Effects of fungal food quality and starvation on the fatty acid composition of Protaphorura fimata...

Comparative Biochemistry and Physiology Part B 138(2004) 41–52

1096-4959/04/$ - see front matter� 2004 Elsevier Inc. All rights reserved.doi:10.1016/j.cbpc.2004.02.009

Effects of fungal food quality and starvation on the fatty acidcomposition ofProtaphorura fimata (Collembola)

D. Haubert *, M.M. Haggblom , S. Scheu , L. Ruessa, b a a¨

Institut fur Zoologie, Technische Universitat Darmstadt, Schnittspahnstr. 3, 64287 Darmstadt, Germanya ¨ ¨Department of Biochemistry and Microbiology, Cook College, Rutgers University, 76 Lipman Drive, New Brunswick,b

NJ 08901-8525, USA

Received 9 October 2003; received in revised form 29 January 2004; accepted 4 February 2004

Abstract

The lipid pattern of animals is influenced by species, life stage, environmental conditions and diet. We investigatedthe effects of food quality and starvation on the phospholipid(PLFA) and neutral lipid(NLFA) fatty acid pattern of thecollembolan Protaphorura fimata. Collembolans were fed with two common soil fungi,Agrocybe gibberosa andChaetomium globosum, of which the cellular lipid composition was analysed.A. gibberosa was grown on agar withdifferent nitrogen contents, resulting in altered fatty acid patterns and C:N ratios, i.e. fungi of different food quality.Collembolans did not mirror the lipid composition of the fungal diet as the pattern of major NLFAs inP. fimata wasvice versa. Presumably, altered food quality of fungi caused compensatory responses by the collembolans, therebydiminishing the fungal signal. In a further experimentP. fimata (previously maintained withC. globosum) was keptwithout food for up to 4 weeks. Starvation resulted in a decline in the total amount of NLFAs; however, it did not affectthe fatty acid pattern, indicating that NLFAs were degraded indiscriminately. Generally, the PLFA profile of thecollembolans changed only slightly due to variations in diet quality or starvation.� 2004 Elsevier Inc. All rights reserved.

Keywords: Springtails; Neutral lipids; Phospholipids; Starvation; Food quality

1. Introduction

Lipid composition of microorganisms and fungiis frequently used as a biomarker in environmentalsamples. Phospholipid fatty acids(PLFAs) arecomponents of cell membranes and have beenused to quantify and classify microorganisms insoil (Tunlid and White, 1992; Frostegard and˚Baath, 1996; Zelles, 1999). Different groups of˚ ˚microorganisms synthesize specific fatty acidsthrough various biochemical pathways, whichmakes PLFAs effective taxonomic markers to

*Corresponding author. Tel.:q49-6151-165104; fax:q49-6151-166111.

E-mail address: [email protected](D. Haubert).

define community composition(Baath et al., 1998;Navarrete et al., 2000). The unsaturated fatty acids18:2v-6,9 (linoleic acid) and 18:1v-9 (oleic acid)have been applied to estimate fungal biomass(Frostega˚rd and Baath, 1996; Mikola and Setala,¨ ¨1998). In contrast to membrane PLFAs, neutrallipid fatty acids(NLFAs) are predominantly stor-age lipids and often constitute a high percentageof the total lipid content in animals. Due to theirrole as energy reserves NLFAs are closely relatedto nutritional requirements and metabolism(Stan-ley-Samuelson et al., 1988). Recently, fatty acidswere suggested as a tool to investigate soil foodwebs, where direct observation of animals is dif-ficult because of their small size and their cryptic

42 D. Haubert et al. / Comparative Biochemistry and Physiology Part B 138 (2004) 41–52

habitat. Ruess et al.(2002) reported for nematodes,that the lipid composition is controlled by both thenematode species and its diet and concluded thatmonitoring fatty acid patterns of soil animals maytherefore provide an effective tool to detect trophicinteractions in belowground food webs.

Collembola are widespread in soil systems andplay an important role in decomposition processes(Visser, 1985). They are known to feed on soilmicrobiota, including bacteria, fungi, actinomy-cetes and algae(Rusek, 1998). Several studieshave demonstrated the importance of fungi incollembolan nutrition(Visser and Whittaker, 1987;Chen et al., 1995; Klironomos and Kendrick,1995). Food selection in collembolans is influ-enced by physiological factors, such as the nutri-tional status of the fungal hyphae, or their agewith low growing or senescent hyphae being lesspreferred (Booth and Anderson, 1979; Leonard,1984). Differences in fungal food quality(e.g.nitrogen content, fungal host species, mixed diet)affect the reproduction and frequency of moultingin collembolans (Joosse and Testerink, 1977;Walsh and Bolger, 1990; Chen et al., 1995; Scheuand Simmerling, 2004). Starvation strongly affectsthe metabolism of collembolans. Generally, thelack of energy supply results in changes in bodywater content, respiration and reproduction(Hub-ert and Sustr, 2001). Also, food depletion nega-ˇ

tively affects drought and cold tolerance ofcollembolans(Lavy et al., 1997).

Lipids, as stored triacylglycerols, play an essen-tial role as a reserve of metabolic energy, inaddition to the function of phospholipids asmembrane components(Canavoso et al., 1998).However, only a few studies have investigated thefatty acid profiles of collembolan lipids(Stranskyet al., 1986; Holmstrup et al., 2002). To use fattyacids as biomarkers in food webs the factors thataffect their trophic transfer and pattern need to beanalysed. Lipid composition is not fixed and inparticular diet and development may excert stronginfluences on the shape of fatty acid profiles. Therelative abundance of fatty acids in insects isdetermined by the specific type of biosyntheticpathway of the given species(Hanson et al., 1985;Ghioni et al., 1996), the life stage(Ogg andStanley-Samuelson, 1992; Sayah et al., 1997), theenvironmental conditions(Joanisse and Storey,1996) and the diet(Thompson et al., 1973; Fer-nando-Warnakulasuriya et al., 1988; Howard andStanley-Samuelson, 1996). The composition of

NLFAs in the fat body results from differentprocesses including the storage of dietary lipids,de novo synthesis, degradation and subsequentrelease for mobilisation to sites where they aremetabolised(Beenakkers et al., 1985). The patternof PLFAs, which are structurally and functionallyinvolved in biomembranes, is individuallyarranged in specific tissues and controlled at thecellular level (Stanley-Samuelson et al., 1988).The aim of our experiments was to investigate theinfluence of food quality and starvation on(i) thelipid content of collembola and(ii) the pattern ofneutral lipid and phospholipid fatty acids. Varia-tions in dietary nutrient supply is a situationfrequently faced by soil living animals and theresults were expected to provide insight into theresponse in lipid metabolism of collembola.

2. Materials and methods

2.1. Fungi

Two soil fungal species were used as food forcollembolans in the experiments: the ascomyceteChaetomium globosum Kunze and the basidiomy-cete Agrocybe gibberosa (Fr.) Fay. C. globosumwas cultivated on potato dextrose agar(PDA,Merck). A. gibberosa was grown on Pachlewskaagar adjusted to different N concentrations in orderto alter the C:N ratio. Pachlewska agar is a nutrientrich medium that contains 20 g glucose, 5 gmaltose, 0.5 g MgSOØ7 H O, 1 g KH PO , 504 2 2 4

mg thiamine HCl, 0.5 ml 1% wyv Fe citrate, 0.5ml 2% wyv ZnSO and 15 g agar per 1 l distilled4

water. As nitrogen source we added 0.5, 0.75, 1.5and 3 g ammonium–tartrate per 1 l(equivalent to75, 150, 300, 600mg Nyg agar). For the analysisof fatty acids,A. gibberosa was grown in liquidcultures of the same media andC. globosum wasgrown on PDA covered by a membrane filter(Millipore, 0.8 mm). Fungi were kept at 158C(three replicates per treatment).

2.2. Collembolans

Collembolans of the speciesProtaphorura fima-ta were taken from laboratory cultures fed withbakers yeast. Specimens were put into plasticmicrocosms(diameter 7 cm, height 4.5 cm) con-taining a layer of plaster mixed with activatedcharcoal (2:1) at the bottom. Microcosms werekept at 15 8C in darkness and kept moist with

43D. Haubert et al. / Comparative Biochemistry and Physiology Part B 138 (2004) 41–52

distilled water. Each microcosm contained 10 indi-viduals ofP. fimata. This low density was chosento avoid cannibalism due to crowding. Eggs, pel-lets and exuvia were removed once a week. Col-lembolans were fed with fungal diet ad libitum.Round pieces of fungal mats with a diameter of10 mm were cut out of the agar cultures understerile conditions, put into the microcosms andwere renewed once a week.

2.3. Experimental set-up

In the first experiment the effect of starvationon the fatty acid pattern ofP. fimata was investi-gated. The collembolans were fed for 6 weekswith C. globosum. Then the fungus was removedand animals were kept 4 weeks without foodsupply. Specimens were sampled at the beginning(day 0), after 2, 4, 8, 14, 21 and 28 days. Thesecond experiment investigated changes in thefatty acid pattern due to different food qualityrepresented by the different C:N ratio of the agarused as culture media for the fungus(75, 150, 300or 600 mg Nyg agar). P. fimata was fed withA. gibberosa ad libitum for 6 weeks. For bothexperiments three replicates per treatment wereperformed and the collembolans were frozen aty20 8C until analysis.

2.4. Analysis of fatty acid patterns

Cellular lipids of Collembolans harvested fromthe microcosms were divided into different lipidclasses using a method described in Zelles(1999).Collembolan biomass was extracted by shaking in5 ml single phase extraction solvent(chloroformymethanoly0.05 M phosphate buffer(pH 7.4)1:2:0.8) overnight. The solvent was then trans-ferred to new tubes and samples were re-extractedby shaking for 2–3 h with additional 2.5 ml.Extraction solvents of both steps were combined,0.8 ml distilled H O and 0.8 ml CHCl were added2 3

and samples centrifuged at 1500 rpm for 5 min.Samples were allowed to stand and separate. Thenthe top two phases were removed and the chloro-form fraction of each sample were transferred to asilica acid column(0.5 g silicic acid, mesh size100–200 mm). Lipids were eluted with 5 mlchloroform (neutral lipids), 8 ml acetone(glyco-lipids) and 5 ml methanol(phospholipids). Neutrallipids and phospholipids were used for data anal-ysis. Separation of these two lipid classes is

suggested to be adequate, as silica acid columnswere washed with acetone for glycolipids inbe-tween elution of the two fractions. Frostegard et˚al. (1991) report 85% recovery of the total phos-phate in the eluates of the methanol fractions andsuggest the method for extraction and digestion togive reliable results. The chloroform and methanolfraction was reduced by evaporation(50 8C, vac-uum 200 hPa) in a Labconco RapidVap.

Fungal biomass ofA. gibberosa was harvestedfrom liquid cultures(Pachlewska media). For C.globosum membrane filters were stripped of theagar with the adhering fungal mats and hyphaeharvested by scraping from the filter with a sterilescalpel.

Chloroform and methanol fractions of the col-lembolan samples and total fungal biomass weresaponified and methylated following the procedu-res given for the Sherlock Microbial IdentificationSystem(MIDI, Newark, DE, USA). Saponificationof lipids was conducted in a sodium hydroxide–methanol solution(45 g NaOH, 150 ml CH OH,3

150 ml distilled H O) at 100 8C for 30 min,2

followed by acid methanolysis in HCl–methanol(325 ml 6.0 N HCl, 275 ml CH OH) at 80 8C3

for 10 min. The fatty acid methylesters wereextracted into hexane–methyl tertiary butyl ether(1:1) and washed with aqueous NaOH(10.8 gNaOH, 900 ml distilled H O). The lipid-containing2

phase was then transferred to test tubes and storedat y20 8C.

Fatty acid methyl esters were analysed by gaschromatography(GC) using the Sherlock Micro-bial Identification System(MIDI Inc.) consistingof a Hewlett Packard 5890 Series II gas chromat-ograph and flame ionisation detector equipped withan HP Ultra 2 phenyl methyl silicone fused capil-lary column (25 m=0.2 mm i.d., film thickness0.33 mm), an automated sampler and computerwith associated software(Sherlock Pattern Rec-ognition Software, MIDI). The fatty acid methylesters were identified on the basis of their retentiontimes and quantified. To verify correct identifica-tion of fatty acid methyl esters(chain length andsaturation) a range of samples containing NLFAsand all samples with PLFAs were analysed by GC-mass spectrometry using an Agilent Series 6890GC System and 5973 Mass Selective Detector,equipped with a HP5MS capillary column(30m=0.25 mm i.d., film thickness 0.25mm). Gen-erally, the relative amounts of identified fatty acidswere calculated according to data from GC analy-


Table 1Cellular lipids(fatty acids in %"S.D.) of the fungusA. gibberosa grown on Pachlewska agar with 75, 150, 300 or 600mg Nyg agar

Fatty acid composition 75 mg Nyg agar 150 mg Nyg agar 300 mg Nyg agar 600 mg Nyg agar

Saturated12:0 0.7"0.9 0.7"0.7 0.2"0.4 0.2"0.414:0 1.1"0.2 0.7"0.1 0.7"0.1 0.7"0.115:0 2.8"0.5 a 1.3"0.4 b 0.8"0.1 b 0.8"0.1 b16:0 39.3"7.1 a 20.1"1.3 b 19.7"1.3 b 18.4"2.6 b17:0 1.2"0.1 a 0.4"0.4 b 0.2"0.3 b 0 b18:0 4.2"0.9 a 2.4"0.5 b 1.5"0.2 b 1.3"0.4 b

Monoenoic17:1v-8 0 b 0.7"0.2 a 0.2"0.3 ab 0.2"0.3 ab17:1v-9 0.6"0.6 0 0.2"0.3 018:1v-9 2.7"0.7 3.4"1.3 2.2"0.9 2.3"0.8

Polyenoic18:2v-6,9 47.4"8.1 b 70.3"2.4 a 74.5"1.8 a 76.1"2.2 a

C16:C18 0.75"0.22 a 0.26"0.02 b 0.25"0.02 b 0.23"0.04 b

Unsaturation index 0.98"0.16 b 1.45"0.04 a 1.51"0.02 a 1.55"0.05 a

Major FAs in bold. Means within rows sharing the same or no letters are not significantly different from each other(Tukey’s HSD,P-0.05).

sis using the Microbial Identification System(MIDI ). GC mass spectrometry was applied toclarify correct identification of fatty acids. If bio-mass in samples was low, i.e. for most PLFAfractions, the combined data obtained with the twoinstruments were used.

The fluidity of lipids was expressed as the fattyacid unsaturation index(UI). This was calculatedas follows:

wUIs (C:1 * 1)q(C:2 * 2)q(C:3 * 3)

xq(C:4 * 4) y100

where C:1, C:2, C:3 and C:4 represent the propor-tion (%) of fatty acids with 1, 2, 3 and 4 doublebonds, respectively.

2.5. C:N analysis

To determine the C:N ratio of the fungi, driedsamples were weighed into tin capsules and totalC and N content was measured by an elementalanalyser(NA 1500, Carlo Erba, Milan, Italy).

2.6. Statistical analysis

Differences in fatty acid profiles of fungi andcollembolans were analysed using ANOVA. Ifsignificant effects were suggested, differencesbetween means were analysed by Tukey’s highest

significant difference test. Statistical analyses wereperformed using SAS(SAS Institute, Cary, NC,USA).

3. Results

3.1. Fatty acid pattern

Total cellular lipids ofA. gibberosa contained10 predominant FAs with a carbon chain lengthranging from 12 to 18(Table 1). The majorsaturated fatty acid was palmitic acid(16:0) at18.4–39.3% of total FAs. Linoleic acid(18:2v-6,9) ranged from 47.4 to 76.1%. Concentrationsof 12:0, 14:0, 15:0, 17:0, 17:1v-8, 17:1v-9, 18:0and 18:1v-9 were low.C. globosum contained 7predominant FAs with a carbon chain length rang-ing from 15 to 18(data not presented). The majorsaturated fatty acid was palmitic acid at 15% oftotal FAs. The most abundant monoenoic fatty acidwas oleic acid at 22.7% and the only polyenoicfatty acid was linoleic acid at 56.7% of total FAs.Proportions of 15:0, 17:0, 18:0 and 17:1v-9 werelow (altogether 5.6%).

Cellular fatty acids of collembolans were divid-ed into neutral lipids(NLFAs) and phospholipids(PLFAs). Generally, the collembolans contained agreater diversity of FAs than their diet.P. fimatafed C. globosum ad libitum contained 13 NLFAswith a carbon chain length ranging from 14 to 20(Table 2). The major saturated fatty acid waspalmitic acid at 23.5% of total NLFAs. Oleic acid


Table 2Neutral lipid fatty acid pattern(fatty acids in %"S.D.) of the collembolanP. fimata starved for up to 28 days

Fatty acid Starvation period(days)composition

0 2 4 8 14 21 28

Saturated12:0 0 0.7"0.7 0 0 0 0 014:0 0.5"0.4 ab 0.9"0.1 a 0.7"0.1 ab 0.8"0.2 ab 0 b 0.5"0.4 ab 0.6"0.5 ab15:0 1.8"0.1 2.1"0.2 1.7"0.1 1.7"0.3 2.2"0.5 2.2"0.1 1.9"0.316:0 23.5"0.4 23.7"0.3 22.8"0.8 23.0"1.7 24.1"2.5 23.0"1.0 21.8"0.817:0 1.1"0.1 1.2"0.1 1.2"0.2 1.3"0.2 1.7"0.3 1.0"0.8 1.5"0.318:0 4.7"0.2 c 5.2"0.6 bc 5.8"0.2 abc 6.9"0.7 a 6.6"0.8 ab 6.1"0.7 abc 6.1"0.5 abc20:0 0 b 0.2"0.3 ab 0.5"0.4 ab 0.8"0 a 0 b 0 b 0 b

Monoenoic16:1v-7 3.3"0.2 b 3.8"0.4 ab 4.0"0.7 ab 4.0"0.3 ab 4.5"1.0 ab 5.4"0.9 a 5.2"0.8 ab17:1v-8 0.2"0.4 0.7"0 0.7"0 0.5"0.5 0.8"0.8 1.1"1.0 1.0"0.917:1v-9 0.2"0.4 0 0 0 0 0 018:1v-9 34.3"0.2 32.1"1.5 33.2"1.1 31.8"2.3 32.1"1.2 32.4"0.5 32.2"2.2

Polyenoic18:2v-6,9 27.4"1.0 26.5"0.6 26.3"2.1 25.6"0.9 25.2"1.6 25.7"1.3 26.2"0.818:3v-6,9,12 1.1"0.1 1.3"0.3 1.1"0.3 1.7"1.0 1.2"0.2 0.8"0.7 1.1"0.120:2v-6,9 0.2"0.3 0 0.5"0.4 0.3"0.5 0 0 0.2"0.420:4v-6,9,12,15 1.8"0.3 1.7"0.2 1.6"0.2 1.7"0.5 1.7"0.4 2.0"0.2 2.3"0.3

C16:C18 0.43"0.01 0.46"0.02 0.44"0.03 0.46"0.03 0.49"0.05 0.48"0.02 0.45"0.02

Unsaturation index 1.03"0.02 1.00"0.01 1.01"0.04 1.00"0.05 0.98"0.06 1.01"0.02 1.03"0.02


was the most abundant monoenoic fatty acid(34.3%) and linoleic acid the most abundant poly-enoic fatty acid(27.4%). The remaining 14.8% ofNLFAs consisted of 14:0, 15:0, 17:0, 18:0, 16:1v-7, 17:1v-8, 17:1v-9, 18:3v-6,9,12, 20:2v-6,9 and20:4v-6,9,12,15. Several long-chain polyenoic FAsoccurred, in particular with C , such as arachidon-20

ic acid. The PLFA pattern ofP. fimata fed C.globosum ad libitum consisted of 11 FAs with acarbon chain length ranging from 12 to 20(Table3). The major saturated acids were palmitic acidand stearic acid with 20.4% and 22.7% of totalPLFAs, respectively. Oleic acid was the mostabundant monoenoic fatty acid(21.1%) and lino-leic acid the most abundant polyenoic fatty acid(22.4%). The remaining 13.4% consisted of 12:0,14:0, 15:0, 17:0, 20:1v-6, 20:3v-6,9,12 and20:4v-6,9,12,15.

3.2. Starvation

Starvation ofP. fimata for 4 weeks affected thebody mass and water content of the collembolans.Both wet and dry mass were significantly affected(Ps0.0002,Ps0.0053, respectively) and rangedbetween 83–243 and 27–54 mgyind, respectively

(data not presented). Wet mass was 189 mgyindat day 0, and, except for a significant decrease to44% of the initial mass on day 8, remained at asimilar level until the end of the experiment. Drymass of ad libitum fed collembolans was 54 mgyind (day 0). It was reduced to 64% of initial masswithin the first 2 days and decreased further to51% after 14 days, thereafter it remained almostconstant.

During starvation the total amount of NLFAsdeclined significantly (Ps0.02), whereas theamount of PLFAs was unchanged(Ps0.29, Fig.1). A starvation period of 28 days did not signifi-cantly affect the three most abundant NLFAs(palmitic, oleic and linoleic acid;Ps0.49, Ps0.44, Ps0.49, respectively) of the collembolans(Table 2). However, some of the less frequentNLFAs were affected by starvation. Palmitoleicacid (16:1v-7) on day 0 had a significant lowerproportion then on day 21(Ps0.021) and stearicacid (18:0) significantly increased during the first14 days(Ps0.006). Starvation did not affect theC16:C18 ratio of the fatty acids(Ps0.45) andthere was no change in the unsaturation index(Ps0.53, Table 2). Within PLFAs, oleic acid wasthe only acid, which significantly decreased during


Fig. 1. Relative amount of phospholipid fatty acids(PLFAs)and neutral lipid fatty acids(NLFAs) per gram dry mass ofthe collembolanP. fimata starved for up to 28 days. Bars("S.D.) sharing the same letters are not significantly different(Tukey’s HSD,P-0.05).

Table 3Phospholipid fatty acid pattern(fatty acids in %"S.D.) of the collembolanP. fimata starved for up to 28 days

Fatty acid Starvation period(days)composition

0 2 4 8 14 21 28

Saturated12:0 0.1"0.2 0 0.5"0.9 0 0 0 014:0 1.0"1.0 0.8"1.1 1.3"0.5 0.6"0.5 0.9"0.9 0.6"0.3 0.4"0.515:0 0.5"0.9 1.3"1.8 0.3"0.5 0 0 0.1"0.2 016:0 20.4"4.3 20.2"0.5 15.7"3.7 20.2"2.0 20.1"5.1 17.6"1.9 14.8"0.017:0 2.1"2.1 1.7"2.3 2.6"2.2 2.2"2.5 2.6"0.3 2.3"0.3 3.1"0.418:0 22.7"4.2 25.9"1.8 26.0"2.5 27.6"7.5 27.2"1.6 22.3"3.0 20.4"1.6

Monoenoic18:1 v-9 21.1"3.9 a 20.0"2.5 ab 16.0"0.6 ab 21.1"3.3 a 12.8"1.4 b 13.2"0.0 b 17.2"0.7 ab20:1v-6 1.2"2.2 0.7"1.0 1.1"1.9 0 0.8"0.7 0.5"0.5 0

Polyenoic18:2v-6,9 22.4"3.6 19.1"3.0 28.0"4.0 24.0"5.7 27.5"2.1 31.2"4.4 32.3"0.720:2v-6,9 0 0 0 0 0 0.3"0.5 020:3v-6,9,12 1.2"2.2 0.7"1.0 2.3"0.5 0.6"1.0 1.6"1.4 3.2"0.4 3.7"0.120:4v-6,9,12,15 7.2"4.5 9.7"6.4 6.3"0.4 3.8"3.3 6.5"2.0 8.7"0.5 8.2"0.8

C16:C18 0.31"0.03 0.31"0.01 0.28"0.06 0.28"0.01 0.30"0.08 0.26"0.03 0.21"0

Unsaturation index 1.00"0.18 1.00"0.18 1.05"0.11 0.86"0.23 1.00"0.18 1.21"0.11 1.26"0.01


starvation(Table 3, Ps0.0023). The long-chainpolyunsaturated PLFAs 20:3v-6,9,12 and 20:4v-6,9,12,15 increased(Ps0.025,Ps0.086, respec-tively) between day 8 and 28 with food depletion.Starvation did not significantly affect the C16:C18ratio (Ps0.18) or theUI (Ps0.17).

3.3. Nitrogen content

Concentrations of 75 and 150mg Nyg agarresulted in a significantly higher C:N ratio ofA.gibberosa (16 and 19, respectively) compared to300 and 600mg Nyg agar, where the C:N ratiowas 9(P-0.0001). Also, the fatty acid pattern ofthe fungusA. gibberosa grown on agar with 75mg Nyg agar had a higher proportion of palmiticacid (Ps0.0004) and a lower content of linoleicacid (Ps0.0002) compared to fungi grown onagar with higher N content(Table 1). Linoleicacid generally increased and palmitic acid gener-ally decreased with increasing N concentrations.Less frequent fatty acids also significantly differedwith a decline in 15:0, 17:0 and 18:0 with higherN concentration in the agar(Ps0.0002,Ps0.001,Ps0.0008, respectively). Additionally, the UIincreased and C16:C18 ratio decreased from 0.98to 1.55 and 0.75 to 0.23, respectively(P-0.0001,Ps0.0014, respectively).

P. fimata fed with A. gibberosa had a wet massbetween 186 and 253 mgyind and dry mass from29 to 51 mgyind. The N content of the fungus didnot affect the collembolan wet mass(Ps0.51).However, dry mass at intermediate N concentra-


Fig. 2. Relative amount of phospholipid fatty acids(PLFAs)and neutral lipid fatty acids(NLFAs) per gram dry mass ofthe collembolanP. fimata reared on the fungusA. gibberosagrown on agar media with 75, 150, 300 or 600mg Nyg agar.Bars ("S.D.) sharing the same letters are not significantlydifferent (Tukey’s HSD,P-0.05).

tions(150 and 300mg Nyg agar) was significantlyhigher than on high and low N concentrations(Ps0.0001).

The N content of the fungus significantly affect-ed the relative amount of both NLFAs and PLFAsof collembolans. With increasing N concentrationsof the food the amount of PLFAs increased andthe amount of NLFAs decreased(Ps0.0007,Ps0.006, respectively; Fig. 2). The relative amountof PLFAs increased 11 times and that of NLFAswas reduced by 85% in animals fed with fungiwith high N content, compared to the collembolansreared on fungi with low N content.

Among the neutral lipids the proportion ofpalmitic acid significantly increased(P-0.0001)and linoleic acid decreased(Ps0.0002) with high-er N content of the fungal food source(Table 4).The proportion of oleic acid was significantlygreater(P-0.0001) at intermediate N levels(150and 300mg Nyg agar). The C16:C18 ratio signif-icantly increased from 0.36 to 0.66, whereas theUI decreased from 1.06 to 0.77(Ps0.009, P-0.0001, respectively). Changes in PLFAs were lessprevalent (Table 5). However, theUI increasedsignificantly from 0.53 to 1.10(Ps0.046) withincreasing N content of the food source.

4. Discussion

4.1. Fungi

The lipid composition ofC. globosum and A.gibberosa as analysed in this study was similar tothat of other fungal species(Stahl and Klug, 1996;Ruess et al., 2002). The major cellular fatty acidsin fungi are 16:0, 18:0, 18:1 and 18:2(Sancholleand Dalpe, 1993), which were well represented in´both species. Fungal PLFAs and whole cell fattyacids, regardless of culture media, do not containC 20(Weete, 1980; Johnk and Jones, 1994; Heringet al., 1999; Chen et al., 2001), as also observedin our analyses.

The lipid content of fungal species is variableand depends on growth conditions, species, stageof fungal development and age(Weete, 1980). Inour experiment the fatty acid composition ofA.gibberosa varied with the C:N ratio of the agar.The C:N ratio generally reflects the resource qual-ity, with low C:N ratios being associated withhigh, and high C:N ratios with low quality(Weete,1980). This likely affects fungal physiology, result-ing in an altered metabolism and, therefore inchanges in fatty acid patterns. Low C:N ratios inthe medium favor protein synthesis and high ratiosfavor lipid accumulation in fungi, i.e. fungi grownunder nitrogen limiting conditions increase theirlipid content (Weete, 1980). Our results indicatethat this accumulation differs between fatty acids,as the fatty acid pattern changed with resourcequality, which was also apparent in a decrease ofthe C16:C18 ratio and an increase of the unsatur-ation index. Additionally, the morphology ofA.gibberosa was altered. Grown on 75mg Nyg agar,the fungus grew thin and rapidly over the medium,whereas with more N added, fungal mats werethicker and growth was slower.

4.2. Collembolans

We found 15 different NLFAs inP. fimata. Themost abundant were 16:0, 18:1v-9 and 18:2v-6,9,these three together represented 78–85% of totalNLFAs. Prevalence of C 16 and C 18 FAs hasbeen documented in lipids of all insect orders(Fast, 1970; Thompson et al., 1973; Stanley-Samuelson et al., 1988). Holmstrup et al.(2002)found 14 different FAs inFolsomia candida, with91% of the total NLFAs being 16:0, 18:1v-9,18:2v-6,9, 18:0, 20:4v-6,9,12,15 and 20:5v-


Table 4Neutral lipid fatty acid pattern(fatty acids in %"S.D.) of the collembolanP. fimata reared on the fungusA. gibberosa grown onPachlewska agar with 75, 150, 300 or 600mg Nyg agar


Saturated12:0 0.2"0.3 0.5"0.5 0.5"0.5 0.7"1.214:0 0.6"0.2 b 0.4"0.4 b 0.7"0.6 b 2.1"0.1 a15:0 2.4"0.1 b 2.0"0.2 b 2.3"0.1 b 4.2"0.6 a16:0 20.7"0.7 c 22.7"1.1 bc 24.6"0.2 b 30.6"1.8 a17:0 0.9"0.1 0.7"0.1 0.6"0.6 0.7"1.218:0 4.6"0.4 b 4.4"0.6 b 6.3"0.2 a 7.3"0.6 a20:0 0.8"0.2 0.8"0.1 0.8"0.7 0

Monoenoic16:1v-7 3.4"0.7 4.5"0.4 6.6"0.4 5.5"5.117:1v-8 0.9"0.4 a 0 b 0 b 0 b18:1v-9 35.0"1.4 b 43.7"1.0 a 40.7"0.4 a 29.1"2.7 c

Polyenoic18:2v-6,9 26.4"1.7 a 16.8"0.9 b 13.9"1.7 b 18.3"2.7 b18:3v-6,9,12 1.1"0.2 a 0.9"0.2 a 1.2"0.3 a 0 b20:2v-6,9 0.9"0.1 a 0.8"0.1 a 0 b 0 b20:4v-6,9,12,15 2.2"0.2 1.8"0.3 1.7"0.2 1.5"1.3

C16:C18 0.36"0.02 b 0.41"0.02 b 0.50"0.02 ab 0.66"0.16 a

Unsaturation index 1.06"0.04 a 0.93"0.02 b 0.86"0.05 bc 0.77"0.02 c


Table 5Phospholipid fatty acid pattern(fatty acids in %"S.D.) of the collembolanP. fimata reared on the fungusA. gibberosa grown onPachlewska agar with 75, 150, 300 or 600mg Nyg agar


Saturated12:0 2.5"1.7 a 0.1"0.3 b 0 b 0.3"0.4 b14:0 2.1"2.2 1.73"0.6 1.7"1.6 1.5"0.215:0 1.5"2.3 1.4"1.0 0.8"0.7 1.8"0.116:0 29.2"5.6 26.2"9.5 18.2"5.0 18.4"0.317:0 3.6"5.3 2.5"0.3 2.8"1.0 5.1"0.118:0 28.5"5.3 20.6"5.8 20.8"1.5 20.2"0.1

Monoenoic18:1v 9 13.6"1.1 18.5"4.6 17.8"3.6 15.6"0.420:1v 6 0 0 0 3.0"2.1

Polyenoic18:2v 6,9 15.6"9.6 21.7"3.2 20.8"3.6 23.2"1.420:2v 6,9 0 0 3.2"3.1 020:3v 6,9,12 0.6"1.0 2.4"2.3 4.7"1.3 3.5"0.220:4v 6,9,12,15 1.7"2.9 4.7"4.5 9.2"2.2 9.0"0.8

C16:C18 0.53"0.21 0.44"0.19 0.32"0.13 0.31"0.02

Unsaturation index 0.53"0.31 a 0.88"0.25 ab 1.17"0.16 b 1.10"0.05 ab


3,6,9,12,15. These FAs represented 92% of thetotal NLFAs in our study. However, we did notdetect the 20:5v-3,6,9,12,15. Dietary fatty acidscan be absorbed and incorporated without modifi-

cation into body tissues of insects(Stanley-Samuelson et al., 1988). Ichneumonid wasps wereshown to match precisely the fatty acid profiles oftheir hosts, retaining no characteristic pattern of


their own (Thompson, 1973). This suggests thatthe lipid components of collembolans may reflectthe pattern of their food source. In our experimentsthe collembolans generally had a more diverse FAprofile then their diet. They contained more long-chain polyenoic FAs, especially with 20 carbonchain length, which is typical for animals. Ourdata indicate thatP. fimata is able to synthesise16:1v-7, 18:3v-6,9,12, 20:0, 20:2v-6,9 and20:4v-6,9,12,15 fatty acids, as these were notpresent in the fungal diet. Several insect speciesconvert C polyunsaturated FAs to their C prod-18 20

ucts, with elongation and desaturation pathwaysresembling those in mammals(Dadd, 1983; Stan-ley-Samuelson and Nelson, 1993). Linoleic acid,the major precursor in this metabolic pathway, andan essential fatty acid for many animals, was inample supply in the fungal host tissue. Therefore,we do not know if the collembolans were capableof the methyl-desaturation of oleic acid or if theyincorporated the dietary fatty acid. De novo syn-thesis of linoleic and linolenic acid via oleic acidhas been reported in some insects(Blomquist etal., 1982; Stanley-Samuelson et al., 1988).

4.3. Influence of starvation on P. fimata

Generally, starvation leads to the use of foodreserves, such as glycogen and lipids. In the firstperiod of food depletion collembolans are knownto use glycogen and afterwards neutral lipids ofstorage fats(Testerink, 1981). In our investigationthe collembolans were starved for up to 28 daysand wet mass, dry mass and fatty acid compositionwere analysed. Dry mass showed a strong decline,whereas wet mass showed minor changes exceptfor a decrease on day 8–44% of the initial mass.In part this may have been caused by the fact thatwet mass is not constant due to differences inmetabolic status or age of individual collembolans(Hilligsøe and Holmstrup, 2003), and physiologi-cal variation during moulting(Verhoef, 1981).However, the data suggest a high loss in bodywater content after 8 days of starvation. Corre-spondingly, a distinct decrease in body size wasobserved, indicating that the water loss may becaused by elevated evaporation due to an increasein surface-to-volume ratio(Verhoef and Li, 1983;Hilligsøe and Holmstrup, 2003). After 14 dayswet mass of the collembolans was almost as highas at the start of the experiment. As dry massdeclined continuously and wet mass remained

more or less constant, the relative body watercontent of the collembolans increased during thestarvation period. This has been observed for othercollembolans previously(Vannier and Verhoef,1978; Verhoef and Li, 1983; Lavy et al., 1997;Hilligsøe and Holmstrup, 2003). Verhoef and Li(1983) suggested this to be a mode of bloodvolume regulation inOrchesella cincta (Collem-bola), which was in agreement with the constantblood osmolarity they observed.

Neutral lipids represent a large energy reserve,which is used when food is in short supply(Holmstrup et al., 2002). These NLFAs are locatedin fat bodies, which are associated with the midgutof collembola(Hopkin, 1997). Energy can also bestored as glycogen, which is used in periods ofshort food deprivation. Testerink(1981) docu-mented, that the lipid content ofO. cincta staysconstant during the first 3 days of starvation, thendeclines until day 7 and afterwards remains con-stant. He also measured the glycogen content,which decreased within three days to zero. This isconsistent with our results, where dry mass ofP.fimata declined rapidly during the first 3 days andneutral lipid levels stayed constant, indicating thatduring this periodP. fimata used glycogen asenergy source. The decrease in storage lipids andbody mass due to food depreviation is generallystrongest in the first week(Testerink, 1981; Lavyet al., 1997; Hilligsøe and Holmstrup, 2003). Weobserved a comparable pattern in our experiment,where storage lipids and individual dry massdeclined until the second week, and then remainedconstant. Animals respond to food deprivation byslowing down physiological processes, such asmetabolic rate(Verhoef and Li, 1983; Lavy et al.,1997), respiration rate(Hubert and Sustr, 2001)ˇ

and moulting rate(Booth and Anderson, 1979;Verhoef et al., 1988). Our results suggest that inP. fimata it takes approximately 2 weeks to adaptits metabolism to food deprivation. In contrast toNLFAs, the amount of PLFAs was not affected bystarvation, likely due to their function as structuralcomponents of cell membranes, which are not usedfor energy metabolism.

In general, the proportion of most NLFAs didnot change during starvation suggesting that fattyacids were metabolised indiscriminately. Similarly,Canavoso et al.(1998) reported a remarkabledecline in lipid stores in Triatominae(Heterop-tera), without noticeable changes in their compo-sition. The PLFA pattern also remained very


constant during starvation except for oleic acid,which may have been used as a precursor for thelong-chain polyunsaturated FAs that increased after1 week of food depletion.

4.4. Influence of different nitrogen content of thefood source

Many fungi adapt physiologically to a widerrange of concentrations of nitrogen, which maylead to variations in the nutritional quality ofhyphae and affect food preferences of fungal feed-ing collembolans(Shaw, 1988). The C:N ratio ofhyphae is an important determinant of food quality,with a low ratio indicating high protein content.In P. fimata the content of NLFAs was significantlyincreased when fed with fungi with low N content.Lavy and Verhoef(1996) observed for the collem-bolanO. cincta that it compensates for low dietarynitrogen by consuming more food to maintain arelatively constant amount of protein and as aconsequence accumulates more fat. This mecha-nism of nutritional compensation also occurs inother insects (Simpson and Abisgold, 1985;Karowe and Martin, 1989; Simpson and Simpson,1990). Nutritional compensation was likelyresponsible for the increase in the amount ofNLFAs in P. fimata fed with a low nitrogen dietin our experiment. However, the amount of PLFAsincreased with increasing nitrogen content ofhyphae. Presumably, the N content of the foodsource also affected the physiological status of thecollembolans. It has been documented to stimulatethe growth and fecundity of fungivores(Boothand Anderson, 1979; Lavy and Verhoef, 1996).Food richer in nitrogen likely induces the produc-tion of more cells, i.e. membranes, which containPLFAs as major compounds. The activated metab-olism also may need more mitochondria to provideenergy resources for the cells. We observed ahigher proportion of C eicosanoids and other20

unsaturated fatty acids, which were mirrored by asignificant increase in the unsaturation index ofPLFAs with enhanced N availability in the fungalfood source. TheUI is a measure for membranefluidity, and changes indicate altered structural orfunctional demands in biomembranes, likelycaused by the alterations in collembolan metabo-lism due to variations in food quality.

Not only the ratio of storage lipids(NLFAs) tomembrane lipids(PLFAs) differed due to the Ncontent of the food source, also the NLFA pattern

of the collembolans was affected significantly. TheUI decreased from 1.06 to 0.77 with decreasingC:N ratio, mainly because of the significantincrease in 16:0 and the significant decrease in18:1v-9 and 18:2v-6,9. In a different system,growing a flagellate under altered N regimes,Parrish et al.(1998) also observed an increase insaturated FAs with low food N content. However,the obvious changes in the NLFA profile of thecollembolans cannot be explained by the changesin the lipid profile of the fungal food, as thepattern of major FAs in the fungus was vice versa.This suggests that the fatty acid pattern ofP.fimata does not depend entirely on its food source.Similarly, the fatty acid profile of the culturemedium was different from the tissue fatty acidpattern in the Tenebrionid beetleZophobas atra-cus, indicating that the insect fatty acid patterndoes not mirror exactly the fatty acid pattern ofthe diet (Howard and Stanley-Samuelson, 1996).Firstly, this might be because the fatty acid com-position is adjusted to meet the need of individualtissues. Secondly, collembolans consume less foodwhen the N content of the food source is high(Lavy and Verhoef, 1996) and as a consequencethey ingest less fatty acids. The fatty acids neededto build membranes are used directly and are notstored as neutral lipids and, therefore are notdetected in the NLFA fraction. Only the excessFAs are stored, which may diminish the signal ofthe fungal pattern.

In conclusion, starvation did not affect the fattyacid composition of the storage lipids in collem-bolans, whereas the C:N ratio of the fungal foodsource altered the NLFA profile distinctly. How-ever, the NLFA pattern ofP. fimata was moreaffected by the N content than by the FA profileof the fungal resource, likely due to compensatoryresponses to variations in dietary nitrogen. Thisindicates that the same fungus grown on differentsubstrates may induce different fatty acid patternsin the storage lipids of its fungal grazers. Gener-ally, the PLFA pattern of the collembolans wasonly slightly altered by food quality or starvation.This implies that PLFAs, i.e. membrane lipids, areless affected by nutritional variations than NLFAs.However, PLFA patterns may change in the long-term due to altered metabolism caused by fooddeprivation. Our results suggest that the influenceof dietary lipid composition on the pattern ofconsumers may be altered by their metabolism andphysiological status diminishing the signal of indi-


vidual fatty acids. We conclude that fatty acidpatterns are not fixed and that diet and develop-ment exert strong influence on fatty acid profiles.More information on the regulation of these pro-cesses is needed to assign specific fatty acids asbiomarkers in trophic cascades.

Acknowledgments

The financial support of D.H. by the DeutscheForschungsgemeinschaft(DFG, Graduiertenkolleg340, TU Darmstadt) is gratefully acknowledged.The work of L.R. was funded by a grant of theDFG (RU 780y2-1, 2-2), Bonn, Germany.

References

Baath, E., Frostegard, A., Dıaz-Ravina, M., Tunlid, A., 1998.˚ ´ ˜˚ ˚ ˚Microbial community-based measurements to estimateheavy metal effects in soil: the use of phospholipid fattyacid patterns and bacterial community tolerance. Ambio 27,58–61.

Beenakkers, A.M.T., van der Horst, D.J., van Marrewijk,W.J.A., 1985. Insect lipids and lipoproteins and their role inphysiological processes. Prog. Lipid Res. 24, 19–67.

Blomquist, G.J., Dwyer, L.A., Chu, A.J., Ryan, R.O., deRenobales, M., 1982. Biosynthesis of linoleic acid in atermite, cockroach and cricket. Insect Biochem. 12,349–353.

Booth, R.G., Anderson, J.M., 1979. The influence of fungalfood quality on the growth and fecundity ofFolsomiacandida (Collembola: Isotomidae). Oecologia 38, 317–323.

Canavoso, L.E., Bertello, L.E., de Lederkremer, R.M., Rubiolo,E.R., 1998. Effect of fasting on the composition of the fatbody lipid of Dipetalogaster maximus, Triatoma infestansand Panstrongylus megistus (Hemiptera: Reduviidae). J.Comp. Physiol. B168, 549–554.

Chen, B., Snider, R.J., Snider, R.M., 1995. Food preferenceand effects of food type on the life history of some soilCollembola. Pedobiologia 39, 496–505.

Chen, J., Ferris, H., Scow, K.M., Graham, K.J., 2001. Fattyacid composition and dynamics of selected fungal-feedingnematodes and fungi. Comp. Biochem. Physiol. B130,135–144.

Dadd, R.H., 1983. Essential fatty acids: insects and vertebratescompared. In: Mittler, T.E., Dadd, R.H.(Eds.), MetabolicAspects of Lipid Nutrition in Insects. Westview Press,Boulder, CO, pp. 107–142.

Fast, P.G., 1970. Insect lipids. Progress in the chemistry offats 11, 181–198.

Fernando-Warnakulasuriya, G.J.P., Tsuchida, K., Wells, M.A.,1988. Effect of dietary lipid content on lipid transport andstorage during larval development ofManduca sexta. InsectBiochem. 18, 211–214.

Frostegard, A., Tunlid, A., Baath, E., 1991. Microbial biomass˚ ˚ ˚measured as total lipid phosphate in soil of different organiccontent. J. Microbiol. Meth. 14, 151–163.

Frostegard, A., Baath, E., 1996. The use of phospholipid fatty˚ ˚ ˚acid analysis to estimate bacterial and fungal biomass insoil. Biol. Fertil. Soils 22, 59–65.

Ghioni, C., Bell, J.G., Sargent, J.R., 1996. Polyunsaturatedfatty acids in neutral lipids and phospholipids of somefreshwater insects. Comp. Biochem. Physiol. B114,161–170.

Hanson, B.J., Cummins, K.W., Cargill, A.S., Lowry, R.R.,1985. Lipid content, fatty acid composition and the effectof diet on fats of aquatic insects. Comp. Biochem. Physiol.B80, 257–276.

Hering, O., Nirenberg, H.I., Kohn, S., Deml, G., 1999. Char-¨acterisation of isolates ofFusarium oxysporum Schlecht. f.sp.vasinfectum (Atk.) Snyd. & Hans., races 1-6, by cellularfatty acid analysis. J. Phytopathol. 147, 509–514.

Hilligsøe, H., Holmstrup, M., 2003. Effects of starvation andbody mass on drought tolerance in the soil collembolanFolsomia candida. J. Insect Physiol. 49, 99–104.

Holmstrup, M., Hedlund, K., Boriss, H., 2002. Drought accli-mation and lipid composition inFolsomia candida: impli-cations for cold shock, heat shock and acute desiccationstress. J. Insect Physiol. 48, 961–970.

Hopkin, S.P., 1997. Biology of the Springtails – Insecta:Collembola. Oxford University Press, Oxford.

Howard, R.W., Stanley-Samuelson, D.W., 1996. Fatty acidcomposition of fat body and malpighian tubules of thetenebrionid beetleZophobas atratus: Significance in eicos-anoid-mediated physiology. Comp. Biochem. Physiol. B115,429–437.

Hubert, J., Sustr, V., 2001. The effect of starvation on theˇ

metabolic rate and microanatomy ofGalumna elimata (Aca-ri: Oribatida). Europ. J. Entomol. 98, 265–275.

Joanisse, D.R., Storey, K.B., 1996. Fatty acid content andenzymes of fatty acid metabolism in overwintering cold-hardy gall insects. Physiol. Zool. 69, 1079–1095.

Johnk, J.S., Jones, R.K., 1994. Comparison of whole-cell fattyacid compositions in intraspecific groups ofRhizoctoniasolani AG-1. Phytopathology 84, 271–275.

Joosse, E.N.G., Testerink, G.J., 1977. The role of food in thepopulation dynamics ofOrchesella cincta (Linne) (Collem-bola). Oecologia 29, 189–204.

Karowe, D.N., Martin, M.M., 1989. The effects of quantityand quality of diet nitrogen on the growth, efficiency offood utilisation, nitrogen budget, and metabolic rate of fifth-instarSpodoptera eridania larvae(Lepidoptera: Noctuidae).J. Insect Physiol. 35, 699–708.

Klironomos, J.N., Kendrick, W.B., 1995. Relationships amongmicroarthropods, fungi, and their environment. Plant Soil170, 183–197.

Lavy, D., Verhoef, H.A., 1996. Effects of food quality ongrowth and body composition of the collembolanOrchesellacincta. Physiol. Entomol. 21, 64–70.

Lavy, D., Nedved, O., Verhoef, H.A., 1997. Effects of starva-tion on body composition and cold tolerance in the collem-bolanOrchesella cincta and the isopodPorcellio scaber. J.Insect Physiol. 43, 973–978.

Leonard, M.A., 1984. Observations on the influence of cultureconditions on the fungal preference ofFolsomia candida(Collembola: Isotomidae). Pedobiologia 26, 369–372.

Mikola, J., Setala, H., 1998. No evidence of trophic cascades¨ ¨in an experimental microbial-based soil food web. Ecology79, 153–164.


Navarrete, A., Peacock, A., Macnaughton, S.J., Urmeneta, J.,Mas-Castella, J., White, D.C., Guerrero, R., 2000. Physio-`logical status and community composition of microbial matsof the Ebro delta(Spain) by signature lipid biomarkers.Microb. Ecol. 39, 92–99.

Ogg, C.W., Stanley-Samuelson, D.W., 1992. Phospholipid andtriacylglycerol fatty acid composition of the major life stagesand selected tissues of the tobbaco hornwormManducasexta. Comp. Biochem. Physiol. B101, 345–351.

Parrish, C.C., Wells, J.S., Yang, Z., Dabinett, P., 1998. Growthand lipid composition of scallop juveniles,Placopectenmagellanicus, fed the flagellateIsochrysis galbana withvarying lipid composition and the diatomChaetocerosmuelleri. Marine Biol. 133, 461–471.

Ruess, L., Haggblom, M.M., Garcıa Zapata, E.J., Dighton, J.,¨ ´2002. Fatty acids of fungi and nematodes – possiblebiomarkers in the soil food chain? Soil Biol. Biochem. 34,745–756.

Rusek, J., 1998. Biodiversity of Collembola and their func-tional role in the ecosystem. Biodivers. Conserv. 7,1207–1219.

Sancholle, M., Dalpe, Y., 1993. Taxonomic relevance of fatty´acids of arbuscular mycorrhizal fungi and related species.Mycotaxon 69, 187–198.

Sayah, F., Karlinsky, A., Breuzet, M., 1997. Lipid and fattyacid composition of the fat body during the female repro-ductive cycle ofLabidura riparia (Insecta Dermaptera). J.Comp. Physiol. 167B, 502–507.

Scheu S., Simmerling F., 2004. Growth and reproduction offungal feeding Collembola as affected by fungal species andmixed diets, Oecologia 139, in press.

Shaw, P.J.A., 1988. A consistent hierarchy in the fungal feedingpreferences of the CollembolanOnychiurus armatus. Pedo-biologia 31, 179–187.

Simpson, S.J., Abisgold, J.D., 1985. Compensation by locustsfor changes in dietary nutrients: behavioural mechanisms.Physiol. Entomol. 10, 443–452.

Simpson, S.J., Simpson, C.L., 1990. The mechanisms ofnutritional compensation by phytophagous insects. In: Ber-nays, E.A.(Ed.), Insect–Plant Interactions II. CRC Press,Boca Raton, FL, pp. 112–160.

Stahl, P.D., Klug, M.J., 1996. Characterisation and differenti-ation of filamentous fungi based on fatty acid composition.Appl. Environm. Microbiol. 62, 4136–4146.

Stanley-Samuelson, D.W., Jurenka, R.A., Cripps, C., Blomqu-ist, G.J., de Renobales, M., 1988. Fatty acids in insects:Composition, metabolism, and biological significance. Arch.Insect Biochem. Physiol. 9, 1–33.

Stanley-Samuelson, D.W., Nelson, D.R., 1993. Insect Lipids:Chemistry, Biochemistry and Biology. University of Nebras-ka Press, Lincoln and London.

Stransky, K., Budesinsky, M., Streibl, M., 1986. Lipid com-pounds from the extract of the springtailTetrodontophorabielanesis (Waga). Coll. Czech. Chem. Comm. 51, 948–955.

Testerink, G.J., 1981. Starvation in field population of litter-inhabiting Collembola. Methods for determining foodreserves in small arthropods. Pedobiologia 21, 427–433.

Thompson, A.C., Davis, F.M., Henson, R.D., Gueldner, R.C.,Hedin, P.A., Henderson, C.A., 1973. Lipids and fatty acidsof the Southwestern corn borer,Diatraea grandiosella. J.Insect Physiol. 19, 1817–1823.

Thompson, S.N., 1973. A review and comparative character-isation of the fatty acid composition of seven insect orders.Comp. Biochem. Physiol. 45, 467–485.

Tunlid, A., White, D.C., 1992. Biochemical analysis of bio-mass, community structure, nutritional status, and metabolicactivity of microbial communities in soil. In: Stotzky, G.,Bollag, J.-M. (Eds.), Soil Biochemistry. Marcel Dekker,New York, pp. 229–262.

Vannier, G., Verhoef, H.A., 1978. Effect of starvation ontranspiration and water content in the populations of twoco-existing collembola species. Comp. Biochem. Physiol.A60, 483–489.

Verhoef, H.A., 1981. Water balance in Collembola and itsrelation to habitat selection: water content, haemolymphosmotic pressure and transpiration during an instar. J. InsectPhysiol. 27, 755–760.

Verhoef, H.A., Li, K.W., 1983. Physiological adaptions to theeffects of dry summer periods in Collembola. In: Lebrun,P., Andre, H.M., De Mets, A., Gregoire-Wibo, C., Wauthy´ ´G. New Trends in Soil Biology–Proceedings of the VIIIthInternational Soil Zoology Colloquium 83. Dien-Brichart,Ottignies-Louvain-la-Neuve, Belgium, pp. 345–356.

Verhoef, H.A., Prast, J.E., Verweij, R.A., 1988. Relativeimportance of fungi and algae in the diet and nitrogennutrition of Orchesella cincta (L.) and Tomocerus minor(Lubbock) (Collembola). Funct. Ecol. 2, 195–201.

Visser, S., 1985. Role of soil invertebrates in determining thecomposition of soil microbial communities. In: Fitter, A.H.,Atkinson, D., Read, D.J., Usher, M.B.(Eds.), EcologicalInteractions in Soil. Blackwell Scientific Publishers, Oxford,pp. 297–317.

Visser, S., Whittaker, J.B., 1987. Feeding preferences forcertain litter fungi byOnychiurus subtenuis (Collembola).Oikos 28, 320–325.

Walsh, M.I., Bolger, T., 1990. Effects of diet on the growthand reproduction of some Collembola in laboratory cultures.Pedobiologia 334, 161–171.

Weete, J.D., 1980. Lipid biochemistry of fungi and otherorganisms. Plenum Press, New York.

Zelles, L., 1999. Fatty acid patterns of phospholipids andlipopolysaccharides in the characterisation of microbial com-munities in soil: a review. Biol. Fertil. Soils 29, 111–129.

Effects of fungal food quality and starvation on the fatty acid composition of Protaphorura fimata...

Documents

Transcript of Effects of fungal food quality and starvation on the fatty acid composition of Protaphorura fimata...