Treat(Div)a 6 0.32048179 4.612395 ***

Time 2 0.217112495 3.124697 ns

Error 70 0.069482731

Table 8: Results of the nested ANOVA for bacterial diversity (ns: not

significant; ***:p<0.001) (a Treatment nested within diversity)

DF Effect DF error F p-level

Time 2 60 4.537769 *

Treatment 9 60 10.63394 ***

Time x Treatment 18 60 2.598584 **

Table 9: Results of the two-way ANOVA for bacterial diversity (*:p<0.05;

**:p<0.01; ***:p<0.001)


69

Table 10: Results of the Tukey HSD post hoc test following the two-

way ANOVA for bacterial diversity (ns: not significant; *:p<0.05;

**:p<0.01; ***:p<0.001)

Decomposition rate

C-content

A

B

Figure 6: (A) Average C-content (mg) (± standard error) in each treatment over the whole

experiment; (B) Average C-content (mg) (± standard error) per diversity level

A B C D S E I J L N A ns ns ns ns * ns *** ns *** B ns ns ns ** ** ns *** ns *** C ns ns ns ns ns ns ns ns ** D ns ns ns ** ns ns ** ns *** S ns ** ns ** ns ns ** ns *** E * ** ns ns ns ns ns ns ns I ns ns ns ns ns ns ns ns ** J *** *** ns ** ** ns ns ns ns L ns ns ns ns ns ns ns ns ** N *** *** * *** *** ns ** ns **

100

150

200

250

300

350

10 20 30 40 50 65 75100

150

200

250

300

350

100

150

200

250

300

350

10 20 30 40 50 65 7510 20 30 40 50 65 75

150160170180190200210220230


mg

C

150160170180190200210220230

150160170180190200210220230


mg

C



days

mg

C


70

Figure 6A shows the average C-content in each treatment over the whole experiment.

A slower decomposition process will be represented by a higher average C-content.

An overview of the temporal C-dynamics in each treatment is given in the appendix.

A generally slower C-loss in the treatments with the 2-species combinations

(treatments E, G, I, J) was seen (fig. 6A and 6B). According to the post hoc test

following nested ANOVA (overall test in table 11), the difference between the 2-

species level and the other diversity levels were significant (all p< 0.001); also a low

significant difference between diversity level 0 (controls) and the 1-species diversity

level was found (p<0.05). Only the differences between the 2-species combinations

and the control were clearly confirmed by the two-way ANOVA (table 12 and 13).

The results of weight are very similar to those for C-content (see also De Mesel et al.,

2003) and are therefore not presented here.

Table 11: Results of the nested ANOVA for C-content (*:p<0.05;

***:p<0.001) (a Treatment nested within diversity)


Time 6 154 204.0253 ***

Treatment 10 154 8.360312 ***

Time x treatment 60 154 1.501799 *

Table 12: Results of the two-way ANOVA for C-content (*:p<0.05;

***:p<0.001)

DF MS F p

Diversity 3 0.054855677 19.30672 ***

Treat(Div)a 7 0.006239223 2.195925 *

Time 6 0.508193444 178.8611 ***

Error 214 0.002841274


71

Table 13: Results of the Tukey HSD post hoc test following two-way

ANOVA for C-content (ns: not significant; *:p<0.05; ***:p<0.001)

N-content

As for C-content, higher average N-content in figure 7A agrees with slower

decomposition rates. Details on N-content per treatment are represented in the

appendix.

Differences for N-content between the control treatment and nematode treatments

were found to be significant by both nested ANOVA (significance level with diversity

DF MS F p-level

Diversity 3 13.91021609 5.929644 ***

Treat(Div)a 7 21.03011271 8.964712 ***

Time 6 96.11691385 40.9727 ***

Error 214 2.345877222

Table 14: Results of the nested ANOVA for N-content (***:p<0.001)

(aTreatment nested within diversity)


Time 6 154 77.37736 ***

Treatment 10 154 15.21042 ***


Table 15: Results of the two-way ANOVA for N-content (***:p<0.001)

A B C D S E G I J L N A ns ns ns ns ns * * ns ns ns B ns ns ns ns ns ns ns ns ns ns C ns ns ns ns ns ns ns ns ns ns D ns ns ns *** ns ns ns ns ns *** S ns ns ns *** *** *** *** *** * ns E ns ns ns ns *** ns ns ns ns *** G * ns ns ns *** ns ns ns ns *** I * ns ns ns *** ns ns ns ns *** J ns ns ns ns *** ns ns ns ns *** L ns ns ns ns * ns ns ns ns * N ns ns ns *** ns *** *** *** *** *


72

A

B

Figure 7: (A) Average N-content (mg) (± standard error) in each treatment over the whole

experiment; (B) Average N-content (mg) (± standard error) per diversity level

Table 16: Results of the Tukey HSD post hoc test following two-way ANOVA

for N-content (ns: not significant; *:p<0.05; **:p<0.01; ***:p<0.001)

A B C D S E G I J L N A ns ns *** ** ns ns ns ns * ns B ns ns *** ns ns ns *** *** *** ns C ns ns *** ns ns ns ns ns *** ns D *** *** *** *** *** *** * * ns *** S ** ns ns *** ns ns *** *** *** ns E ns ns ns *** ns ns ns ns ** ns G ns ns ns *** ns ns ** ** *** ns I ns *** ns * *** ns ** ns ns ns J ns *** ns * *** ns ** ns ns ns L * *** *** ns *** ** *** ns ns *** N ns ns ns *** ns ns ns ns ns ***

0

2

4

6

8

10

12

10 20 30 40 50 65 75

mg

N

0

2

4

6

8

10

12

0

2

4

6

8

10

12

10 20 30 40 50 65 7510 20 30 40 50 65 75

mg

N

10

20

30

40

50


mg

N

10

20

30

40

50

10

20

30

40

50


mg

N




73

levels respectively p<0.05; p<0.05 and p<0.001) and two-way ANOVA (table 15 and

16) (fig. 7B). The significant treatment effect pointed out by nested ANOVA (table

14) was due to significant differences between monospecific treatments D and,

respectively, A (p<0.01), B (p<0.001) and C (p<0.01), and between 3-species

treatments L and N (p<0.05). These significant differences were confirmed by two-

way ANOVA (table 16).

Discussion Development of nematode populations

All four nematode species used in this experiment belong to the deposit feeders

(Wieser, 1953) and are assumed to feed mainly on bacteria (Moens and Vincx, 1997).

Despite their similar functional position, some ecological differences between the

Panagrolaimidae and Monhysteridae have been described. Panagrolaimidae are

considered extreme colonizers and enrichment opportunist, quickly colonizing

organically enriched sites with high microbial activity and disappearing as microbial

activity decreases (Bongers and Bongers, 1998). Monhysteridae also occur in sites

with high resource availability, however they can persist under food-poor conditions;

as such they belong to the general opportunists (Bongers et al., 1995). Their

generation time is slightly longer in comparison with Panagrolaimidae. Due to these

differences, P. paetzoldi was expected to reach peak densities at the start of the

experiment, while the monhysterids were thought to have a slower increase in

numbers, but to persist throughout the experiment. This was however not confirmed

by our observations: densities of P. paetzoldi were much higher during the course of

the experiment, both in the monospecific treatments and in the combination

treatments. Another remarkable observation was that monhysterid densities were

considerably lower in the combination treatments compared to the monospecific

treatments. One reason could have been that the densities of individual species in the

initial inoculum of the combination treatments were lower than in the monospecific

treatments: 36 individuals in the monospecific treatments, compared to 18 and 12 in,

respectively, the 2- and 3- species combinations. This might have caused some delay

in the population development of each species; however this discrepancy should

vanish over subsequent generations, as was the case for P. paetzoldi. Moreover, an


74

additional experiment, with larger nematode inocula, confirmed inhibitory

interactions between two of the monhysterids used here, D. meyli and D.

dievengatensis (pers. observation, unpublished). In a similar experimental set-up as

used in this study, a number of petri dishes were inoculated with either D. meyli or D.

dievengatensis, or stayed without nematodes (controls). After an incubation of 7 days,

half of the petri dishes with D. meyli or D. dievengatensis were inoculated with the

other species, the other half were further incubated without manipulation.

Simultaneously, half of the controls received D. meyli, the other half D.

dievengatensis. Two weeks later, nematode densities were counted and compared. A

lower number of either species was consistently found in the combination treatments

when compared to single-species treatments, with the differences mostly being

significant.

Similar observations have been reported by Ilieva-Makulec (2001), who studied the

life strategy of two terrestrial bacteria-feeding nematodes, Acrobeloides nanus and

Dolichorhabditis dolichura. They explained this mutual inhibition by food shortage in

the combination treatments. Nematode densities in the monospecific treatments show,

however, that, even when all species require the same bacterial diet, higher densities

can be sustained with the resources available. Additionally, a food-selection

experiment (Moens et al., 1999) and analysis of the bacterial community in the

monospecific treatments (De Mesel et al., 2004) indicated different feeding

preferences between species, which should allow them to share the food sources. We

also found similar aerobic activity and even higher proteolytic activity in the

combination treatments compared with the monospecific treatments, which does not

at all indicate a suppression of the bacterial community.

Information on inhibition within benthic fauna is in general very scarce, while

facilitative interactions between species have been reported more often (e.g. David,

1996; Levin et al., 2001; Cardinale et al., 2002). Chandler and Fleeger (1987) were

the first to study interactions within meiofauna. They conducted experiments with

copepods in estuarine sediments and found both inhibition and facilitation between

species. They suggest that the observed inhibitory interactions were due to mucus

trails secreted by one of the copepod species, which not only altered sediment

structure, but might also have evoked a chemical repulsion. Many free living marine

nematodes also secrete mucus tracks (Riemann and Schrage, 1978, Riemann and

Helmke, 2002). It seems unlikely that any structural modification of the sediment


75

could have caused the mutual suppression in population development because (i)

species-specific densities would probably be too low to affect the sediment structure

of the total microcosm and (ii) one would expect one species (the ‘modifier’) to

dominate the system and suppress the other species. This was obviously not the case

in our study: a clear mutual inhibition was observed. The idea that some kind of

chemical interference between species occurred can not be refuted. Chemical cues

could emanate from the mucus tracks, or from other glandular secretions. Nematodes

are susceptible to chemical signals (Huettel, 1986) since they are important in the

reproduction (Bone, 1982) or for tracing food sources (Moens et al., 1999). Up till

now, patchy food distribution (Trotter and Webster, 1984; Moens et al., 1999) was

assumed to be the main cause of the heterogeneous nematode distribution in marine

sediments, but chemical interaction might enhance this phenomenon. Accumulation of

chemicals in the closed system in this experiment could have prevented growth of

nematodes in the whole microcosm.

Effect of nematodes on microbial activity and bacterial diversity

The activity of the microbial community was measured by the respiration rate and the

proteolytic activity. The highest respiration rate was observed in the control. This was

statistically confirmed by analysis at both diversity and treatment level. Respiration

was assumed to mainly represent fungal activity (Padgett et al., 1985; De Mesel et al.,

2003). However, some interference of bacterial activity in the respiration

measurements will have occurred, especially at the start of the experiment. Still, the

high respiration rates in the control coincide with observations of stronger growth of

fungi. In terrestrial systems an inhibition of fungal growth in the presence of

nematodes has been described, either as a consequence of direct effects or through

interactions with bacteria (Kaufman, 1983; Grewal, 1991; Grewal and Richardson,

1991; Grewal and Hand, 1992). Similar interactions in marine environments have not

yet been studied; however we believe they might have occurred in our microcosms.

Nematodes might have facilitated bacterial communities, for instance through

secretion of mucus trails (Riemann and Schrage, 1978) or any other metabolites.

Antagonistic interactions between bacteria and fungi during decomposition of aquatic

plant litter have been described (Wohl and McArthur, 2001); with the reverse effect of


76

bacteria on fungi being more pronounced (Mille-Lindblom and Tranvik, 2003).

Facilitation of the bacterial community in the nematode treatments might thus have

antagonized the development of fungi. Fungi are important in the first stages of

cordgrass decay and their suppression could have slowed down the decomposition

process (see below).

Proteolytic activity would mainly represent anaerobic metabolism, since it increases

while respiration rate decreases. Aquatic fungi are generally aerobic (Rheinheimer,

1992), suggesting that proteolytic activity provides information on anaerobic bacterial

activity; however some interference of fungi can not be excluded. Proteolytic activity

was considerably lower in the presence of P. paetzoldi than in the other treatments

within the same diversity level. Their densities were much higher than those of the

monhysterids, so this reduction can be ascribed to overgrazing of the bacterial

community. The same applies to bacterial diversity: the high grazing pressure in the

presence of P. paetzoldi reduced diversity within the bacterial community (see also

De Mesel et al., 2004). While, in general, a positive trend in the relationship between

number of nematode species and enzymatic activity was seen, the link with bacterial

diversity was negative. Lack of correlation between bacterial diversity and bacterial

activity has been reported before (Troussellier et al., 2002).

Impacts on the decomposition rates

Despite the low nematode densities in the absence of P. paetzoldi, and the dominance

of the latter species when present, we did find a diversity-related impact on the C-loss

of the cordgrass detritus. As explained before (De Mesel et al., 2003), in the

monospecific treatments a slower decomposition rate was observed compared to the

control. An even more severe suppression of C-loss occurred when two nematode

species were pooled, but at the 3-species diversity level C-loss was again comparable

to the monospecific treatments, or even higher as for the treatment with the

combination P. paetzoldi, D. dievengatensis and D. meyli.

Nitrogen content of the cordgrass leaves was generally lower in the control treatment.

This could be due to the presence of fungi in the controls (see above), which are

important during the early stages of decomposition. On the other hand, nematodes

consume nitrogen in excess when feeding on bacteria (Anderson et al., 1983; Ferris et


77

al., 1997). After excretion, this nitrogen can either be used by the bacteria, or can bind

to the detritus (White and Howes, 1994; Valiela, 1995), masking any effect of

microbial activity on the removal of nitrogen. Significant differences in nitrogen

content of the cordgrass leaves between nematode treatments were found, however no

link with diversity nor with specific species or species combinations can be made. For

instance, when D. meyli and D. dievengatensis were applied separately to the

microcosms, N-content was significantly higher than in the control treatment, but

when these species were pooled, the difference with the control was less pronounced;

adding the third monhysterid (D. oschei) again yielded significant difference with the

control.

No link between bacterial diversity and decomposition rates was found. According to

O’Donnell et al. (2001), the central problem in microbial ecology is to understand the

link between genetic diversity and community structure, and between community

structure and function. Studies on functional implications of altering microbial

diversity in soil systems is still scarce, however, a degree of redundancy is expected

(Griffiths et al., 2000; Nannipieri et al., 2003).

Within treatments a positive correlation between respiration rate and both C-content

and, to a lesser extent, N-content was found; links with proteolytic activity were less

clear. This indicates that fungi would largely depend on the cordgrass substrate, and

thus would be mainly responsible for the decomposition of the detritus. This is not

unusual, since fungi can penetrate solid substrates and digest them from within

(Newell, 1996); bacteria on the other hand might have a better ability to utilize

dissolved organic compounds (Wohl and McArthur, 2001; Mille-Lindblom and

Tranvik, 2003). These organic compounds leached into the sediment at the start of the

experiment (Valiela, 1995). This does however not exclude that bacteria might have

impacted the decomposition rate to some extent, but their effect could be obscured by

the fungi.

Differences in respiration rates between treatments did not necessarily result in

(predictable) differences in elemental losses; the same goes for proteolytic activity.

Riemann and Helmke (2002) suggested that nematodes and bacteria growing on their

mucus tracks, ‘share’ enzymes to break down organic matter. This way, nematodes

could directly assimilate organic compounds from the detritus. On the other hand,

differences in C-use efficiencies between bacterial communities might also have

affected our results (Degens, 1998 and references therein).


78

Conclusions

The main aim of this paper was to evaluate the importance of nematode diversity or

nematode species in affecting the microbial activity and the decomposition rate.

Our data did not support any hypothesis predicting enhancement of process rates with

an (initial) increase in species-numbers (e.g. rivet hypothesis, optimum hypothesis,

modified null hypothesis). A key mechanism in enhancing ecosystem function would

be interspecific facilitation (Loreau and Hector, 2001; Mulder et al., 2001; Fridley,

2001; Cardinale et al., 2002). Facilitation was not at all observed in our experiments;

instead we found clear evidence of inhibitory interaction between the monhysterid

species. Redundancy among the monhysterids seems unlikely as it was not

consistently confirmed for all parameters. N-loss for instance, was very different

between monhysterid treatments. Differences between the monhysterids and P.

paetzoldi were obvious. The extremely high densities of P. paetzoldi resulted in a

reduction in bacterial activity and diversity; however this was not translated into a

slower decomposition rate. Also within the treatments exclusively containing

monhysterid nematode species no clear link between microbial activity and

decomposition rate was found. The lack of completely separate measurements of

bacterial and fungal activity might have complicated the establishment of any such

link. Additionally, different C-use efficiencies of the bacterial communities, and the

possibility that (some) nematode species could directly use organic detrital

compounds, should be considered.

Our results rather support the idiosyncratic diversity model, which describes an

unpredictable effect of changing species composition or species diversity on

ecosystem processes. However, in concordance with Mikola and Setälä (1998), who

also found an idiosyncratic response of microbivore nematode diversity in a terrestrial

soil system, we should note that our experiments only contained a small number of

species. It is possible that idiosyncratic effects are mainly visible in low diversity

systems. Still, differences between bacterivore nematode species and species

combinations could not be predicted at the start of the experiment, indicating that no a

priori redundancy within the bacterivores can be expected. Additionally, several other

experimental studies, using other benthic systems, confirm this idiosyncratic theory

(e.g. Emmerson et al., 2001; Downing and Leibold, 2002).


79

Acknowledgements

Annick Van Kenhove is acknowledged for the chemical analyses and Prof. Dr.

Jacques Vanfleteren for the use of the Victor Multilabel counter. We thank Prof. Dr.

Ann Vanreusel, Dr. Peter Herman and Dr. Marleen De Troch for the valuable

comments on the manuscript. The first and second author acknowledge grants from

the Flemish Institute for the Promotion of Scientific-Technological Research (I.W.T.).

The last author is a postdoctoral fellow with the Flemish Fund for Scientific Research.

Further financial support was obtained from Ghent University in BOF- projects

1205398 and 011060002.

Submitted for publication as:

De Mesel, I, Vanaverbeke, J, Vincx, M, Moens, T (submitted) Development of a

nematode community on decomposing organic matter (Spartina anglica) under

contrasting conditions.

Chapter 5

Development of a nematode community on decomposing organic matter (Spartina anglica) under contrasting

conditions

Chapter 5: Colonization of cordgrass detritus under contrasting conditions

83

Abstract

Cordgrass detritus has often been described as a microhabitat sustaining high densities

of nematodes. We performed litterbag experiments in which the development of the

nematode community on decaying cordgrass detritus was studied under stable

conditions in the laboratory and in the field, where incubations were subject to

environmental fluctuations, mainly associated with the tidal currents.

Nematodes showed a clear preference for the detritus under controlled conditions in

the lab, and the nematode assemblage on the cordgrass differed considerably from the

surrounding sediments. By contrast, in the field nematodes were rather distributed at

random, probably through passive transport with the tides, showing that active habitat

colonization was of minor importance in the hydrodynamically impacted system.

We assumed that the contrasting incubation conditions in the lab and the field would

affect the structural and functional aspects of the nematode community. In the lab,

incubations were performed in the dark, preventing growth of microalgae, an

important food source for mainly epistrate feeders and deposit feeders. Epistrate

feeders were however equally abundant in the field and in the lab, indicating that

other food sources, probably bacteria, can be consumed. Deposit feeders, dominating

field samples, were mainly replaced by microvores in the lab; still some species were

able to adapt to a principally bacterial diet. Nematode diversity was not affected by

resource availability, nor by the daily hydrodynamical disturbance.


85

Introduction

Marine sediments are generally characterized by high densities of meiofauna. In both

subtidal and intertidal habitats, their distribution can be very patchy (Findlay 1981,

1982). At large scales (m-km) this patchiness is due to gradients in physical factors,

for example salinity, tidal exposure, sediment composition (Findlay, 1981; Coull,

1988). The small scale heterogeneous distribution would mainly be caused by biotic

interactions such as competition, either inter- or intraspecific (Heip, 1975, 1976;

Service and Bell, 1987), facilitation (Heip and Engels, 1977; Fleeger and Gee, 1986;

Chandler and Fleeger, 1987), the presence of biogenic structures (Bell et al., 1978;

Sun et al., 1993; Wetzel et al., 1995; Fenchel, 1996) and the heterogeneous

distribution of their food sources (Buzas, 1969; Boucher and Chamroux, 1976). In a

laboratory experiment, Lee et al. (1977) offered a number of micro-algae to a

meiofaunal inoculum in a so called ‘cafeteria’-design. He reported a selective

attraction of different taxa towards each of the food sources. A similar set-up was

used to seek for the response of a limited number of nematode species towards a

selection of bacterial and diatomaceous food items (Trotter and Webster, 1984;

Moens et al., 1999). These experiments suggested that nematodes can perceive and

move towards their preferred food source. Differences in food availability could also

cause spatial differences and temporal dynamics in the trophic composition of the

nematode community (Danovaro and Gambi, 2002)

In the dynamic environment of intertidal salt marshes, however, small-scale spatial

distribution of meiofauna may not only be influenced by biotic interactions but also

by abiotic factors. Meiofauna generally have limited active dispersal capacities

(Fegley, 1987 and references therein) and can be passively dispersed by tidal currents

(Bell and Sherman, 1980; Sherman and Coull, 1982; Fleeger et al., 1984; Bertelsen,

1998; Powers 1998; Commito and Tita, 2002). Wave action may thus facilitate patch

colonization (Ullberg and Ólafsson, 2003), but, at the same time hamper the

establishment of ‘stable’ populations.

The present study aimed to assess the relative importance of biotic and abiotic forces

in structuring nematode communities colonizing patches of freshly deposited

cordgrass detritus. Cordgrass (Spartina spp) is often the dominant vegetation of the

lower parts of tidal marshes in temperate areas and its detritus can support high


86

nematode densities (Reice and Stiven, 1983; Alkemade et al., 1993). These detrital

microhabitats are often characterized by a high relative abundance of characteristic

taxa, mainly Monhysteridae (Bouwman et al., 1984; Warwick 1987; Alkemade et al.,

1993). They may stimulate the decomposition rate of the organic matter (Abrams and

Mitchell, 1980; Anderson et al., 1981; Findlay and Tenore, 1982; Alkemade et al.,

1992 a, b, 1993), although this has been questioned recently (De Mesel et al., 2003).

Cordgrass detritus in litterbags was placed on salt marsh sediment both under

controlled laboratory conditions and in a field site. Samples were taken monthly

during a three-month period. We tested whether (i) detritus availability could impact

the distribution of the nematode community in an undisturbed laboratory

environment, (ii) this distribution could be affected by natural disturbances in the

field, and (iii) the contrasting environmental conditions in the field and in the lab

would alter the nematode community structure.

Materials and Methods

Experimental set up

Litterbag incubations were performed simultaneously in a salt marsh (Paulina Salt

Marsh, Westerschelde, The Netherlands) and under controlled conditions in the

laboratory. The litterbags (10 cm * 12 cm) were made of gauze with a mesh size of

125µm, allowing free movement of micro-and meiofauna. One set of bags was filled

with cordgrass detritus, the other with plastic strips which served as controls.

Green cordgrass leaves were collected in the salt marsh (1st October 2001). They were

thoroughly rinsed with fresh water and dried to the air to remove the attached fauna.

The leaves were cut in fragments of approximately 10 cm and 4.1 ± 0.1g was put in

each litterbag. The control litterbags contained plastic strips with a similar surface as

the cordgrass fragments.

In the salt marsh 4 experimental fields with a surface of approximately 4 m2 each

were selected at the edges of Spartina-patches in the middle of the Spartina

dominated zone, which was flooded twice a day. In each field, 30 litterbags, 15 of

which containing detritus and 15 controls were incubated. The litterbags were

separated approximately 30 cm from each other, preventing mutual influence.


87

In the laboratory, plastic dishes (13 cm * 17 cm * 15 cm) were filled with sediment

originating from the Paulina Salt Marsh and incubated in the dark at 12°C for one

week allowing the system to stabilize. Afterwards, half of the dishes received a

detritus litterbag, the others control litterbags. A suspension of sediment from the salt

marsh in artificial sea water (Dietrich and Kalle, 1957) was added when the litterbags

were placed on the dishes and after every sampling in order to provide a fresh fauna

inoculum. Each time a few hundred nematodes were applied to all dishes. Our results

showed however that these inocula did not considerably affect the composition of the

nematode communities. The sediment was water saturated throughout the experiment

and a constant salinity was maintained.

Litterbags were incubated at 12°C in the dark.

Sampling

Samples were taken monthly over a 3-month period (November 2001, December

2001 and January 2002). At each sampling occasion one detritus litterbag and one

control were removed at random from each experimental field in the salt marsh,

resulting in four replicates per sampling. Simultaneously, the upper cm of the

sediment in each field was sampled with Perspex cores (∅ 3.6 cm).

In the laboratory, 4 dishes with cordgrass and 4 with controls were sampled every

month, on the same days as the field samplings. From every dish, sediment

surrounding the litterbags was collected as described above.

Nematodes were rinsed of the cordgrass leaves and plastic strips with tap water. The

rinse water and the sediment were preserved on a 4% buffered formaldehyde solution.

Cordgrass fragments were dried, weighed, an preserved at -20°C for C- and N-

analysis, which was performed with a Carlo-Erba elemental analyzer type NA-1500.

Fungal biomass in the cordgrass leaves was assessed by ergosterol measurements

(Gessner and Newell, 1997)

Nematode community


al., 1985). Nematodes from the sediment, the detritus and the controls were stained


88

with Rose Bengal and counted under a dissecting microscope. At each sampling

occasion 200 nematodes were picked randomly from two replicates from each

treatment (sediment, detritus, controls) and mounted on Cobb slides for identification

to species level. The contribution of each species to the assemblage was expressed as

a relative abundance in order to facilitate comparison of the assemblages from the

detritus, the controls (both ind./litterbag) and the sediment (ind/10cm²). For the

functional analysis of the community, nematodes were grouped in 6 feeding guilds:

microvores, deposit-feeders, ciliate feeders, epigrowth feeders, facultative predators

and predators (Moens and Vincx, 1997). This system is based on the classification of

Wieser (1953) but expanded with information obtained from observations.

Diversity was expressed with the Shannon-Wiener Index (H’) (Shannon and Weaver,

1949):

H’= ∑=

n

iii pp

1ln

n represents the number of species in the sample and pi the relative abundance of the

ith species.


Univariate two- or three-way analysis of variance (ANOVA) was used to test for

differences of nematode densities, feeding types and diversity between sites (field vs

laboratory), sampling dates, treatments (detritus vs controls vs sediment) and

combinations of these factors. Fourth root transformation on nematode densities was

performed to meet the required assumptions. When ANOVA indicated significant

differences the Tukey HSD test was used for pairwise post hoc comparison. These

tests were done with the Statistica software.

Nematode community structure was analysed by means of Multidimensional Scaling

(MDS) using the PRIMER software package. The MDS was based on the Bray-Curtis

similarity index, calculated from the relative densities. Analysis of Similarities

(ANOSIM) was applied to detect differences between sites and differences between

treatments within sites. Similarity percentage analysis (SIMPER) was used to weigh

the contribution of each species to similarity/dissimilarity between sites and

treatments.


89

Results

Nematode densities

ANOVA showed significant differences in nematode densities between the sediment

samples (p<0.001) (table 1). The Tukey HSD test showed that in the laboratory,

nematode densities in the sediment from dishes with detritus were very similar to

densities from dishes with controls, and both densities were significantly lower than

field densities (both p<0.001) (fig. 1A).

Figure 1: Nematode densities (A) in the sediment samples (average of 4 replicates ± standard

error) and (B) on the detritus and the controls (average of 4 replicates ± standard error)

According to the ANOVA test (table 2), nematode densities in litterbags (controls and

detritus) were significantly higher in the laboratory compared to the field (p<0.01) and

on the detritus compared to the controls (p<0.001). No significant effect of ‘time’ was

observed. Highly significant differences were found for the combined effect ‘site’ x

‘treatment’ (p<0.001). The Tukey HSD post hoc test showed that in the laboratory,

0

2000

4000

6000

8000

10000

12000

November December January

sediment field

sediment lab(detritus)sediment lab(control)

02000400060008000

10000120001400016000


detritus fielddetritus labcontrol fieldcontrol lab

A

B0

2000

4000

6000

8000

10000

12000


sediment field


sediment field


02000400060008000

10000120001400016000

November December January0

2000400060008000

10000120001400016000




A

B

ind/

10cm

2in

d/lit

terb

ag


90

nematode densities were significantly higher on detritus compared to the controls

(p<0.001) and compared to both the detritus and the controls in the field (both

p<0.001) (fig. 1B). In the field, densities on the controls and the detritus were similar.

Table 1: Results of the two-way ANOVA for nematode densities in the

sediment (ns: not significant, ***:p<0.001) a type: sediment core from (1) the field, (2) control dishes in the lab and (3)

dishes with detritus in the lab

Table 2: Results of the three-way ANOVA for nematode densities on the detritus

and the controls (ns: not significant, **:p <0.01, ***:p<0.001) a treatment: (1) detritus and (2) control

Effect DF effect DF error F p-level

Time 2 27 12.55228 ***

Typea 2 27 40.21584 ***

Time x Type 4 27 1.96637 ns

Effect DF effect DF error F p-level

Site 1 36 10.17099 **

Time 2 36 0.12346 ns

Treatmenta 1 36 18.45973 ***

Site x Time 2 36 1.84033 ns

Site x Treatment 1 36 16.6994 ***

Time x Treatment 2 36 0.20101 ns

Time x Treatment x Site 2 36 0.32266 ns


91

Nematode community Nematode assemblage

In the MDS plot based on the nematode species assemblages, a separation of the field

and lab samples was clear (fig. 2). ANOSIM indicated that the difference between

sites was significant (p<0.001). By means of SIMPER analysis (results not shown)

Daptonema biggi and Tripyloides marinus were identified as characteristic species for

the field samples. Leptolaimus limicolus, Anoplostoma viviparum and Axonolaimus

typicus were important members of the communities in the laboratory.

Metachromadora remanei and Ptycholaimellus ponticus were abundant at both sites.

Differences between sites were mainly determined by D. biggi and L. limicolus.

Figure 2: MDS ordination plot based on the nematode species assemblages

ANOSIM on the samples of the laboratory showed a highly significant difference

between nematode community composition on the detritus and in the sediment

(p<0.01); the difference between controls and the sediment was low significant

(p<0.05)(table 3). SIMPER analysis showed that L. limicolus, who dominated the

communities on the detritus, explained most of the variation between the detritus and

the sediment community. Ptycholaimellus ponticus was the most characteristic

species on the controls and M. remanei and A. typicus in the cores. Anoplostoma

viviparum was a prominent member of all communities in the lab (fig. 3A).

sediment field

sediment lab

detritus fieldcontrol field

detritus labcontrol lab

nov

nov

novnov

novnov

decdec

dec

dec

dec

dec

dec

jan jan

janjan

jan jan

jan

nov

nov

nov

nov

dec

janjan

decnov

dec

jandec nov

jandec

jan

Stress: 0.15detritus fieldcontrol field


nov

nov

novnov

novnov

decdec

dec

dec

dec

dec

dec

jan jan

janjan

jan jan

jan

nov

nov

nov

nov

dec

janjan

decnov

dec

jandec nov

jandec

jan

Stress: 0.15

sediment field

sediment lab

detritus fieldcontrol field


nov

nov

novnov

novnov

decdec

dec

dec

dec

dec

dec

jan jan

janjan

jan jan

jan

nov

nov

nov

nov

dec

janjan

decnov

dec

jandec nov

jandec

jan

Stress: 0.15detritus fieldcontrol field


nov

nov

novnov

novnov

decdec

dec

dec

dec

dec

dec

jan jan

janjan

jan jan

jan

nov

nov

nov

nov

dec

janjan

decnov

dec

jandec nov

jandec

jan

Stress: 0.15


92

A

B

Figure 3: Composition of the nematode community on detritus, controls and in the sediment.

The 5 dominant species in (A) the laboratory and (B) the field samples are represented.

For the field samples ANOSIM showed no significant differences between treatments.

All treatments were characterized by the presence of D. biggi; other species, such as

P. ponticus, M. remanei, Desmolaimus zeelandicus and Tripyloides marinus were also

characteristic for all treatments (fig 3B).

Table 3: Results of the ANOSIM routine to analyze differences in

nematode assemblages between detritus, controls and sediment in the

laboratory (ns: not significant, *:p<0.05, **:p<0.01)

R-statistic p-level

detritus vs control 0.107 ns

detritus vs sediment 0.381 **

control vs sediment 0.228 *

0%

20%

40%

60%

80%

100%

detritus control sediment

Daptonema biggiPtycholaimellus ponticus

Metachromadora remaneiTripyloides marinusDesmolaimus zeelandicusothers

0%

20%

40%

60%

80%

100%

detritus control sediment0%

20%

40%

60%

80%

100%


Daptonema biggiPtycholaimellus ponticus

Metachromadora remaneiTripyloides marinusDesmolaimus zeelandicusothers

0%

20%

40%

60%

80%

100%


Leptolaimus limicolusPtycholaimellus ponticusMetachromadora remaneiAnoplostoma viviparumAxonolaimus typicusothers

0%

20%

40%

60%

80%

100%


Leptolaimus limicolusPtycholaimellus ponticusMetachromadora remaneiAnoplostoma viviparumAxonolaimus typicusothers


93

Functional community structure

Table 4: Results of the three-way ANOVA for relative abundances of microvores (MV),

deposit-feeders (DF), epistrate feeders (EF), ciliate feeders (CF), predators (P) and facultative

predators (FP) (ns: not significant, *:p<0.05, **:p<0.01, ***:p<0.001)

An overview of the results of the three-way ANOVA is represented in table 4. The

proportion of microvores, ciliate feeders, predators and facultative predators was

significantly higher in the lab compared to the field (p<0.001, p<0.05; p<0.05;

p<0.001 respectively), while the opposite was found for deposit feeders (p<0.001)

(fig. 4). Similar percentages of epigrowth feeders were found at both sites. No

significant differences were found between treatments within one site.

MV DF CF EF P FP

DF effect DF error

1 18

1 18

1 18

1 18

1 18

1 18

F 64,62328 32,48896 4,88594 0,15354 5,002934 34,50508 Site

p-level *** *** * ns * *** DF effect DF error

2 18

2 18

2 18

2 18

2 18

2 18

F 3,84566 6,39942 1,297919 11,21461 0,432588 0,33567 Tim

e

p-level * ** ns *** ns ns DF effect DF error

2 18

2 18

2 18

2 18

2 18

2 18

F 2,19227 1,80725 0,533884 2,09577 0,325256 1,14379

Trea

tmen

ta

p-level ns ns ns ns ns ns

DF effect DF error

2 18

2 18

2 18

2 18

2 18

2 18

F 4,19017 2,11368 0,134558 0,56834 0,618706 0,27625 Site

x

Tim

e

p-level * ns ns ns ns ns DF effect DF error

2 18

2 18

2 18

2 18

2 18

2 18

F 1,08413 6,11429 0,118275 1,93536 0,846145 0,77165 Site

x

Trea

tmen

t

p-level ns ** ns ns ns ns

DF effect DF error

4 18

4 18

4 18

4 18

4 18

4 18

F 1,14245 1,63149 2,527191 0,43745 0,081061 1,76394 Tim

e x

Trea

tmen

t

p-level ns ns ns ns ns ns DF effect DF error

4 18

4 18

4 18

4 18

4 18

4 18

F 0,85902 1,0576 1,243217 1,3911 0,610571 1,51127 Site

x

Tim

e x

Tr

eatm

ent

p-level ns ns ns ns ns ns


94

Figure 4: Trophic community composition in the lab and in the field. MV: microvores, DF:

deposit feeders, EF: epistrate feeders, CF: ciliate feeders, P: predators, FP: facultative

predators

Diversity

In the laboratory, a total of 49 species were counted; in the field 53. Diversity was

rather similar in the field and in the lab, as shown by the Shannon-Wiener Index (fig.

5). ANOVA indicated significant differences for the combined effect ‘time’ x

‘treatment’ (table 5), but the Tukey HSD post hoc test did not reveal any significant

pairwise differences.

Table 5: Results of the three-way ANOVA for the Shannon-Wiener Indices (ns: not

significant, *: p < 0.05). a Treatment: (1) detritus, (2) controls and (3) sediment

DF effect DF error F p-level

Site 1 18 2.58911 ns

Time 2 18 2.582196 ns

Treatmenta 2 18 0.23249 ns

Site x Time 2 18 2.927248 ns

Site x Treatment 2 18 0.380592 ns

Time x Treatment 4 18 3.611206 *

Site x Time x Treatments 4 18 0.43276 ns

0%10%20%30%40%50%60%70%80%90%

100%

field lab

FPPEFCFDFMV

0%10%20%30%40%50%60%70%80%90%

100%

0%10%20%30%40%50%60%70%80%90%

100%

field lab

FPPEFCFDFMV

FPPEFCFDFMV


95

Figure 5: Shannon-Wiener Indices in (A) field and (B) laboratory samples (average of 2

replicates ± 1 standard error)

Discussion

Deposits of decomposing salt marsh vegetation have been described as

(micro)habitats supporting very high densities of nematodes (Montagna and Ruber,

1980; Reice and Stiven, 1983; Buth and de Wolf, 1985; Hemminga and Buth, 1991;

Alkemade et al., 1993, 1994). We observed a similar pattern in the laboratory, where

the total nematode community showed a clear preference for the detritus. Spartina

litter incubated in the field, however, supported only moderate nematode densities,

comparable to the controls.

The available resources did not seem to affect the total nematode community in the

field. Here, detritus was solely used as free space. Similar patterns have been

described for macrophyte detritus in intertidal systems (Reice and Stiven, 1983; Buth

and de Wolf, 1985). Nematodes however can show species-specific responses, which

are masked when considering them as one taxon (Heip et al., 1985; Warwick et al.,

1988; Alkemade et al., 1993; Ólafsson, 1992).

A

B1

1.21.41.61.8

22.22.42.62.8

3


detrituscontrolsediment

11.21.41.61.8

22.22.42.62.8

3



A

B1

1.21.41.61.8

22.22.42.62.8

3

11.21.41.61.8

22.22.42.62.8

3



11.21.41.61.8

22.22.42.62.8

3

11.21.41.61.8

22.22.42.62.8

3




96

Spartina patches in salt marshes are often characterised by a number of (dominant)

species which are much less prominently present in the underlying and surrounding

sediments (Bouwman et al., 1984; Warwick, 1987). Again, only under controlled

laboratory conditions, a similar picture was obtained: the nematode assemblage on the

decaying cordgrass detritus differed considerably from that in the surrounding

sediments. A dominance of Leptolaimus limicolus on the detritus and of

Metachromadora remanei and Axonolaimus typicus in the sediment, and the

appearance of Anoplostoma viviparum explained most of this dissimilarity.

Remarkably, the nematode assemblage on the controls showed a low significant

difference with the sediments, mainly due to the high densities of Ptycholaimellus

ponticus. This species is very common in muddy intertidal sediments, especially in

the upper few mm, below which it is almost absent (Coull, 1988; Steyaert et al.,

2003). Therefore we assume that the dominance of P. ponticus on the controls is due

to a contamination by the upper sediment layer attaching to the plastic strips. In the

field, no preference for any microhabitat was observed; all treatments were dominated

by Daptonema biggi and also other species reaching high densities, such as P.

ponticus and Tripyloides marinus, were characteristic for all treatments. These

observations indicate the overriding importance of the dynamic environmental

conditions in shaping the nematode communities in the field. The physical

(hydrodynamical) disturbance regime in the field site, which is flooded twice a day,

appears to hamper the (species-specific) attraction towards Spartina, and the

community build-up in the resource-rich habitat. A strong dispersal capacity of tidal

currents on the nematode community has been described before (Sherman and Coull,

1980; Armonies, 1988). Alkemade et al. (1993) similarly reported that the majority of

nematodes colonizing Spartina litter in another small Westerschelde marsh belonged

to typical sediment species.

In the same study however correlations between species-specific densities and the

decomposition rate of the detritus were reported. Nematodes are assumed to affect

decomposition processes through interaction with the bacterial community developing

on the detritus (Riemann and Schrage, 1978; Abrams and Mitchell, 1980; Alkemade

et al., 1992a,b, 1993; De Mesel et al., 2003). In our study, a positive correlation

between the decomposition rate of Spartina and densities of Axonolaimus typicus was

found only in the field samples. The relevance of this correlation can be questioned as

A. typicus makes up only a very small proportion of the total nematode community


97

and a similar correlation was not found in the laboratory, where A. typicus reached

much higher densities. In the lab, decomposition rate was rather linked to fungal

biomass, although the correlation was not significant (fig. 6A and B). Fungi are

important in the early stages of decomposition of cordgrass (Newell et al., 1989;

Valiela, 1995) and their biomass was significantly higher in the lab compared to the

field. Since there is no obvious trophic link between the free-living nematodes and

fungi, no nematode effect on the decomposition rate was observed.

Figure 6: Decomposition rate expressed as g C loss per day (bars) (average of 4 replicates ±

standard error) and fungal biomass in g (line) (average of 2 replicates ± standard error) in (A)

the laboratory and (B) the field.

The high hydrodynamic impacts at the field site studied may also explain why we did

not find the typical nematode species, mainly Monhysteridae, that are often reported

in high abundances on decomposing Spartina detritus (Bouwman et al., 1984;

Warwick, 1987; Alkemade et al., 1994). Several species of the genera

Diplolaimelloides, Diplolaimella, (Geo)monhystera and Thalassomonhystera are

common on phytodetritus within dense Spartina vegetation a little higher up in the

November December JanuaryDec

ompo

sitio

n ra

te (g

C d

ay-1

)

00.0050.01

0.0150.02

0.0250.03

0.0350.04

0.045

00.020.040.060.080.10.120.140.16 Fungal biom

ass (g)

A

B

Dec

ompo

sitio

n ra

te (g

C d

ay-1

)

00.0050.01

0.0150.02

0.0250.03

0.0350.04

0.045

00.020.040.060.080.10.120.140.16

Fungal biomass (g)



ompo

sitio

n ra

te (g

C d

ay-1

)

00.0050.01

0.0150.02

0.0250.03

0.0350.04

0.045

00.0050.01

0.0150.02

0.0250.03

0.0350.04

0.045

00.020.040.060.080.10.120.140.16

00.020.040.060.080.10.120.140.16 Fungal biom

ass (g)

A

B

Dec

ompo

sitio

n ra

te (g

C d

ay-1

)

00.0050.01

0.0150.02

0.0250.03

0.0350.04

0.045

00.020.040.060.080.10.120.140.16

Fungal biomass (g)


ompo

sitio

n ra

te (g

C d

ay-1

)

00.0050.01

0.0150.02

0.0250.03

0.0350.04

0.045

00.0050.01

0.0150.02

0.0250.03

0.0350.04

0.045

00.020.040.060.080.10.120.140.16

00.020.040.060.080.10.120.140.16

Fungal biomass (g)



98

marsh (Moens unpubl.). These habitats are presumably more sheltered and tidal

currents are either reduced, reducing resuspension of the nematodes in the water

column, and thus their passive distribution (Armonies, 1988), or they are not flooded

daily. The latter situation can be compared with the laboratory experiment in which

active migration of the nematodes can determine their micro-distribution. The lack of

these species in the inoculum explains why they were not found in the lab samples.

The stable laboratory conditions resulted in the development of different nematode

communities, with other species dominating the assemblages, namely Leptolaimus

limicolus in the lab and Daptonema biggi in the field. In addition to these differences

in species composition, we had also expected different community attributes in the

field versus the laboratory. First of all, the resource characteristics in field and lab

settings show important differences. While we added the same amount of organic

matter in lab and field, lab incubations were performed in the dark and hence did not

allow a significant primary production by microphytobenthos. Diatoms and other

microalgae are an important food source for a variety of nematodes, mainly belonging

to the deposit and epigrowth feeders (Wieser, 1953; Moens and Vincx, 1997). Due to

the absence of microphytobenthos, food items in the laboratory were probably

dominated by bacteria, especially associated with the decaying organic matter.

Bacteria serve as food sources for microvores, deposit feeders and, to a lesser extent,

facultative predators (Moens and Vincx, 1997). We expected to find a reflection of the

food availability in the trophic structure of the nematode communities; however these

assumptions were only partly confirmed. High densities of deposit feeders were found

in the field, which is common in muddy sediments. Next to the high densities of

Daptonema biggi, also Daptonema oxycerca, Desmolaimus zeelandicus and

Praeacanthonchus punctatus occurred in considerable densities in all treatments. In

the laboratory, however, the proportion of deposit feeders was significantly lower than

in the field. Only Axonolaimus typicus and, to a lesser extent, D. zeelandicus were

present in all treatments. This indicates some flexibility within the trophic group of

the deposit-feeders: some species did survive on a mainly bacterial diet, while others,

such as Daptonema species and P. punctatus, clearly required microphytobenthos as a

food source. In the lab, deposit feeders seemed to be partly replaced by microvores,

with Leptolaimus limicolus reaching high densities in all microhabitats, especially on

the detritus, and also Halalaimus minusculus being prominently present in the

sediment and on controls. These species have very small buccal cavities through


99

which particles no larger than bacteria can pass. The life strategy of microvores is thus

adapted to bacteria as major food source, and therefore they might be competitively

stronger than deposit feeders when prey is dominated by bacteria. Remarkably, the

expected lack of epigrowth feeders in the lab due to the absence of their presumed

food source was not observed; they made up about 25% of the total nematode

community, similar to the field. High numbers of mainly M. remanei and P. ponticus

were found, but also Microlaimus globiceps and Microlaimus nanus were quite

abundant. This suggests a trophic plasticity of the epigrowth feeders. Bacteria are the

most plausible alternative food source for these species although this has been

doubted in the past (Tietjen and Lee, 1973; Jennings and Deutch, 1975; Deutsch,

1978). A recent study, however, showed a positive correlation between epigrowth

feeders and dividing bacterial cells, rather than with chloropigment concentrations

(Danovaro and Gambi, 2002). Due to their well developed buccal armature, M.

remanei and P. ponticus, have been assumed to predate on other nematode species

(Wieser, 1953), however this seems unlikely since no such observations have yet been

done. Overall densities of predators and facultative predators were rather low both in

the field and the lab, but still they made up a significantly larger proportion of the

total community in the lab in comparison with the field samples. The stable conditions

in the lab will have allowed (facultative) predators, which generally have long

reproduction times, to establish a population.

Community complexity and diversity among sites and treatments were remarkably

similar. One could have expected the resource characteristics to affect community

diversity, resulting in differences between litter treatments, controls and sediment.

The daily currents can be regarded as disturbances (Garstecki and Wickham, 2003)

generating and maintaining higher diversity compared to the stable lab conditions

(Connell, 1978). However, when disturbance is too strong, this can result in a negative

effect on diversity (Connell, 1978). Since the distribution of the nematodes already

illustrated that the tidal currents might have severely affected the establishment of

their community, we assume that disturbance was indeed too intense and prevented

the development of a more diverse nematode community.

To conclude, we found that nematodes show a clear (species-specific) habitat

preference when only biotic factors play part. In the unstable, intertidal system

however, this active dispersion is of minor importance and nematodes are distributed

at random by the tidal currents. These findings are in agreement with Hogue (1982),


100

who noticed that biotic factors tend to aggregate meiofauna, while abiotic factors tend

to disperse them.

Both structural and functional characteristics of the nematode community structure

were clearly affected by the thoroughly different environmental conditions in the lab

and in the field, while diversity was similar at both sites. The trophic structure of the

nematode communities reflected to a certain extent food availability in the two sites,

but our data also showed that species-specific flexibility in food uptake within trophic

guilds is present, especially for the epigrowth feeders and to a lesser extent for the

deposit feeders.

Acknowledgements

We greatly thank Gunter Van Ryckegem for the ergosterol analyses. The first author

acknowledges a grant from the Flemish Institute for the Promotion of Scientific-

Technological Research (IWT). The last author is a postdoctoral fellow with the

Flemish Fund for Scientific Research. Further financial support was obtained from

Ghent University in BOF projects 1205398 and 011060002.

Conclusions

Conclusions

103

Conclusions Species manipulation experiment

Microcosm studies in ecological research

Both experiments in this thesis are (partly) based on microcosm incubations under

controlled laboratory conditions. Microcosm experiments have been widely used in

ecological studies during the last few decades (Ives et al., 1996) to unravel

phenomena observed in the field or to generate hypotheses about the behavior and

functioning of real ecosystems (Drake et al., 1996). They are a generally accepted

research tool in soil ecology (Moore et al., 1996). While in other ecological branches

the limitations of scale can be a problem, this is generally much less the case in soil

systems. Organisms of interest are mostly small so that their actual habitat size can be

approached. Some scientists have criticized the (over)simplification of natural

systems in laboratory set-ups (Carpenter, 1996; Kampichler et al., 2001); however this

can also be considered as the strength of this approach (Lawton, 1996). Natural

ecosystems are very complex, and consist of a large number of interacting processes.

Understanding individual processes seems essential to get insight into the whole

ecosystem. Integrating these observations into field studies could then lead to a better

insight into functions and processes (Carpenter, 1996; Moore et al., 1996).

Development of nematode populations

Some findings in microcosm experiments could simply never have been detected in

field studies. This applies for instance for the interference interactions between the

monhysterid nematode species. Remarkably, despite the competition, they manage

to establish and maintain a rather stable community, though with low (overall)

densities. Based on traditional competition models one would expect a competitively

inferior species to be excluded from the community under the stable conditions of this

laboratory experiment.

Food partitioning is one of the processes which may allow coexistence of

bacterivorous nematode species in field sediments (Grewal and Wright, 1992; Moens

et al., 1999) that could have occurred in our systems as well, as shown by the

differences in the bacterial community structure in the monospecific treatments. Still,

Conclusions

104

this does not provide an explanation for the lower (overall) densities. Based on our

results, we believe that chemical interference should be considered. In our closed

laboratory set-up, accumulation of chemicals could have hampered general growth; in

field sediments this might protect monospecific aggregates from intrusion of other

species. Aggregation is thought to be a way of minimizing interspecific competition,

thus of enhancing the coexistence of highly diverse communities with functionally

similar species (Ettema, 1998). It would be very interesting to further examine these

kinds of interactions in order to formulate a satisfactory explanation on their nature.

Another peculiar observation in our microcosms was the extremely high densities of

Panagrolaimus paetzoldi. As they are extreme colonizers, they generally occur in

sites with freshly deposited organic matter where bacterial activity is high. We

observed however that P. paetzoldi can maintain high densities even when bacterial

activity was suppressed. We assume they might be able to directly assimilate

detrital compounds, for instance through enzyme sharing with the bacterial

community (Riemann and Helmke, 2002). In any case, this species was better adapted

to utilize the available resources than the monhysterid nematodes.

Effect of bacterivorous nematodes on decomposition rates and microbial activity

The idea for studying the effect of bacterivorous nematode diversity on decomposition

processes originated from studies describing a stimulation of decay rates in the

presence of bacterivores (e.g. Abrams and Mitchell, 1980; Findlay and Tenore, 1982;

Anderson et al., 1983; Alkemade et al., 1992 a, b). The outcome of our experiment

was thus rather surprising. No stimulation of the decomposition rate, nor of the

microbial community was observed in the presence of nematodes. We assume that

the impact of bioturbation, generally considered one of the main pathways of

stimulation (Alkemade et al., 1992b) was minor in our microcosms since fragments of

cordgrass leaves were incubated on a sediment surface, as they generally decompose

in situ, instead of mixed with the sediment.

The question then still remains why the decomposition process was slowed down

in the presence of nematodes. Based on our results, we propose a hypothesis which

could explain the different rates in decomposition in the control compared to the

nematode treatments (see also schematic overview in figure 1). We have indications

(observations and data on the respiration rate) that the microbial community in the

controls was dominated by fungi. Fungi are important in the early stages of the

Conclusions

105

decomposition process of cordgrass. In the presence of the nematodes, fungal growth

seemed to be suppressed. This could have had two different causes: either nematodes

directly antagonized the fungi (cfr Kaufman, 1983), or they interacted by favoring the

Figure 1: Schematic representation of the hypothesis on the decomposition process in

absence and presence of nematodes. The dotted lines illustrate N-regeneration by the

nematodes and its subsequent adsorption to the detritus. (DOM: Dissolved Organic Matter,

POM: Particulate Organic Matter)

bacterial community, for instance by facilitating bacterial growth on mucus tracks

(Riemann and Schrage, 1978), which could in turn interfere with fungal growth

(Mille-Lindblom and Tranvik, 2003; Wohl and McArthur, 2001). Both steps in the

latter pathway have already been established in aquatic systems. Bacteria, thus partly

replacing fungi in the nematode treatments, can more easily use dissolved organic

compounds, rather than particulate substrates (Mille-Lindblom and Tranvik, 2003).

Leaching of organic compounds from detritus is generally the first step in a

decomposition process (Valiela, 1995) and was probably enhanced in our experiment

by the treatment of the cordgrass leaves prior to the incubation. Leaching was mainly

confirmed by the high N- loss in all treatments during the first 10 days of the

in ABSENCE of NEMATODES in PRESENCE of NEMATODES

Fungi

Fungi

Bacteria

Bacteria

Nematodes

---

-

+-

C N

DOM

POM

C N

DOM

N

in ABSENCE of NEMATODES in PRESENCE of NEMATODES

Fungi

Fungi

Bacteria

Bacteria

Nematodes

-----

-

+-

C N

DOM

POM

C N

DOM

N

Conclusions

106

experiment. Bacteria will thus have first degraded this dissolved organic matter,

before utilizing the more inaccessible compounds from the detritus, resulting in the

slower decomposition rate observed in the presence of nematodes in our experiments.

By contrast, fungi have the ability to penetrate solid organic substrates and digest

them from within (Newell, 1996). This could have caused the fast initial

decomposition rates in the controls. Within the nematode treatments, different

characteristics of the bacterial communities (e.g. composition, efficiency,….),

interactions with the nematodes (e.g. enzyme sharing,….) and interference of fungi

and bacteria in the microbial activity measurements probably obscured clear

relationships between microbial activity and decomposition rates in the nematode

treatments.

Impact of bacterivorous nematodes on the bacterial community composition

Bacterial communities in our microcosms were shaped by a combination of bottom-up

and top-down impacts. The types of bacteria were primarily determined by the

substrate as the same bacterial species occurred in different treatments. However,

clear species specific top-down impacts of the bacterivorous nematodes on the

pool of bacteria were evident, even at relatively low nematode densities as in the

presence of the monhysterids. We assume that each species grazed on a particular part

of the bacterial community, as has already been suggested by some controlled

laboratory experiments (Grewal and Wright, 1992; Moens et al., 1999). The effect of

nematode diversity on the composition of the bacterial community was not analyzed,

but we did check the impact on bacterial diversity. First, a lower bacterial diversity

in the presence of two- and three- species combinations was noted. This might be

the summed effect of selective grazing pressure of the individual nematode species,

suppressing their preferred food sources below the detection limit of DGGE.

Secondly, within each nematode diversity level, bacterial diversity was

considerably lower when Panagrolaimus paetzoldi was present. This is most likely

due to the high densities of this species, resulting in overgrazing of the bacterial

community, rather than to characteristics of the species itself.

Does species diversity matter in bacterivorous nematode communities?

One of the most discussed theories on the relationship between species diversity and

the functioning of ecosystems predicts redundancy among functionally similar species

Conclusions

107

(Walker, 1992; Schläpfer et al., 1999). Based on this hypothesis, we expected the

bacterivorous nematode species used in our experiment to perform the same function

and no or a minor impact of diversity on the decomposition process and the associated

microbial community. This was however not confirmed, as our results suggest an

idiosyncratic response of the system upon changing diversity or species

composition. It can of course be argued that diversity levels in our experiment were

low compared to natural systems. Idiosyncratic effects of changing community

composition may become apparent mainly in experiments with low numbers of

species. It would thus be interesting to integrate higher diversity levels, or even

natural nematode communities, into a next phase of this study. Still, the effect of

species or species combinations on the decomposition process in our microcosms

could not have been predicted at the start of the experiment. In other words,

redundancy of nematodes within the same functional group, or positive effects when

adding species to a system can not be assumed a priori. Mikola and Setälä (1998)

came to similar conclusions when testing the effect of bacterivorous and fungivorous

nematode species on decomposition rates in a terrestrial system.

As for the effect on bacterial diversity, the species-specific effect of Panagrolaimus

paetzoldi on the microbial activity will have been due to their high densities, which

resulted in overgrazing, rather than being caused by intrinsic properties of the species.

Suggestions for future research

As explained in the introductory paragraph, it is important to link microcosm

experiments to field studies in order to validate their significance and to understand

the functioning of the ecosystem. This step could however not be established in the

framework of this thesis. Before doing so, it is recommended to further expand our

knowledge under laboratory experiments. As mentioned above, it would be interesting

to integrate higher diversity levels and natural nematode communities into the study.

A next step might then be to simulate tidal floods, temperature fluctuations, day

and night rhythm, etc in experiments with manipulated species composition and

natural nematode assemblages in order to test their impact on the observations under

controlled conditions (interspecific interactions, the relationship between nematodes

and bacteria, effects of presence of nematodes on decomposition rates). However,

before doing so, a field study might be required to assess under which tidal regime the

nematode species from our experiment occur. Our colonization experiment has for

Conclusions

108

instance shown that these species are absent in the lower parts of the salt marsh. They

presumably prefer sites higher up the marsh which are flooded only occasionally

(Alkemade et al., 1993; Moens, pers. comm.). Another observation in the field

samples of the colonization experiment was that fungal growth was considerably less

compared to laboratory samples; this could thus indicate that the bacterial community

is favored under fluctuating conditions. The impact of the nematodes on the

bacterial community and on the decomposition rate may be altered in a bacteria-

dominated microbial community. It might therefore also be interesting to integrate a

bacterial based food web into the laboratory incubations, using a different substrate,

so that the impact of bacterivorous nematodes in both soil systems can be compared.

Integrating species manipulation in field experiments may be difficult to achieve. Due

to the open character of the system, and the supply of nematodes from surrounding

sediments with the tidal action, excluding species from an experimental field seems

impossible. Including natural communities in the laboratory experiments could

however enable an evaluation of the processes observed under controlled conditions.

Colonization experiment

Nematode distribution

Under controlled laboratory conditions nematodes from salt marsh sediments have

the ability to actively colonize and exploit a resource-rich microhabitat, such as

freshly deposited cordgrass detritus, but they can not take advantage of this ability

in a physically disturbed field environment. This is most probably due to the daily

tidal disturbance which suspends the nematodes and distributes them over the

sediments surface.

The species composition in our samples showed that the occurrence of mainly

Monhysterida and Rhabditida on phytodetritus may not be as universal as previously

thought, and is presumably restricted to higher, more sheltered parts of intertidal areas

which are not flooded daily (Alkemade et al., 1993; Moens, pers. comm.). These more

stable conditions could favor their growth and might allow active migration of the

nematode to determine their micro-distribution, similar to our laboratory incubations

in this experiment.

Conclusions

109

Functional aspects of the nematode community

Some functional aspects of the nematode community in the laboratory samples

deviated from our expectations. Epigrowth feeders were assumed to be hampered due

to growth inhibition of their supposed food sources, namely microphytobenthos,

under dark laboratory conditions. However, their proportion of the total community

did not differ from field samples. Hence, they must be able to utilize other resources,

presumably bacteria. Since mostly the same species occurred at both sites, this

indicates that the functional position of a species might differ according to food

availability.

The main food items for deposit feeders have been identified as bacteria and diatoms.

While some deposit-feeders adapted very well to a bacterial diet in the laboratory (e.g.

Axonolaimus typicus), others remained highly dependent on the presence of diatoms

as a food source (e.g. Daptonema biggi). These observations point to substantial

variation in diet between nematode species within functional groups.

These observations on both the epistrate feeders and deposit feeders confirm the

above formulated conclusions drawn from the species manipulation experiment

in that the current trophic classification of free living marine nematodes is

inadequate to make predictions on the function of the nematode species in an

ecosystem.

Diversity of the nematode community

Several mechanisms have been proposed by which nematode diversity in soils could

be maintained (Ettema, 1998). Physical disturbance is one of them and could be an

important process in maintaining high diversity of nematodes with similar resource

needs in intertidal environments, such as salt marshes. However, equally high

nematode diversity under controlled laboratory conditions and in field samples

was found. In other words, disturbance does not seem to be essential in maintaining

these high diversities in the studied system; biotic interactions obviously led to a

stable coexistence of comparable numbers of (functionally similar) species.

Resource partitioning might explain the coexistence of the nematode species in our

model system. Still, stronger competition in the laboratory could be expected as the

spectrum of available food items was restricted: most species depend on bacteria for

Conclusions

110

their growth since the development of the other main food source, microphytobenthos,

was hampered. However, this did not result in a reduction of nematode diversity. In

accordance with conclusions of the species manipulation experiment, chemical

interactions supporting coexistence of highly diverse communities could be

considered. Accumulations of chemicals might have been less severe than in the

species manipulation experiments, as the scale of the microcosms was considerably

larger, but nonetheless it might be part of the reason for the lower densities in the

sediment in the laboratory than in the field.

Appendix

Appendix

113

10 20 30 40 50 65 750

0,2

0,4

0,6

0,8

1

Res

pira

tion

rate

(µm

ol O

2h-1

)

Days10 20 30 40 50 65 75

0

0,2

0,4

0,6

0,8

1

Res

pira

tion

rate

(µm

ol O

2h-1

)

Days

Res

pira

tion

rate

(µm

ol O

2h-1

)

0

0,2

0,4

0,6

0,8

1

10 20 30 40 50 65 75Days

Res

pira

tion

rate

(µm

ol O

2h-1

)

0

0,2

0,4

0,6

0,8

1

10 20 30 40 50 65 75Days

0

0,2

0,4

0,6

0,8

1

10 20 30 40 50 65 750

0,2

0,4

0,6

0,8

1

10 20 30 40 50 65 7510 20 30 40 50 65 75Days

Res

pira

tion

rate

(µm

ol O

2h-1

)

0

0,2

0,4

0,6

0,8

1

10 20 30 40 50 65 75Days

Res

pira

tion

rate

(µm

ol O

2h-1

)

0

0,2

0,4

0,6

0,8

1

10 20 30 40 50 65 75Days

0

0,2

0,4

0,6

0,8

1

10 20 30 40 50 65 750

0,2

0,4

0,6

0,8

1

10 20 30 40 50 65 7510 20 30 40 50 65 75Days

A

B

C

D

S

A

B

C

D

S

E

G

I

J

S

E

G

I

J

S

L

N

S

L

N

S

1 nematode species

2 nematode species

3 nematode species

Temporal changes in respiration rate per treatment (Chapter 4 - p. 64)

Appendix

114

0

5000

10000

15000

20000

25000

30000

10 20 30 40 50 65 75

Prot

eoly

tic a

ctiv

ity (c

ount

s min

-1)

Days

0

5000

10000

15000

20000

25000

30000

10 20 30 40 50 65 75

Prot

eoly

tic a

ctiv

ity (c

ount

s min

-1)

0

5000

10000

15000

20000

25000

30000

10 20 30 40 50 65 750

5000

10000

15000

20000

25000

30000

0

5000

10000

15000

20000

25000

30000

10 20 30 40 50 65 7510 20 30 40 50 65 75

Prot

eoly

tic a

ctiv

ity (c

ount

s min

-1)

Days

0

5000

10000

15000

20000

25000

30000

10 20 30 40 50 65 75

Prot

eoly

tic a

ctiv

ity (c

ount

s min

-1)

Days

0

5000

10000

15000

20000

25000

30000

10 20 30 40 50 65 75

Prot

eoly

tic a

ctiv

ity (c

ount

s min

-1)

0

5000

10000

15000

20000

25000

30000

10 20 30 40 50 65 750

5000

10000

15000

20000

25000

30000

10 20 30 40 50 65 7510 20 30 40 50 65 75

Prot

eoly

tic a

ctiv

ity (c

ount

s min

-1)

Days

0

5000

10000

15000

20000

25000

30000

35000

10 20 30 40 50 65 75

Prot

eoly

tic a

ctiv

ity (c

ount

s min

-1)

Days

0

5000

10000

15000

20000

25000

30000

35000

10 20 30 40 50 65 75

Prot

eoly

tic a

ctiv

ity (c

ount

s min

-1)

0

5000

10000

15000

20000

25000

30000

35000

10 20 30 40 50 65 750

5000

10000

15000

20000

25000

30000

35000

0

5000

10000

15000

20000

25000

30000

35000

10 20 30 40 50 65 7510 20 30 40 50 65 75

Prot

eoly

tic a

ctiv

ity (c

ount

s min

-1)

Days

A

B

C

D

S

A

B

C

D

S

E

G

I

J

S

E

G

I

J

S

L

N

S

L

N

S

1 nematode species

2 nematode species

3 nematode species

Temporal changes in proteolytic activity per treatment (Chapter 4 - p. 66)

Appendix

115

Shan

non-

Wie

ner I

ndex

(H’)

1,51,7

1,92,1

2,32,5

2,72,9

3,1

30 50 65Days

Shan

non-

Wie

ner I

ndex

(H’)

1,51,7

1,92,1

2,32,5

2,72,9

3,1

30 50 65Days

1,5

1,7

1,9

2,1

2,3

2,5

2,7

2,9

3,1

30 50 65Days

Shan

non-

Wie

ner I

ndex

(H’)

1,5

1,7

1,9

2,1

2,3

2,5

2,7

2,9

3,1

30 50 651,5

1,7

1,9

2,1

2,3

2,5

2,7

2,9

3,1

30 50 65Days

Shan

non-

Wie

ner I

ndex

(H’)

1,51,7

1,92,12,3

2,52,7

2,93,1

30 50 65

Shan

non-

Wie

ner I

ndex

(H’)

Days

1,51,7

1,92,12,3

2,52,7

2,93,1

30 50 65

Shan

non-

Wie

ner I

ndex

(H’)

Days

1 nematode species

2 nematode species

3 nematode species

A

B

C

D

S

A

B

C

D

S

E

G

I

J

S

E

G

I

J

S

L

N

S

L

N

S

Temporal changes in bacterial diversity per treatment (Chapter 4 - p. 68)

Appendix

116

100

150

200

250

300

350

400

10 20 30 40 50 65 75

mg

C

Days

100

150

200

250

300

350

400

10 20 30 40 50 65 75100

150

200

250

300

350

400

10 20 30 40 50 65 75

mg

C

Days

100

150

200

250

300

350

400

10 20 30 40 50 65 75

mg

C

Days

100

150

200

250

300

350

400

10 20 30 40 50 65 75100

150

200

250

300

350

400

10 20 30 40 50 65 75

mg

C

Days

100

150

200

250

300

350

400

10 20 30 40 50 65 75

mg

C

Days

100

150

200

250

300

350

400

10 20 30 40 50 65 75100

150

200

250

300

350

400

10 20 30 40 50 65 75

mg

C

Days

1 nematode species

2 nematode species

3 nematode species

A

B

C

D

S

A

B

C

D

S

E

G

I

J

S

E

G

I

J

S

L

N

S

L

N

S

Temporal changes in C-content per treatment (Chapter 4 - p. 70)

Appendix

117

mg

N

0

2

4

6

8

10

12

Days10 20 30 40 50 65 75

mg

N

0

2

4

6

8

10

12

Days10 20 30 40 50 65 7510 20 30 40 50 65 75

0

2

4

6

8

10

12

10 20 30 40 50 65 75

mg

N

Days

0

2

4

6

8

10

12

10 20 30 40 50 65 7510 20 30 40 50 65 75

mg

N

Days

0

2

4

6

8

10

12

10 20 30 40 50 65 75

mg

N

Days

0

2

4

6

8

10

12

10 20 30 40 50 65 750

2

4

6

8

10

12

10 20 30 40 50 65 75

mg

N

Days

A

B

C

D

S

A

B

C

D

S

E

G

I

J

S

E

G

I

J

S

L

N

S

L

N

S

1 nematode species

2 nematode species

3 nematode species

Temporal changes in N-content per treatment (Chapter 4 - p. 71)

Appendix

118

Overview DGGE-profiles (Chapter 4 - p. 68)

D. meyli

D. oschei

P. paetzoldi

D. dievengatensis

Control

30 days50 days65 days





D. meyli

D. oschei

P. paetzoldi

D. dievengatensis

Control











Appendix

119

D. meyli + D. oschei

P. paetzoldi + D. dievengatensis

D. oschei + D. dievengatensis

D. meyli + D. oschei + D. dievengatensis

D. meyli + P. paetzoldi + D. dievengatensis






D. meyli + D. oschei

P. paetzoldi + D. dievengatensis

D. oschei + D. dievengatensis

D. meyli + D. oschei + D. dievengatensis

D. meyli + P. paetzoldi + D. dievengatensis











Overview DGGE-profiles (Chapter 4 - p. 68)

Cited literature

Cited literature

123

Cited literature

Abrams BI, Mitchell MJ (1980) Role of nematode-bacterial interactions in

heterotrophic systems with emphasis on sewage sludge decomposition. Oikos 35: 404-

410

Alkemade R, Wielemaker A, Hemminga MA (1992a) Stimulation of Spartina anglica

leaves by the bacterivorous marine nematode Diplolaimelloides bruciei. J Exp Mar

Biol Ecol 159: 267-278

Alkemade R, Wielemaker A, de Jong SA, Sandee AJJ (1992b) Experimental evidence

for the role of bioturbation by the marine nematode Diplolaimella dievengatensis in

stimulating the mineralization of Spartina anglica detritus. Mar Ecol Prog Ser 90:

149-155

Alkemade R, Wielemaker A, Hemminga MA (1993) Correlation between nematode

abundance and decomposition rate of Spartina anglica leaves. Mar Ecol Prog Ser 99:

293-300

Alkemade R, Wielemaker A, Herman PMJ, Hemminga MA (1994) The population

dynamics of Diplolaimelloïdes bruciei, a nematode associated with the salt marsh

plant Spartina anglica. Mar Ecol Prog Ser 105: 277-284

Aller RC, Aller JY (1992) Meiofauna and solute transport in marine muds. Limnol

Oceanogr 37: 1018-1033

Altschul SF, Gish W, Miller W, Myers EW, Lipman DJ (1990) Basic local alignment

search tool. J Mol Biol 215: 403-410

Anderson RV, Coleman DC, Cole CV, Elliott ET (1981) Effect of the nematodes

Acrobeloides sp. and Mesodiplogaster lheritieri on substrate utilization and nitrogen

and phosphorus mineralization in soil. Ecology 62: 549-555

Anderson RV, Gould WD, Woods LE, Cambardella C, Ingham RE, Coleman DC

(1983) Organic and inorganic nitrogenous losses by microbivorous nematodes in soil.

Oikos 40: 75-80

Cited literature

124

Armonies W (1988) Hydrodynamic factors affecting behaviour of intertidal

meiobenthos. Ophelia 28: 183-193

Beare MH, Parmelee RW, Hendrix PF, Cheng W, Coleman DC, Crossley DA (1992)

Microbial and faunal interactions and effects on litter nitrogen and decomposition in

agroecosystems. Ecol Monogr 62: 569-591

Bell SS, Sherman KM (1980) A field investigation of meiofauna dispersal: tidal

resuspension and implications. Mar Ecol Prog Ser 3: 245-249

Bell SS, Watzin MC, Coull BC (1978) Biogenic structure and its effect on the spatial

heterogeneity of meiofauna in a salt marsh. J Exp Mar Biol Ecol 35: 99-107

Bertelsen RD (1998) Active and passive settling by marine benthic nematodes. Diss

Abstr Int B Sci Eng 58: 3425

Bone LW (1982) Reproductive chemical communication of helminthes. II.

Aschelminthes. Internat J Invert Reprod 5: 311-321

Bongers T, Bongers M (1998) Functional diversity of nematodes. Appl Soil Ecol 10:

239-251

Bongers T, Alkemade R, Yeates GW (1991) Interpretation of disturbance-induced

maturity decrease in marine nematode assemblages by means of the Maturity Index.

Mar Ecol Prog Ser 76: 135-142

Bongers T, de Goede RGM, Korthals GW, Yeates GW (1995) Proposed changes of c-

p classification for nematodes. Russ J Nematol 3: 61-62

Bonkowski M, Griffiths B, Scrimgeour C (2000) Substrate heterogeneity and

microfauna in soil organic ‘hotspots’ as determinants of nitrogen capture and growth

of ryegrass. Appl Soil Ecol 14: 37-53

Boucher G, Chamroux S (1976) Bacteria and meiofauna in an experimental sand

ecosystem. I. Material and preliminary results. J Exp Mar Biol Ecol 24: 237-249

Cited literature

125

Bouwman LA, Romeyn K, Kremer DR, Van Es FB (1984) Occurrence and feeding

biology of some nematode species in estuarine aufwuchscommunities. Cah Biol Mar

25: 287-303

Buth GJC, de Wolf L (1985) Decomposition of Spartina anglica, Elytrigia pungens

and Halimione portulacoides in a Dutch salt marsh in association with faunal and

habitat influences. Vegetatio 62: 337-355

Buth GJC, Voesenek LACJ (1988) Respiration of standing and fallen plant litter in a

Dutch salt marsh. In Vegetation structure in relation to carbon and nutrient economy.

Verhoeven JTA, Heil GW, Werger MJ (eds) S.P.B. Academic Publishing, The Hague:

51-60

Buzas MA (1969) Foraminiferal species densities and environmental variables in an

estuary. Limnol Oceanogr 14: 411-422

Cardinale BJ, Palmer MA, Collins SL (2002) Species diversity enhances ecosystem

functioning through interspecific facilitation. Nature 415: 426-429

Carpenter SR (1996) Microcosm experiments have limited relevance for community

and ecosystem ecology. Ecology 77: 677-680

Chandler GT, Fleeger JW (1987) Facilitative and inhibitory interactions among

estuarine meiobenthic harpacticoid copepods. Ecology 68: 1906-1919

Clarke KR (1993) Non-parametric multivariate analyses of change in community

structure. Aus J Ecol Syst 3: 169-192

Clarke KR, Gorley RN (2001) PRIMER v5: User Manual/Tutorial. PRIMER-E,

Plymouth

Clarke KR, Warwick RM (2001) Change in marine communities: an approach to

statistical analysis and interpretation, 2nd edition. PRIMER-E, Plymouth

Coldeira MC, Ryel RJ, Lawton JH, Pereira JS (2001) Mechanisms of positive

biodiversity-production relationships: insights provided by delta C-13 analysis in

experimental Mediterranean grassland plots. Ecol Letters 4: 439-443

Cited literature

126

Coleman DC, Reid CPP, Cole CV (1983) Biological strategies of nutrient cycling in

soil systems. In Advances in ecological research. Volume 13. Macfayden A, Ford ED

(eds) Academic Press, New York, USA

Collins S (1995) The measurement of stability in grasslands. Trends Ecol Evol 10: 95-

96

Commito JA, Tita G (2002) Differential dispersal rates in an intertidal meiofauna

assemblage. J Exp Mar Biol Ecol 268: 237-256

Connell JH (1978). Diversity in tropical rain forests and coral reefs. Science 199:

1302-1311

Coull BC (1988) Ecology of the marine meiofauna. In Introduction to the study of

meiofauna. Higgins RP, Thiel H (eds) Smithsonian Institution Press, Washington DC,

London: 18-38

Coull BC (1990) Are members of the meiofauna food for higher trophic levels? Trans

Am Microsc Soc 109: 233-246

Cullen DJ (1973) Bioturbation of superficial marine sediments by interstitial

meiobenthos. Nature 242: 323-324

Danovaro R, Gambi C (2002) Biodiversity and trophic structure of nematode

assemblages in seagrass systems: evidence for a coupling with changes in food

availability. Mar Biol 141: 667-677

David AR (1996) Association among ascidians: facilitation of recruitment in Pyura

spinifera. Mar Biol 126: 34-41

Degans H, Zöllner E, Van der Gucht K, De Meester L, Jürgens K (2002) Rapid

Daphnia-mediated changes in microbial community structure: an experimental study.

FEMS Microbiol Ecol 42: 137-149

Degens BP (1998) Decreases in microbial functional diversity do not result in

corresponding changes in decomposition under different moisture conditions. Soil

Biol Biochem 30: 1989-2000

Cited literature

127

De Mesel I, Derycke, S, Swings, J, Vincx, M, Moens, T (2003) Influence of

bacterivorous nematodes on the decomposition of cordgrass. J Exp Mar Biol Ecol

296: 227-242

De Mesel I, Derycke S, Moens T, Van der Gucht K, Vincx M, Swings J (2004) Top-

down impact of bacterivorous nematodes on the bacterial community structure: a

microcosm study. Env Microbiol (in press)

Deutsch A (1978) Gut structure and digestive physiology of two marine nematodes,

Chromadorina germanica, Bütschli, 1874, and Diplolaimella sp. Biological Bulletin.

Marine Biological Laboratory, Woods Hole 155: 317-335

Dietrich G, Kalle K (1957) Allgemeine Meereskunde. Eine Einführung in die

Ozeanographie. Gebrüder Borntraeger, Berlin, Nikolassee

Downing AL, Leibold MA (2002) Ecosystem consequences of species richness and

composition in pond food webs. Nature 416: 837-841

Drake JA, Huxel GR, Hewitt CL (1996) Microcosms as models for generating and

testing community theory. Ecology 77: 670-677

Ehrlich PR, Ehrlich AH (1981) Extinction. The causes and consequences of the

disappearance of species. Random House, New York

Eichner CA, Erb RW, Timmis KN, Wagner-Döbler I (1999) Thermal gradient gel

electrophoresis analysis of bioprotection from polluted shocks in the activated sludge

microbial community. Appl Environ Microbiol 65: 102-109

Emmerson MC, Solan M, Emes C, Paterson DM, Rafaelli D (2001) Consistent

patterns and the idiosyncratic effects of biodiversity in marine ecosystems. Nature

411: 73-77

Ettema CH (1998) Soil nematode diversity: species coexistence and ecosystem

function. J Nematol 30: 159-169

Ewel JJ, Mazzarino MJ, Berish CW (1991). Tropical soil fertility changes under

monocultures and successional communities of different structure. Ecol Appl 1: 289-

302

Cited literature

128

Fegley SR (1987) Experimental variation of near-bottom current speeds and its effects

on depth distribution of sand-living meiofauna. Mar Biol 95: 183-191

Fenchel T (1996) Worm burrow and oxic microniches in marine sediments. 2.

Distribution patterns of ciliated protozoa. Mar Biol 127: 297-301

Ferris H, Venette RC, Lau SS (1997) Population energetics of bacterial-feeding

nematodes: carbon and nitrogen budgets. Soil Biol Biochem 29: 1183-1194

Findlay SEG (1981) Small-scale spatial distribution of meiofauna on a mud –and

sandflat. Est Coast Shelf Sci 12: 471-484

Findlay SEG (1982) Influence of sampling scale on apparent distribution of

meiofauna on a sandflat. Estuaries 5: 322-324

Findlay S, Tenore KR (1982) Effect of a free-living marine nematode (Diplolaimella

chitwoodi) on detrital carbon mineralization. Mar Ecol Prog Ser 8: 161-166

Fleeger JW, Gee JM (1986) Does interference competition determine the vertical

distribution of meiobenthic copepods? J Exp Mar Biol Ecol 95:173-181

Fleeger JW, Chandler GT, Fritzhugh GR, Phillips FE (1984) Effects of tidal currents

on meiofauna densities in vegetated salt marsh sediments. Mar Ecol Prog Ser 19: 49-

53

Freckman DW (1988) Bacterivorous nematodes and organic-matter decomposition.

Agr Ecosyst Environ 24 : 195-217

Fridley JD (2001) The influence of species diversity on ecosystem productivity: How,

where, and why? Oikos 93: 514-526

Fromin N, Hamelin J, Tarnawski S, Roesti D, Jourdain-Miserez K, Forestier N,

Teyssier-Cuvelle S, Gillet F, Aragno M, Rossi P (2002) Statistical analysis of

denaturing gel electrophoresis (DGE) fingerprinting patterns. Environ Microbiol 4:

634-643

Cited literature

129

Garstecki T, Wickham SA (2003) The response of benthic rhizopods to sediment

disturbance does not support the intermediate disturbance hypothesis. Oikos 103: 528-

536

Gee JM (1989) An ecological and economic review of meiofauna as food for fish.

Zool J Linnean Soc 96: 243-261

Gessner MO, Newell SY (1997) Bulk quantitative methods for the examination of

eukaryotic organoosmotrophs in plant litter. In Manual of environmental

microbiology. Hurst CJ, Knudsen GR, McInerney MJ, Stetzenbach LD, Walter MV

(eds) ASM press, Washington DC: 295-308

Givnish TJ (1994) Does diversity beget stability? Nature 371: 113-114

Green C (1971) Mating and host finding behaviour of plant nematodes. In Plant

parasitic nematodes. Vol. II. Zuckerman BM, Rohde RA (eds) Academic Press, New

York: 247-266

Grewal PS (1991) Relative contribution of nematodes (Caenorhabditis elegans) and

bacteria towards the disruption of flushing patterns and losses in yield and quality of

mushrooms (Agaricus bisporus). Ann Appl Biol 119: 438-499

Grewal PS, Hand P (1992) Effects of bacteria isolated from a saprophagous rhabditid

nematode Caenorhabditis elegans on mycelial growth of Agaricus bisoporus. J App

Bacteriol 72: 173-179

Grewal PS, Richardson PN (1991) Effects of Caenorhabditis elegans (Nematoda:

Rhabditidae) on yield and quality of the cultivated mushroom Agaricus bisporus. Ann

App Biol 118: 381-394

Grewal PS, Wright DJ (1992) Migration of Caenorhabditis elegans (Nematoda:

Rhabditidae) larvae towards bacteria and the nature of the bacterial stimulus. Fund

Appl Nematol 15: 159-166

Griffiths BS, Bonkowski M, Dobson G, Caul S (1999) Changes in soil microbial

community structure in the presence of microbial-feeding nematodes and protozoa.

Pedobiologia 43: 297-304

Cited literature

130

Griffiths BS, Ritz K, Bardgett RD, Cook R, Christensen S, Ekelund F, Sorensen SJ,

Baath E, Bloem J, de Ruiter PC, Dolfing J, Nicolardot B (2000) Ecosystem response o

pasture soil communities to fumigation induced microbial diversity reductions: an

examination of the biodiversity-ecosystem function relationship. Oikos 90: 279-294

Groenendijk AM (1984) Primary production of four dominant salt-marsh angiosperms

in the SW Netherlands. Vegetatio 57: 143-152

Hahn MW, Höfle MG (2001) Grazing of protozoa and its effect on populations of

aquatic bacteria. FEMS Microbiol Ecol 35: 113-121

Hamerlynck O, Vanreusel A (1993) Mesacanthion diplechma (Nematoda:

Thoracostomopsidae), a link to higher trophic levels? J Mar Biol Ass UK 73: 453-456

Hedlund BP, Geiselbrecht AD, Bair T, Staley JT (1999) Polycyclic aromatic

hydrocarbon degradation by a new marine bacterium, Neptunomonas naphthovorans

gen. nov., sp. nov. Appl Environ Microbiol 65:251-259

Heip C (1975) On the significance of aggregation in some benthic marine

invertebrates. In Proceedings of the ninth European Marine Biology Symposium.

Barnes H (Ed) Aberdeen University Press, Aberdeen: 527-538

Heip C (1976) The spatial pattern of Cyprideis torosa (Jones, 1850) (Crustacea:

Ostracoda). J mar biol Ass UK 56:179-189

Heip C, Engels P (1977). Spatial segregation in copepod species from a brackish

water habitat. J exp mar Biol Ecol 26: 77-96

Heip C, Vincx M, Vranken G (1985) The ecology of marine nematodes. Oceanogr

Mar Biol Annu Rev London 23: 399-489

Hemminga MA, Buth GJC (1991) Decomposition in salt marsh ecosystems of the SW

Netherlands: the effects of biotic and abiotic factors. Vegetatio 92: 73-83

Hemminga MA, Klap VA, van Soelen J, Boon JJ (1993) Effect of salt marsh inundation

on estuarine particulate organic matter characteristics. Mar Ecol Prog Ser 99: 153-161

Cited literature

131

Hemminga MA, Klap VA, van Soelen J, de Leeuw J, Boon JJ (1992) Shifts in seston

characteristics after inundation of a European coastal salt marsh. Limnol Oceanogr 37:

1559-1564

Herman PMJ, Vranken G (1988) Studies of the life-history and energetics of marine

and brackish-water nematodes. II. Production, respiration and food uptake by

Monhystera disjuncta. Oecologia 77: 457-463

Hillebrand H (2003) Opposing effects of grazing and nutrients on diversity. Oikos

100: 592-600

Hippe H, Andreesen JR, Gottschalk G (1992) The genus Clostridium non-medical. In

The Prokaryotes. A Handbook on the Biology of Bacteria: Ecophysiology, Isolation,

Identification, Applications. Second edition. Balows A, Trüper HG, Dworkin M,

Harder W, Schleifer K-H (eds) Springer Verlag, New York: 1804-1866

Hogue EW (1982) Sediment disturbance and the spatial distributions of shallow water

meiobenthic nematodes on the open Oregon Coast. J Mar Res 40: 551-573

Hopper BE, Meyers SP (1967) Population studies on benthic nematodes within a

subtropical seagrass community. Mar Biol 1: 85-96

Howarth RW (1993) Microbial processes in salt-march sediments. In Aquatic

Microbiology. An Ecological Approach. Ford TE (ed). Blackwell Scientific

Publications, Boston: 239-259

Huettel RN (1986) Chemical communicators in nematodes. J Nematol 18: 3-8

Ilieva-Makulec K (2001) A comparative study of the life strategies of two bacterial-

feeding nematodes under laboratory conditions. II. Influence of the initial food level

on the population dynamics of Acrobeloides nanus (De Man 1880) Anderson 1968

and Dolichorhabditis dolichura (Schneider 1866) Andrássy 1983. Pol J Ecol 49: 123-

135

Ingham RE, Trofymow JA, Ingham ER, Coleman DC (1985) Interactions of bacteria,

fungi, and their nematode grazers; effects on nutrient cycling and plant growth. Ecol

Monogr 55: 119-140

Cited literature

132

Ives AR, Foufopoulos J, Klopfer ED, Klug JL, Palmer TM (1996) Bottle or big-scale

studies: how do we do ecology? Ecology 77: 681-685

Jennings JB, Deutsch A (1975) Occurrence and possible adaptive significance of b-

glucuronidase and arylamidase (Leucine aminopeptidase) activity in two species of

marine nematodes. Comparative Biochem Physiol 52A: 611-614

Jensen P (1987) Feeding ecology of free-living aquatic nematodes. Mar Ecol Prog Ser

35: 187-196

Jensen P (1996) Burrows of marine nematodes as centres for microbial growth.

Nematologica 42: 320-329

Johannes RE (1965) Influence of marine protozoa on nutrient regeneration. Limnol

Oceanogr 10: 434-442

Kampichler C, Bruckner A, Kandeler E (2001) Use of enclosed model systems in soil

ecology: a bias towards laboratory research. Soil Biol Biochem 33: 269-275

Kaufman TD, Bloom JR, Lukezic FL (1983) Effect of an extract from saprozoic

nematode-infested compost on the mycelial growth of Agaricus brunnescens. J

Nematol 15: 567-571

Kersters K, De Vos P, Gillis M, Swings J, Vandamme P, Stackebrandt E (2003)

Introduction to the Proteobacteria. http://141.150.157.117:8080/prokPUB/index.htm

Lawton JH (1994) What do species do in ecosystems? Oikos 71: 367-374

Lawton JH (1996) The ecotron facility at Silwood Park: the value of “big bottle”

experiments. Ecology 77:665-669

Lee JJ, Tietjen JH, Mastropaolo C, Rubin H (1977) Food quality and the

heterogeneous spatial distribution of meiofauna. Helgoländer Meeresunter 30: 272-

282

Cited literature

133

Levin LA, Boesch DF, Covich A, Dahm C, Erseus C, Ewel KC, Kneib RT, Moldenke

A, Palmer MA, Snelgrove P, Strayer D, Weslawski JM (2001) The function of marine

critical transition zones and the importance of sediment biodiversity. Ecosystems 4:

430-451

Lillebø AI, Flindt MR, Pardal MÂ, Marques JC (1999) The effect of macrofauna,

meiofauna and microfauna on the degradation of Spartina maritima detritus from a

salt marsh area. Acta Oecol 20: 249-258

Loreau M, Hector A (2001) Partitioning selection and complementarity in biodiversity

experiments. Nature 412: 72-76

Magurran AE (1988) Ecological diversity and its measurement. Princeton University

Press, Princeton, NJ

Massana R, Jürgens K (2003) Composition and population dynamics of planktonic

bacteria and bacterivorous flagellates in seawater chemostat cultures. Aquat Microbiol

Ecol 32: 11-22

McClung CR, Patrjiquin DG, Davis RE (1983) Campylobacter nitrofigilis sp. nov., a

nitrogen-fixing bacterium associated with roots of Spartina alterniflora Loisel. Int J

Syst Bacteriol 33: 605-612

Mikola J, Setälä H (1998) Relating species diversity to ecosystem functioning:

mechanistic backgrounds and experimental approach with a decomposer food web.

Oikos 83: 180-194

Mille-Lindblom C, Tranvik LJ (2003) Antagonism between bacteria and fungi on

decomposing aquatic plant litter. Microb Ecol 45: 173-182

Moens T, Vincx M (1997) Observations on the feeding ecology of estuarine

nematodes. J Mar Biol Assoc UK 77: 211-227

Moens T, Vincx M (1998) On the cultivation of free-living marine and estuarine

nematodes. Helgoländer Meeresunters 52: 115-139

Moens T, Vierstraete A, Vincx M (1996) Life strategies in two bacterivorous marine

nematodes: preliminary results. Mar Ecol 17: 509-518

Cited literature

134

Moens T, Verbeeck L, de Maeyer A, Swings J, Vincx M (1999) Selective attraction of

marine bacterivorous nematodes to their bacterial food. Mar Ecol Prog Ser 176: 165-

178

Montagna PA, Ruber E (1980) Decomposition of Spartina alterniflora in different

seasons and habitats of a northern Massachusetts salt marsh, and a comparison with

other Atlantic regions. Estuaries 3: 61-64

Moore JC, de Ruiter PC, Hunt HW, Coleman DC, Freckman DW (1996) Microcosms

and soil ecology: critical linkages between field studies and modelling food webs.

Ecology 77: 694-705

Mulder CPH, Uliassi DD, Doak DF (2001) Physical stress and diversity-productivity

relationships: The role of positive interactions. Proc Nat Acad Sci US 98: 6704-6708

Muylaert K, Van der Gucht K, Vloemans N, De Meester L, Gillis M, Vyverman W

(2002) Relationship between bacterial community composition and bottom-up versus

top-down variables in four eutrophic shallow lakes. Appl Environ Microbiol 68: 4740-

4750

Muyzer G (1999) DGGE/TGGE a method for identifying genes from natural

ecosystems. Curr Opin Microbiol 2: 317-322

Muyzer G, de Waal EC, Uitterlinden AG (1993) Profiling of complex microbial

populations by Denaturing Gradient Gel Electrophoresis analysis of polymerase chain

reaction-amplified genes coding for 16S rRNA. Appl Environ Microbiol 59: 695-700

Nannipieri P, Ascher J, Ceccherini MT, Landi L, Pietramellara G, Renella G (2003)

Microbial diversity and soil functions. Eur J Soil Sci 54: 655-670

Nehring S, Jensen P, Lorenzen S (1990) Tube-dwelling nematodes: tube construction

and possible ecological effects on sediment-water interfaces. Mar Ecol Prog Ser 64:

123-128

Newell SY (1996) Established and potential impacts of eukaryotic mycelial

decomposers in marine/terrestrial ecotones. J Exp Mar Biol Ecol 200: 187-206

Cited literature

135

Newell SY, Fallon RD, Miller JD (1989) Decomposition and microbial dynamics for

standing, naturally positioned leaves of the salt-marsh grass Spartina alterniflora.

Mar Biol 101: 471-481

Newell SY, Fallon RD, Cal Rodriguez RM, Groene LC (1985) Influence of rain, tidal

wetting and relative humidity on release of carbon dioxide by standing-dead salt-

marsh plants. Oecologia 68: 73-79

Nübel U, Garcia-Pichel F, Kühl M, Muyzer G (1999) Quantifying microbial diversity:

morphotypes, 16S rRNA genes, and carotenoids of oxygenic phototrophs in microbial

mats. App Env Microbiol 65: 422-430

Nübel U, Engelen B, Felske A, Snaidr J, Wieshuber A, Amann RI, Ludwig W,

Backhaus H (1996) Sequence heterogeneities of genes encoding 16S rRNAs in

Paenibacillus polymyxa detected by temperature gradient gel electrophoresis. J

Bacteriol 178: 5636-5643

O’Donnell AG, Seasman M, Macrae A, Waite I, Davies JT (2001) Plants and

fertilizers as drivers of change in microbial community structure and function in soils.

Plant and Soil 232: 135-145

Odum EP (1971) Fundamentals of ecology. Saunders College Publishing,

Philadelphia

Ólafsson E (1992) Small-scale spatial distribution of marine meiobenthos: the effects

of decaying macrofauna. Oecologia 90 : 37-42

Padgett DE, Courtney TH, Sizemore RK (1985) A technique for distinguishing

between bacterial and non-bacterial respiration in decomposing Spartina alterniflora.

Hydrobiologia 112: 113-119

Perkins EJ (1958) The food relationships of the microbenthos with particular

reference to that found at Whitstable, Kent. Ann Mag Nat Hist 13: 64-77

Cited literature

136

Pot B, Gillis M, De Ley J (1992) The genus Oceanospirillum. In The Prokaryotes. A

Handbook on the Biology of Bacteria: Ecophysiology, Isolation, Identification,

Applications. Second edition. Balows A, Trüper HG, Dworkin M, Harder W, Schleifer

K-H (eds) Springer Verlag, New York: 3230-3236

Powers SP (1998) Recruitment of soft-bottom benthos (benthic invertebrates,

encrusting community, infaunal community. Diss Abstr Int B Sci Eng 58: 5760

Reice SE, Stiven AE (1983) Environmental patchiness, litter decomposition and

associated faunal patterns in a Spartina alterniflora marsh. Est Coast Shelf Sci 16:

559-571

Reichenbach H (1992) The order Cytophagales. In The Prokaryotes. A Handbook on

the Biology of Bacteria: Ecophysiology, Isolation, Identification, Applications.

Second edition. Balows A, Trüper HG, Dworkin M, Harder W, Schleifer K-H (eds)

Springer Verlag, New York: 3631-3675

Rheinheimer G (1992) Aquatic Microbiology, 4rd ed. John Wiley & Sons, UK

Riemann F, Helmke E (2002) Symbiotic relations of sediment-agglutinating

nematodes and bacteria in detrital habitats: the enzyme-sharing concept. Mar Ecol 23:

93-113

Riemann F, Schrage M (1978) The mucus-trap hypothesis on feeding of aquatic

nematodes and implications for biodegradation and sediment texture. Oecologia 34:

75-88

Rieper-Kirchner M (1989) Microbial degradation of North Sea macroalgae: field and

laboratory studies. Bot Mar 32: 241-252

Romeyn K, Bouwman LA (1983) Food selection and consumption by estuarine

nematodes. Hydrobiol Bull 17: 103-109

Rønn R, McCaig AE, Griffiths BS, Prosser JI (2002) Impact of protozoan grazing on

bacterial community structure in soil microcosms. Appl Environ Microbiol 68: 6094-

6105

Cited literature

137

Ruby EG (1992) The genus Bdellovibrio. In The Prokaryotes. A Handbook on the

Biology of Bacteria: Ecophysiology, Isolation, Identification, Applications. Second

edition. Balows A, Trüper HG, Dworkin M, Harder W, Schleifer K-H (eds) Springer

Verlag, New York: 3400-3415

Schauer M, Massana R, Pedrós-Alió C (2000) Spatial differences in bacterioplankton

composition along the Catalan coast (NW Mediterranean) assessed by molecular

fingerprinting. FEMS Microbiol Ecol 33: 51-59

Schläpfer F, Schmid B, Seidl I (1999) Expert estimates about effects of biodiversity on

ecosystem processes and services. Oikos 84: 346-352

Service SK, Bell SB (1987) Density-influenced active dispersal of harpacticoid

copepods. J Exp Mar Biol Ecol 46: 49-62

Shannon CE, Weaver W (1949) The mathematical theory of communication.

University of Illinois Press, Urbana

Sherman KM, Coull BC (1980) The response of meiofauna to sediment disturbance. J

Exp Mar Biol Ecol 46: 59-71

Silvertown J, Dodd ME, McConway K, Potts J, Crawley M (1994) Rainfall, biomass

variation, and community composition in the Park Grass Experiment. Ecology 75:

2430-2437

Sohlenius B (1968) Influence of microorganisms and temperature upon some

rhabditid nematodes. Pedobiologia 8: 137-145

Steyaert M, Vanaverbeke J, Vanreusel A, Barranguet C, Lucas C, Vincx M (2003)

The importance of fine-scale, vertical profiles in characterising nematode community

structure. Estuar Coast Shelf Sci 58: 353-366

Sun B, Fleeger JW, Carney RS (1993) Sediment microtopography and the small-scale

spatial distribution of meiofauna. J Exp Mar Biol Ecol 167: 73-90

Tenore KR, Tietjen JH, Lee JJ (1977) Effect of meiofauna on incorporation of aged

eelgrass, Zostera marina, detritus by the polychaete Nephthys incisa. J Fish Res Bd

Canada 34: 563-567

Cited literature

138

Tietjen JH (1980) Microbial-meiofaunal interrelationships: a review. In Aquatic

Microbial Ecology. Colwell RR, Foster J (eds) University of Maryland Press,

Maryland: 335-338

Tietjen JH, Lee JJ (1973) Life-history and feeding habits of the marine nematode

Chromadora macrolaimoids Steiner. Oecologia 12: 303-314

Tietjen JH, Lee JJ (1977) Feeding behaviour of marine nematodes. In Ecology of

marine benthos. Coull BC (ed) Univ. South Carolina Press, Columbia: 22-36

Tilman D, Downing JA (1994) Biodiversity and stability in grasslands. Nature 367:

363-365

Tilman D, Wedin D, Knops J (1996) Productivity and sustainability influenced by

biodiversity in grassland ecosystems. Nature 379: 718-720

Trotter DB, Webster JM (1984) Feeding preferences and seasonality of free-living

marine nematodes inhabiting the kelp Macrocystis integrifolia. Mar Ecol Prog Ser 14:

151-157

Troussellier M, Schafer H, Batailler N, Bernard L, Courties C, Lebaron P, Muyzer G,

Servais P, Vives-Rego J (2002) Bacterial activity and genetic richness along an

estuarine gradient (Rhone River plume, France) Aquat Microb Ecol 28: 13-14

Ullberg J, Ólafsson E (2003) Free-living marine nematodes actively choose habitat

when descending from the water column. Mar Ecol Prog Ser 260: 141-149

Valaeys T, Topp E, Muyzer G, Macheret V, Laguerre G, Rigaud A, Soulas G (1997)

Evaluation of denaturing gradient gel electrophoresis in the detection of 16S rDNA

sequence variation in rhizobia and methanotrophs. FEMS Microbiol Ecol 24: 279-285

Valiela I (1995) Marine Ecological Processes. Springer-Verlag, New York: 305-324

Van der Gucht K, Sabbe K, De Meester L, Vloemans N, Zwart G, Gillis M,

Vyverman W (2001) Contrasting bacterioplankton community composition and

seasonal dynamics in two neighbouring hypertrophic freshwater lakes. Environ

Microbiol 3: 680-690

Cited literature

139

Van Hannen EJ, Veninga M, Bloem J, Gons HJ, Laanbroek HJ (1999) Genetic

changes in the bacterial community structure associated with protistan grazers. Arch

Hydrobiol 145: 25-38

Venette RC, Ferris H (1998) Influence of bacterial type and density on population

growth of bacterial-feeding nematodes. Soil Biol Biochem 30: 949-960

Verhoef HA (1996) The role of soil microcosms in the study of ecosystem processes.

Ecology 77: 685-690

von Thun W (1966) Eine Methode zur Kultivierung der Mikrofauna. Veröffn Inst

Meeresforsch Bremerh 2: 277-280

von Thun W (1968) Autökolgische Untersuchungen an freilebenden Nematoden des

Brackwassers. PhD thesis, Universität Kiel

Walker BH (1992) Biodiversity and ecological redundancy. Conserv Biol 6: 18-23

Walker BH (1995) Conserving biological diversity through ecosystem resilience.

Conserv Biol 9: 747- 752

Wardle DA, Bonner KI, Nicholson KS (1997) Biodiversity and plant litter:

experimental evidence which does not support the view that enhanced species

richness improves ecosystem function. Oikos 79: 249-258

Warwick RM (1981) Survival strategies of meiofauna. In Feeding and survival

strategies of estuarine organisms. Jones NV, Wolff WJ (eds) Plenum Press, New

York: 39-52

Warwick RM (1987) Meiofauna: their role in marine detrital systems. Detritus and

microbial ecology in aquaculture. ICLARM conference proceedings 14: 282-295

Warwick RM, Carr MR, Clarke KR, Gee JM, Green RH (1988) A mesocosm

experiment on the effects of hydrocarbon and copper pollution on a sublittoral soft-

sediment meiobenthic community. Mar Ecol Prog Ser 46: 181-191

Cited literature

140

Wetzel MA, Jensen P, Giere O (1995) Oxygen/sulfide regime and nematode fauna

associated with Arenicola marina burrows: new insights in the thiobios case. Mar Biol

124: 301-312

White DS, Howes BL (1994) Nitrogen incorporation into decomposing litter of

Spartina alterniflora. Limnol Oceanogr 39: 133-140

Wieser W (1953) Die Beziehung zwischen Mundhöhlengestalt, Ernährungsweise und

Vorkommen bei freilebenden marinen Nematoden. Eine ökologische-morphologische

Studie. Ark Zoo. 4: 439-484

Wohl DL, McArthur JV (2001) Aquatic actinomycete-fungal interactions and their

effects on organic matter decomposition: a microcosm study. Microb Ecol 42: 446-

457

Zöllner E, Santer B, Boersma M, Hoppe HG, Jürgens K (2003) Cascading predation

effects of Daphnia and copepods on microbial food web components. Freshwater

Biol 48: 2174-2193

Zwart G, Huismans R, van Agterveld MP, Van de Peer Y, De Rijk P, Eenhoorn H,

Muyzer G, van Hannen EJ, Gons HJ, Laanbroek HJ (1998) Divergent members of the

bacterial division Verrucomicrobiales in a temperate freshwater lake. FEMS

Microbiol Ecol 25: 159-169

Documents

Nematodes associated with phytodetritus in salt marshes