Science of the Total Environment 407 (2009) 4283–4288

Contents lists available at ScienceDirect

Science of the Total Environment

j ourna l homepage: www.e lsev ie r.com/ locate /sc i totenv

Mixtures of zinc and phosphate affect leaf litter decomposition by aquaticfungi in streams

Isabel Fernandes, Sofia Duarte, Fernanda Cássio, Cláudia Pascoal ⁎Centre of Molecular and Environmental Biology (CBMA), Department of Biology, University of Minho, Campus de Gualtar, 4710-057 Braga, Portugal

⁎ Corresponding author. Tel.: +351 253604045; fax: +E-mail address: cpascoal@bio.uminho.pt (C. Pascoal)

0048-9697/$ – see front matter © 2009 Elsevier B.V. Adoi:10.1016/j.scitotenv.2009.04.007

a b s t r a c t

a r t i c l e i n f o

Article history:Received 3 January 2009Received in revised form 26 March 2009Accepted 7 April 2009Available online 2 May 2009

Keywords:StreamsAquatic fungiLeaf decompositionZinc and phosphate mixtures

To better understand the impacts of multiple stressors in freshwaters, we investigated the effects of mixturesof zinc and inorganic phosphorus on microbial decomposition of leaf litter. Alder leaves were colonized in astream and placed in microcosms with stream water supplemented or not with 3 concentrations of zinc (Znup to 9.8 mg/l) or phosphate (P–PO4

3− up to 0.5 mg/l), alone and in all possible combinations. We measuredleaf mass loss, and fungal biomass, reproduction and diversity. In control microcosms, 23 species of aquatichyphomycetes were identified on leaves, and the exposure to the highest zinc concentration reduceddiversity to 14 species. Articulospora tetracladia was the dominant species followed by Flagellospora sp. andAlatospora acuminata. The exposure to phosphate increased the contribution of A. acuminata, but this specieswas negatively affected by zinc. Under high zinc stress, Varicosporium elodeae increased its contribution tothe total conidial production.The exposure to high zinc concentration, alone or in mixtures with phosphate, led to shifts in fungalcommunity structure, as indicated by cluster analysis based on sporulation data and denaturing gradient gelelectrophoresis (DGGE) fingerprints of fungal DNA. These changes were accompanied by a reduction in leafdecomposition, particularly in mixtures with high Zn concentration, in which leaf mass loss was 30% lowerthan in the control. This suggests that the co-occurrence of zinc and phosphate may have negative effects onstream ecosystem functioning. However, we did not detect decreased leaf-associated fungal biomass andsporulation, probably because a delay in fungal colonization occurred due to the presence of stressors.

© 2009 Elsevier B.V. All rights reserved.

1. Introduction

Freshwater pollution due to anthropogenic activities has beenrising during the last century but its consequences to ecosystemfunctioning are not yet fully understood (Covich et al., 2004).Urbanization and intensive agriculture can increase the levels ofnitrogen and phosphorus in streams leading to eutrophication withimpacts to freshwater ecosystems (Allan, 1995). This can be furthercomplicated by the presence of metals that can reach streams as by-products of agricultural fertilizers, from mine drainage, industrialemissions and garbage disposal (Rand et al., 1995).

Aquatic hyphomycetes are a polyphyletic group of fungi thatdominate in earlier stages of plant-litter decomposition in streams(Bärlocher, 2005). These fungi produce a variety of extracellularenzymes that attack the structural polysaccharides of plant-litter cellwalls, improving leaf palatability to invertebrate shredders andreleasing fine particulate matter consumed by collectors (Suberkropp,1998). This is accompanied by the release of inorganic and organicnutrients used by autotrophic and other heterotrophic organisms

351 253678980..

ll rights reserved.

(Bärlocher, 2005). Although aquatic hyphomycetes are mainlydocumented in clean streams, they have also been found in streamsaffected by metals (Sridhar et al., 2001; Niyogi et al., 2002; Krausset al., 2005) or mixtures of nutrients and metals (Pascoal and Cássio,2004; Pascoal et al., 2005a). This increases the interest in betterunderstanding the contribution of fungi to plant-litter decompositionin streams impacted by those stressors.

Several studies have shown that an increase in nutrient concentra-tions in the stream water is associated with an increase in fungaldiversity (Gulis and Suberkropp, 2004) and activity, measured asbiomass build-up and spore production (Sridhar and Bärlocher, 2000;Gulis and Suberkropp, 2003a,b; Pascoal and Cássio, 2004), leading tofaster leaf decomposition (Sridhar and Bärlocher, 2000; Grattan andSuberkropp, 2001; Pascoal et al., 2001, 2003; Ardón and Pringle, 2007).In contrast, other studies have found no effect of nutrient enrichment onthe biota associated with leaf litter (Triska and Sedell, 1976; Royer andMinshall, 2001). There is evidence that a Michaelis–Mentenmodel bestdescribes the effects of dissolved inorganic nutrients on microbialactivity and plant-litter decomposition, with low half saturationconstants (Km: 7–21 µg/l SRP, Rosemond et al., 2002; Gulis et al.,2006; Km: 16–52 µg/l N–NO3, Ferreira et al., 2006). On the other hand, areduction in fungal diversity and activity has been found in streamshighly polluted by heavymetals (Sridhar et al., 2001; Niyogi et al., 2001,

4284 I. Fernandes et al. / Science of the Total Environment 407 (2009) 4283–4288

2002). However, the threshold of fungal responses is expected to varywith metal identity and concentration, and the measured fungalparameter (e.g., biomass, reproduction and diversity). Even thoughthere is a considerable amount of information about the effects ofnutrient loading or metals on leaf litter decomposition and associatedcommunities in streams, studies addressing the impacts of the co-occurrence of these two stressors, under controlled conditions, are lessfrequent and presumably absent for aquatic microbial decomposers.

In this study, we examined the effects of zinc and phosphate, aloneand in combination, on decomposition of alder leaves by aquatic fungi.Leaves were immersed in a stream to allow microbial colonization,and placed in microcosms with stream water supplemented or notwith 3 concentrations of zinc and/or phosphate to test their combinedeffects on leaf mass loss and leaf-associated fungal biomass,reproduction and diversity. Previous studies indicate that highconcentrations of zinc alter the structure of fungal community anddecrease leaf decomposition (Duarte et al., 2004), while increasingphosphate concentrations tend to enhance (Sridhar and Bärlocher,2000; Gulis and Suberkropp, 2003a) or have no effect (Rosemondet al., 2002) on leaf-associated fungal activity. We hypothesized thathigh levels of zinc may reduce fungal diversity and/or activity, and thepresence of phosphate would modulate metal effects on microbialdecomposition of leaf litter.

2. Methods

2.1. Sampling site and microcosm experiment

The study site is a reference site located at the spring of the EsteRiver in the Northwest of Portugal (N: 41°34′46 and W: 8°19′60;461 m altitude). At the spring, the Este River is about 30 cm deep and50 cmwide. The stream bottom consists of granitic rocks, pebbles andgravel, and the riparian vegetation is mainly constituted by Eucalyptusglobulus Labill., Pinus pinaster Aiton, Pteridium aquilinum Khun andJuncus sp.

Temperature, pH, conductivity and dissolved oxygen concentrationin the stream water was measured in situ (Multiline F/set 3 no.400327,WTW) andwater samples were collected for quantification ofnitrate (HACH kit, program 355), nitrite (HACH kit, program 371),ammonia (HACHkit, program385) and phosphate (HACHkit, program480). Data from physical and chemical analysis of the stream waterindicated circumneutral pH, lowconductivity, moderate concentrationof nitrate and low concentration of other inorganic nutrients (Table 1),consistent with that generally found at this site (Duarte et al., 2004;Pascoal et al., 2005a).

Air dried alder leaves (Alnus glutinosaGaertn.), collected in autumn2005, were leached for 1 h in deionised water and cut into 22 mm-disks. Sets of 40 disks were placed into 54 fine-mesh bags (16×20 cm,0.5-mm mesh size to minimize the entrance of invertebrates). On 3rdMarch 2006, leaf bags were immersed at the study site to allowmicrobial colonization. On the same day, 3 bags were transported tothe laboratory to estimate the initial mass of the leaves. After 14 days,the remaining bags were retrieved and transported to the laboratory,

Table 1Physical and chemical parameters of the streamwater at the source of the Este River on13rd March 2006, collected for microcosm experiments.

Parameter Value

Temperature (°C)a 10.6pHa 6.7Conductivity (µS/cm)a 46O2 dissolved (mg/l)a 9.4N–NO3

− (mg/l) 0.15N–NO2

− (mg/l) b0.1N–NH3 (mg/l) b0.01P–PO4

3− (mg/l) 0.01

a Measurements were done in situ at 10 a.m.

where leaf disks from each bag were rinsed with deionised water andplaced into 250 ml Erlenmeyer flasks with 150 ml of sterilized streamwater (120 °C, 20 min). A set of 3 flasks was used to determine leafmass loss and fungal parameters at the beginning of the microcosmstudy. Themicrocosmswere supplementedwith ZnCl2 (Sigma) at finalZn concentrations of 0.03 mg/l (Zn1), 0.98 mg/l (Zn2) and 9.8 mg/l(Zn3), and KH2PO4 (Merck) at final P–PO4

3− concentrations of 0.05mg/l(P1), 0.2 mg/l (P2) and 0.5 mg/l (P3), alone or in all possiblecombinations (3 replicates). The microcosms were incubated on ashaker (Certomat BS 3, B. Braun, Biotech International, Germany) at120 rpm,18 °C, during 35 days and solutionswere changed every 7 days.At the end of the experiment, 30 leaf disks of each replicatewere used toestimate leaf mass loss and the remaining disks were freeze-dried(Christ alpha 2–4, B. Braun, Germany) for fungal biomass quantificationand analysis of fungal community by denaturing gradient gel electro-phoresis (DGGE).

2.2. Sporulation rates

Appropriate aliquots of conidial suspensions released from leafdisks in each microcosm after 7 and 21 days of exposure to treatmentswere filtered (0.45-µm pore size, Millipore) and stained with 0.05%cotton blue in lactic acid. At least 300 conidia per filter were identifiedand counted under a microscope to determine the contribution ofeach aquatic hyphomycete species to the total conidial production.

2.3. Fungal DNA extraction, amplification and separation by denaturinggradient gel electrophoresis

DNAwas extracted from 3 leaf disks (2 half disks of each replicate,pooled from 3 replicates) or from 2 plugs (5 mm diameter) of purecultures on solid medium of 5 aquatic hyphomycete species(Articulospora tetracladia Ingold UMB-22.01, Varicosporium elodeaeW. Kegel UMB-20.01, Alatospora acuminata Ingold UMB-173.01, Tri-cladium chaetocladium Ingold UMB-163.01 and Heliscus lugdunensisSacc. and Therry UMB-159.01), using the UltraClean Soil DNA kit(MoBio Laboratories, Solana Beach, CA, USA). The primer pair ITS3GCand ITS4 (White et al., 1990) was used to amplify the ITS2 region offungal genomic rDNA (Nikolcheva and Bärlocher, 2005). The primerITS3GC had a sequence identical to ITS3 with a 40-bp GC tail on the 5′end (Muyzer et al., 1993). For PCR reactions, 4 µM of each primer,3 mM of MgCl2, 1.5 U of Taq DNA polymerase, 2 mM of dNTPs, 1x Taqbuffer [KCl:(NH4)2SO4] and 1 µl of DNA were mixed in a final volumeof 50 µl. The amplification programme started with a denaturation for2 min at 95 °C, followed by 36 cycles of denaturation for 30 s at 95 °C,primer annealing for 30 s at 55 °C and extension for 1 min at 72 °C,concluding with an extension for 5 min at 72 °C (Duarte et al., 2008a)(iCycler Thermal Cycler, BioRad Laboratories, Hercules, CA, USA). ThePCR products were analysed using the DCode™ Universal MutationDetection System (BioRad Laboratories, Hercules, CA, USA). Samplesof 20–45 µl from the amplification products of 380–400 bp wereloaded on 8% (w/v) polyacrylamide gels in 1x Tris–Acetate–EDTA(TAE) with a denaturing gradient from 30% to 70% (100% denaturantcorresponds to 40% formamide and 7 M urea). In each DGGE gel, aDNA mixture of the 5 aquatic hyphomycete species indicated abovewas used to calibrate the gels. The gels were run at 55 V for 16 h at56 °C and stained with ethidium bromide (1 µg/ml). The gel imageswere captured under UV light with an Eagle eye II (Stratagene, La Jolla,CA, USA).

2.4. Fungal biomass

Fungal biomass on decomposing leaf disks was estimated fromergosterol concentration (Gessner, 2005). Lipids were extracted fromsets of 8 leaf disks per replicate by heating (80 °C, 30 min) in KOH inmethanol (8 g/l), purified by solid-phase extraction and eluted in

4285I. Fernandes et al. / Science of the Total Environment 407 (2009) 4283–4288

isopropanol. Ergosterol was quantified by high performance liquidchromatography (HPLC) (Beckmann Gold System) using a LiChro-spher RP18 column (250×4 mm, Merck). The system was runisocratically with HPLC-grade methanol at 1.4 ml/min and 33 °C.Ergosterol peaks were detected at 282 nm and a series of ergosterolstandards in isopropanol (Sigma) was used to estimate ergosterolconcentration in the samples.

2.5. Leaf mass loss

Leaf disks from each replicate of each treatmentwere dried at 50 °Cto constantmass (72 h±24 h) andweighed to the nearest 0.001mg todetermine leaf mass loss.

2.6. Statistical analyses

A two-way ANOVA was used to test the effects of zinc andphosphate concentrations on leaf mass loss, fungal biomass, sporula-tion and diversity, and on the contribution of aquatic hyphomycetespecies to the total conidial production (Zar, 1996). Differencesbetween control and treatments were analysed by a Dunnett's post-test. To achieve normal distribution and homoscedasticity, fungalbiomass, sporulation and diversity were ln-transformed, and thepercentages of leaf mass loss and of contribution of aquatichyphomycete species to the total conidial production were arcsinesquare root transformed (Zar, 1996).

Cluster analyses of fungal community structure, assessed fromDGGE fingerprints and aquatic hyphomycete-sporulating species,were done by unweighted pair-group method average (UPGMA)using the Pearson correlation coefficient (Rademaker and de Bruijn,2004).

Univariate analyses were done with the Statistica 6.0 for Windows(Statsoft, Inc) and DGGE fingerprints were analysed with theGelCompar II program (Applied Maths, Belgium).

Fig. 1. Leaf mass loss (a), and fungal biomass (b) and sporulation (c) on decomposingalder leaves exposed to zinc and phosphate alone or in mixtures. Measurements weredone at the end of the experiment (35 days in microcosms), except for sporulation thatwas measured after 7 and 21 days in microcosms. Treatments: Zn1, 0.03 mg/l; Zn2,0.98 mg/l; Zn3, 9.8 mg/l; P1, 0.05 mg/l; P2, 0.2 mg/l; P3, 0.5 mg/l. M+SEM; n=3;⁎Dunnett's test based on comparison to control, Pb0.05.

3. Results

3.1. Effects of zinc and phosphate on fungal activity

In the absence of any stressor, mass loss of alder leaves was 78%(Fig. 1a), of which 27%was lost during the 14 days of leaf immersion atthe spring of the Este River (data not shown) and 51% was lost during35 days in microcosms. The exposure to zinc led to a significantdecrease in leaf mass loss (two-way ANOVA, Pb0.05 for Zn, Table 2,Fig. 1a). The strongest inhibition effects were observed in Zn1P3,Zn2P2, Zn3, Zn3P1 and Zn3P2 treatments (Dunnett's test, Pb0.05),where leaf mass loss was nearly 30% lower relatively to control(Fig. 1a).

Fungal biomass on decomposing leaves was 142 µg ergosterol g−1

leaf dry mass after stream colonization (data not shown). Fungalbiomass increased significantly with exposure to zinc and phosphateafter 35 days of incubation in microcosms (two-way ANOVA, Pb0.05,Table 2, Fig. 1b). The most significant effects were observed for Zn2P2,Zn2P3, Zn3, Zn3P1 and Zn3P3 treatments (Dunnett's test, Pb0.05), inwhich an increase of almost 2-times was observed.

Sporulation rates of aquatic hyphomycetes on leaves weresignificantly affected by zinc and phosphate after 21 days of exposure,and by zinc and phosphate interactions after 7 days (two-way ANOVA,Pb0.05, Table 2, Fig. 1c). At this time (7 days), sporulation rates weresignificantly inhibited by exposure to P1 and Zn1P3 treatments(Dunnett's test, Pb0.05), while after 21 days a 3- to 4-times increasewas observed mainly when zinc and phosphate were added together(Zn1P1, Zn1P3, Zn2P2, Zn2P3, Zn3P1, Zn3P2 and Zn3P3 treatments;Dunnett's test, Pb0.05, Fig. 1c).

3.2. Effects of zinc and phosphate on the structure of fungal community

From conidial morphology, 32 species of aquatic hyphomyceteswere identified on decomposing alder leaves (Table 3). Articulosporatetracladia was the dominant species, contributing from 35% to 75%to the total conidial production, followed by Flagellospora sp. andAlatospora acuminata. The exposure to zinc and phosphate signifi-cantly affected the contribution of A. acuminata (two-way ANOVA,P=0.04 and Pb0.01, respectively) with an increase in P1 and Zn1P1treatments (Dunnett's test, Pb0.01 and P=0.04, respectively) and adecrease in treatments with the highest zinc concentration (Zn3,Zn3P2 and Zn3P3; Dunnett's test, P=0.01). The exposure to zincsignificantly increased the contribution of V. elodeae, particularly inZn3P3 treatment (two-way ANOVA, Pb0.01; Dunnett's test, P=0.02).

Table 2Two-way ANOVAs on the effects of zinc (Zn) and phosphate (P–PO4

3−) concentrationson leaf mass loss, fungal biomass and sporulation rate.

Parameter Effect df SS MS F P

Leaf mass loss Zn 3 1.27 0.42 3.90 0.018P 3 0.36 0.12 1.11 0.359Zn⁎P 9 1.68 0.19 1.71 0.126Error 32 3.48 0.11

Fungal biomass Zn 3 0.95 0.32 7.83 b0.001P 3 0.58 0.20 4.81 0.007Zn⁎P 9 0.47 0.05 1.30 0.276Error 32 1.30 0.04

Sporulation rate 7 days Zn 3 0.95 0.32 1.71 0.184P 3 0.98 0.33 1.77 0.172Zn⁎P 9 4.33 0.48 2.61 0.022Error 32 5.91 0.19

Sporulation rate 21 days Zn 3 4.12 1.37 9.30 b0.001P 3 2.57 0.86 5.81 0.003Zn⁎P 9 1.23 0.14 0.93 0.517Error 32 4.73 0.15

Data were obtained at the end of the experiment (35 days), except for sporulationwhich was measured after 7 and 21 days.

4286 I. Fernandes et al. / Science of the Total Environment 407 (2009) 4283–4288

The number of aquatic hyphomycete species did not appear tochange after zinc and phosphate exposure, except for the highest zincconcentration, where only 14 fungal species were found compared to23 species in the control (Table 3). The DGGE fingerprint analysis offungal DNA on decomposing leaves showed a total of 20 bands in theabsence of stressors at the end of the experiment (Fig. 2). Theexposure to zinc and phosphate did not seem to affect fungal diversityas assessed by DGGE (15–22 bands) (Fig. 2). DGGE bands matching

Table 3Mean percentage contribution of individual fungal species on decomposing leaves to overamixtures.

Taxon Treatment

Control P1 P2 P3 Zn1 Z

Alatospora acuminata Ingold 7.7 22.0 15.8 15.8 13.1 1Alatospora pulchella Marvanová 3.8 3.1 4.2 2.7 2.5 2Anguillospora filiformis Greath 0.9 0.4 0.1 0.2 0.2 0Articulospora tetracladia Ingold 59.6 47.8 51.6 51.6 49.9 4Clavatospora longibrachiata (Ingold) Sv Nilsson exMarvanová and Sv Nilsson

– – 0.4 3.1 0.1 0

Cylindrocarpon sp. – – b0.1 – b0.1 –

Dimorphospora foliicola Tubaki 0.1 – – – – –

Diplocladiella scalaroides G Arnaud ex MB Ellis – – – – – –

Flagellospora curta J Webster 0.1 – 0.1 b0.1 b0.1 0Flagellospora curvula Ingold b0.1 0.1 0.2 0.1 0.1 0Flagellospora sp. (60 / 2 µm) 20.8 21.3 15.5 17.0 28.8 2Heliscella stellata (Ingold and JV Cox) Marvanová 0.1 – 1.0 2.6 – 0Heliscus lugdunensis Sacc and Therry 0.2 0.2 2.3 0.5 0.7 0Lemonniera aquatica De Wild 0.5 0.5 0.1 0.4 b0.1 0Lunulospora curvula Ingold 0.4 0.4 0.3 0.3 0.3 0Pleuropedium multiseptatum Marvanová and Descals b0.1 – – – – –

Tetracladium breve A Roldán 0.1 0.1 b0.1 b0.1 b0.1 0Tetracladium furcatum Descals b0.1 0.1 0.2 0.1 0.1 0Tetracladium maxilliforme (Rostr) Ingold – 0.1 – – – –

Tetracladium setigerum (Grove) Ingold 0.1 0.4 0.7 0.5 0.4 0Tetracladium sp. b0.1 b0.1 b0.1 – – –

Tricellula aurantiaca (Haskins) Arx – – – – – –

Tricellula sp. 0.1 – – – – –

Tricladiopsis flagelliformis Descals – 0.1 – – – –

Tricladium chaetocladium Ingold b0.1 – 0.2 0.1 0.1 b

Tricladium sp. – – – – b0.1 –

Tricladium splendens Ingold 0.4 0.5 0.5 0.6 0.5 0Tricladium terrestre D Park – – b0.1 b0.1 – –

Tripospermum sp. b0.1 b0.1 b0.1 – – –

Triscelophorus cf. acuminatus Nawawi 0.1 0.6 1.0 1.0 b0.1 0Varicosporium elodeae W Kegel 4.8 2.3 5.7 3.5 3.1 4Ypsilina graminea (Ingold, PJ McDougall and Dann)Descals, J Webster and Marvanová

– b0.1 0.1 b0.1 – –

Species richness 23 20 24 21 20 1

Treatments: Zn1, 0.03 mg/l; Zn2, 0.98 mg/l; Zn3, 9.8 mg/l; P1, 0.05 mg/l; P2, 0.2 mg/l; P3,

with those produced by pure cultures of A. tetracladia (the speciesthat most contributed to the total conidial production) and V. elodeaewere present in all treatments (Fig. 2). However, no bands matchedwith that produced by the pure culture of A. acuminata. Clusteranalyses based on DGGE fingerprints (Fig. 3a) and aquatic hyphomy-cete-sporulating species (Fig. 3b) indicated that communities exposedto higher Zn concentrations (Zn2 and Zn3), alone or in mixtures withphosphate, grouped together and were separated from the remainingtreatments.

4. Discussion

Many aquatic environments are impacted by a set of stressors thatare threatening biodiversity with probable negative consequences forecological processes (Dudgeon et al., 2006). Metals and excessivenutrient input into streams often occur simultaneously, but theircombined effects on aquatic biota are complex and have rarely beenexamined (but see Wang and Dei, 2001).

In this study,we assessed the effects of zinc andphosphate, alone orin combination, on fungal decomposer communities naturally occur-ring in streams by manipulating environmentally realistic concentra-tions of those chemicals (see Pascoal and Cássio, 2004; Pascoal et al.,2003, 2005a). A total of 32 species of aquatic hyphomycetes werefound representing the high fungal diversity commonly associatedwith alder leaves in streams of the Northwest Portugal (Pascoal et al.,2003, 2005a). Zinc appeared to lower diversity, particularly at thehighest metal concentration, in which only 14 fungal species wereidentified sporulating on leaves. The presence of phosphate in themixtures tended to attenuate the negative effects of zinc, although

ll conidia produced after 7 and 21 days of exposure to zinc and phosphate alone or in

n1P1 Zn1P2 Zn1P3 Zn2 Zn2P1 Zn2P2 Zn2P3 Zn3 Zn3P1 Zn3P2 Zn3P3

9.3 8.4 16.1 11.9 9.2 11.0 11.6 0.6 2.1 0.6 0.5.6 1.6 7.9 1.9 1.3 2.0 2.4 0.1 0.2 0.1 0.2.2 0.5 0.5 0.4 0.5 0.1 0.3 – b0.1 – 0.14.8 40.6 35.3 51.5 58.5 60.7 51.9 69.4 58.6 62.0 67.0.1 7.0 2.5 b0.1 – b0.1 0.8 – – b0.1 b0.1

0.1 0.1 – – b0.1 – – – – b0.1– – – – – – – – – –

– – – – – b0.1 – – – –

.1 0.1 b0.1 b0.1 0.1 0.2 b0.1 b0.1 b0.1 b0.1 0.1

.1 1.3 0.6 0.1 – – 0.1 – – – –

5.8 25.2 26.6 26.6 23.3 19.6 22.8 16.8 26.6 26.0 15.0.1 5.7 2.4 b0.1 – 0.3 1.9 – – – –

.5 0.4 0.4 0.6 0.3 0.6 1.2 0.6 1.1 0.6 0.5

.2 b0.1 0.1 0.1 0.1 – 0.2 b0.1 0.1 0.3 0.2

.5 0.3 0.1 0.1 0.2 0.1 0.5 0.1 0.3 0.5 0.3– – – – – – – – – –

.1 0.2 0.1 b0.1 0.1 0.1 0.1 0.1 0.1 0.1 b0.1

.1 b0.1 b0.1 – b0.1 – b0.1 – – – b0.1– – – – – – – – – –

.2 0.3 0.4 0.6 0.4 0.5 0.4 0.4 0.4 0.3 0.3– b0.1 b0.1 0.1 b0.1 – b0.1 – 0.1 0.1– – – – – – – b0.1 – –

– – – – – – – – – –

– – – – – – – – – –

0.1 0.1 b0.1 0.1 0.1 b0.1 b0.1 – b0.1 b0.1 b0.1– – – – – – – – – –

.4 0.5 0.4 0.5 0.2 0.4 0.5 0.1 0.3 0.4 0.6b0.1 b0.1 – – – – – – – –

– – – – – – – – 0.1 –

.6 3.3 3.8 0.5 0.1 0.6 0.5 – b0.1 0.1 –

.5 4.4 2.5 5.1 5.4 3.7 4.8 11.8 10.1 8.6 14.9b0.1 – – – b0.1 b0.1 b0.1 b0.1 – 0.1

9 22 22 19 17 19 21 14 17 17 19

0.5 mg/l.

Fig. 2. DGGE fingerprints of the ITS2 region of rDNA of fungal communities on decomposing alder leaves exposed to zinc and phosphate, alone or inmixtures, before and after 35 daysin microcosms. M, mixture of DNA from pure cultures: AT, Articulospora tetracladia; VE, Varicosporium elodeae; AA, Alatospora acuminata; TC, Tricladium chaetocladium; HL, Heliscuslugdunensis. Lines: 1 and 10, Control; 2, P1; 3, P2; 4, P3; 5, Zn1; 6, Zn1P1; 7, Zn1P2; 8, Zn1P3; 9, before exposure to treatments; 11, Zn2; 12, Zn2P1; 13, Zn2P2; 14, Zn2P3; 15, Zn3; 16,Zn3P1; 17, Zn3P2; 18, Zn3P3. Treatments: Zn1, 0.03 mg/l; Zn2, 0.98 mg/l; Zn3, 9.8 mg/l; P1, 0.05 mg/l; P2, 0.2 mg/l; P3, 0.5 mg/l.

4287I. Fernandes et al. / Science of the Total Environment 407 (2009) 4283–4288

fungal diversity did not attain that of the control (23 species). Thisagrees with field observations reporting that aquatic hyphomycetediversity is low under metal stress (Sridhar et al., 2001; Niyogi et al.,2002; Sridhar et al., 2005) but high in nutrient-enriched streams(inorganic N and P, Gulis and Suberkropp, 2004). However, in ourwork, changes in fungal diversitywere less pronouncedwhen assessedby DNA fingerprinting than by spore identification, suggesting thatsome species remain in leaves but have their reproduction compro-mised (Nikolcheva and Bärlocher 2005; Duarte et al., 2008a).

In our study, the exposure to zinc and/or phosphate also led toshifts in the structure of fungal communities. Phosphate increased thecontribution of A. acuminata to the total conidial production, but thisspecies was negatively affected by elevated zinc concentrations. Underhigh zinc stress, V. elodeae increased its contribution to the totalreleased conidia. These changes in the community composition maybe apparent since they relied on the ability of fungal species toproduce spores, which may vary with species identity (Duarte et al.,2006) and several environmental factors, including nutrient avail-ability (Gulis and Suberkropp, 2003b, 2004) and metal concentration

Fig. 3. Dendograms from DGGE fingerprints of fungal DNA (a) and from aquatichyphomycete-sporulating species (b) on decomposing alder leaves exposed to zinc andphosphate, alone or inmixtures. Clusters were constructed fromUPGMA analyses basedon Pearson coefficient. DNA fingerprinting was done at the end of the experiment(35 days in microcosms), while sporulation data were obtained after 21 days inmicrocosms. Treatments: Zn1, 0.03 mg/l; Zn2, 0.98 mg/l; Zn3, 9.8 mg/l; P1, 0.05 mg/l;P2, 0.2 mg/l; P3, 0.5 mg/l.

in the stream water (Sridhar et al., 2001; Duarte et al., 2004, 2008a).However, fungal communities exposed to higher zinc concentrations,alone or in phosphate mixtures, separated from those exposed to theremaining treatments, as indicated by cluster analysis based onsporulation data and DNA fingerprinting. These findings corroborateprevious observations indicating that water chemistry is a majorfactor structuring fungal communities in streams (Pascoal et al.,2005b; Duarte et al., 2008a).

Shifts in fungal community composition induced by environmentalfactors probably reflect changes towards a better-adapted community,whose performances may differ from the initial community. Fungalactivity and leaf decomposition is reported to be stimulated byphosphate in stream and microcosm experiments, at least whennitrogen is not limiting (Sridhar and Bärlocher, 2000; Grattan andSuberkropp, 2001; Gulis and Suberkropp, 2003a; Pascoal et al., 2003).However, it was not the case of this study, in which phosphateconcentrations (P–PO4

3−) ranged from 0.01 mg/l (control) to 0.5 mg/land nitrate did not appear to be limiting (0.15 mg/l N–NO3

−). Somestudies indicate that the response of leaf decomposition to nutrientenrichment tends to be smaller in plant litter with high quality (Guliset al., 2004). Leaves of Alnus glutinosa are reported to have 2- to 4-times higher N and P contents than those of Eucalyptus globulus orQuercus robur (Sampaio et al., 2001). Therefore, the high nutritivevalue of A. glutinosa leaves (Sampaio et al., 2001) might havecontributed to attenuate the putative effects of phosphate on leafdecomposition in the present study.

Conversely, the exposure of fungal communities to zinc led to adecrease in leaf mass loss, particularly in mixtures with phosphate,suggesting that this nutrient can modulate the impacts of zinc on leafdecomposition. If this also happens in natural ecosystems, mixtures ofnutrients and metals are likely to have complex effects on streamfunctions.

Previous studies show that fungal biomass can be maintained atmetal concentrations (Cu and/or Zn) able to reduce leaf decomposi-tion (Duarte et al., 2004, 2008a). Similarly, we found that leaf massloss was reduced in microcosms exposed to zinc, alone or combinedwith phosphate, but high fungal biomass and sporulationwere found.This suggests that at least some populations of aquatic fungi are ableto tolerate high zinc stress (Krauss et al., 2005; Guimarães-Soareset al., 2006; Azevedo et al., 2007) and/or that a well-establishedfungal community may be less sensitive to the impacts of stressors(Sridhar et al., 2005; Duarte et al., 2008b). Although we did not track

4288 I. Fernandes et al. / Science of the Total Environment 407 (2009) 4283–4288

bacterial activity during this study, bacteria may compete with fungifor resources (Mille-Lindblom et al., 2006; Romani et al., 2006) andhave been consistently found to bemore sensitive tometals than fungi(Rajapaksha et al., 2004; Duarte et al., 2008a). Consequently, it isconceivable that decreased competition between fungi and bacteriaoccur under metal stress that together with the carbon released fromdead bacteria, might have triggered increased fungal growth, over-riding the expected negative impact of zinc, as suggested byRajapaksha et al. (2004). On the other hand, water chemistry isreported to affect the dynamic of fungal colonization by changing notonly the magnitude of spore or biomass production but also dates ofmaximum values (Suberkropp and Chauvet, 1995; Pascoal and Cássio,2004; Pascoal et al., 2005a). If a delay in fungal colonization hadoccurred under stressful conditions, it may help to explain the highbiomass and sporulation rates observed in stressed communities inthis experiment. We encourage the monitoring of microbial coloniza-tion dynamic of leaf litter in further studies to clarify the impacts ofthese stressors on leaf decomposition in streams.

Overall our results showed that high concentrations of zinc, aloneor in mixtures with phosphate, led to changes in the structure offungal communities on decomposing leaves assessed as DNAfingerprinting or spore production. These effects were accompaniedby a reduction in leaf decomposition, suggesting that the co-occurrence of these stressors may have negative effects on streamecosystem functioning. Since conclusions were based on responses ofa single fungal community, further research using communities withdifferent backgrounds may help to better understand the magnitudeof zinc and phosphate impacts on streams.

Acknowledgment

The Portuguese Foundation for the Science and the Technologysupported S. Duarte (SFRH/BD/13482/2003).

References

Allan JD. Stream ecology: Structure and function of running waters. London, UK:Chapman and Hall; 1995.

Ardón M, Pringle CM. The quality of organic matter mediates the response ofheterotrophic biolfilms to phosphorus enrichment of the water column andsubstratum. Freshw Biol 2007;52:1762–72.

Azevedo M-M, Carvalho A, Pascoal C, Rodrigues F, Cássio F. Responses of antioxidantdefenses to Cu and Zn stress in two aquatic fungi. Sci Total Environ 2007;377:233–43.

Bärlocher F. Freshwater fungal communities. In: Dighton J, Oudemans P,White J, editors.The fungal community. third ed. Boca Raton, Florida: CRC Press; 2005.

Covich AP, Austen MC, Bärlocher F, Chauvet E, Cardinale BJ, Biles CL, et al. The role ofbiodiversity in the functioning of freshwater and marine benthic ecosystems.BioScience 2004;54:767–75.

Duarte S, Pascoal C, Cássio F. Effects of zinc on leaf decomposition by fungi in streams:studies in microcosms. Microb Ecol 2004;48:366–74.

Duarte S, Pascoal C, Cássio F. Aquatic hyphomycete diversity and identity affect leaf litterdecomposition in microcosms. Oecologia 2006;147:658–66.

Duarte S, Pascoal C, Alves A, Correia A, Cássio F. Copper and zinc mixtures induce shiftsin microbial communities and reduce leaf litter decomposition in streams. FreshwBiol 2008a;53:91-101.

Duarte S, Pascoal C, Cássio F. High diversity of fungi maymitigate the impact of pollutionon plant litter decomposition in streams. Microb Ecol 2008b;56:688–95.

Dudgeon D, Arthington AH, Gessner MO, Kawabata Z-I, Knowler DJ, Lévêque C, et al.Freshwater biodiversity: importance, threats, status and conservation challenges.Biol Rev 2006;81:163–82.

Ferreira V, Gulis V, Graça MAS. Whole-stream nitrate addition affects litter decom-position and associated fungi but not invertebrates. Oecologia 2006;149:718–29.

Gessner MO. Ergosterol as a measure of fungal biomass. In: Graça MAS, Bärlocher F,Gessner MO, editors. Methods to study litter decomposition: a practical guide.Dordrecht, Netherlands: Springer; 2005. p. 189–96.

Grattan RM, Suberkropp K. Effects of nutrient enrichment on yellow poplar leafdecomposition and fungal activity in streams. J N Am Benthol Soc 2001;20:33–43.

Guimarães-Soares L, Felícia H, Bebianno MJ, Cássio F. Metal-binding proteins andpeptides in aquatic fungi exposed to severe metal stress. Sci Total Environ2006;372:148–56.

Gulis V, Suberkropp K. Effect of inorganic nutrients on relative contributions of fungiand bacteria to carbon flow from submerged decomposing leaf litter. Microb Ecol2003a;45:11–9.

Gulis V, Suberkropp K. Leaf litter decomposition and microbial activity in nutrient-enriched andunaltered reaches of a headwater stream. FreshwBiol 2003b;48:123–34.

Gulis V, Suberkropp K. Effects of whole-stream nutrient enrichment on the concentra-tion and abundance of aquatic hyphomycete conidia in transport. Mycologia2004;96:57–65.

Gulis V, Rosemond AD, Suberkropp K, Weyers HS, Benstead JP. Effects of nutrientenrichment on the decomposition of wood and associated microbial activity instreams. Freshw Biol 2004;49:1437–47.

Gulis V, Ferreira V, Graça MAS. Stimulation of leaf litter decomposition and associatedfungi and invertebrates by moderate eutrophication: implications for streamassessment. Freshw Biol 2006;51:1655–69.

Krauss G, Schlosser D, Krauss G-J. Aquatic fungi in heavy metal and organically pollutedhabitats. In: Deshmukh SK, Rai MK, editors. Biodiversity of fungi: their role inhuman life. Enfield, NH, USA: Science Publishers, Inc.; 2005. p. 221–46.

Mille-Lindblom C, Fisher H, Tranvik LJ. Antagonism between bacteria and fungi:substrate competition and a possible trade-off between fungal growth andtolerance towards bacteria. Oikos 2006;113:233–42.

Muyzer G, de Waal EC, Uitterlinden AG. Profiling of complex microbial populations bydenaturing gradient gel electrophoresis analysis of polymerase chain reaction-amplified genes coding for 16 S rRNA. Appl Environ Microbiol 1993;59:695–700.

Nikolcheva LG, Bärlocher F. Seasonal and substrate preferences of fungi colonizing leaves instreams: traditional versus molecular evidence. Environ Microbiol 2005;7:270–80.

Niyogi DK, Lewis Jr WM, McKnight DM. Litter breakdown in mountain streams affectedby mine drainage: biotic mediation of abiotic controls. Ecol Appl 2001;11:506–16.

Niyogi DK, McKnight DM, Lewis Jr WM. Fungal communities and biomass in mountainstreams affected by mine drainage. Arch Hydrobiol 2002;155:255–71.

Pascoal C, Cássio F. Contribution of fungi and bacteria to leaf litter decomposition in apolluted river. Appl Environ Microbiol 2004;70:5266–73.

Pascoal C, Cássio F, Gomes P. Leaf breakdown rates: a measure of water quality? Int RevHydrobiol 2001;86:407–16.

Pascoal C, Pinho M, Cássio F, Gomes P. Assessing structural and functional ecosystemcondition using leaf breakdown: studies on a polluted river. Freshw Biol2003;48:2033–44.

Pascoal C, Cássio F, Marvanová L. Anthropogenic stress may affect aquatic hyphomycetediversity more than leaf decomposition in a low-order stream. Arch Hydrobiol2005a;162:481–96.

Pascoal C, Marvanová L, Cássio F. Aquatic hyphomycete diversity in streams ofNorthwest Portugal. Fungal Divers 2005b;19:109–28.

Rademaker JLW, de Bruijn FJ. Computer-assisted analysis of molecular fingerprintprofiles and database construction. In: Kowalchuk GA, de Bruijn FJ, Head IM,Akkermans ADL, van Elsas JD, editors. Molecular microbial ecology manual. seconded. Dordrecht, Netherlands: Klumer Academic Publishers; 2004. p. 1397–446.

Rajapaksha RMCP, Tobor-Kapłon MA, Bååth E. Metal toxicity affects fungal and bacterialactivities in soil differently. Appl Environ Microbiol 2004;70:2966–73.

Rand GM, Wells PG, McCarty LS. Introduction to aquatic toxicology: effects,environmental fate, and risk assessment. In: Rand GM, editor. Fundamentals ofaquatic toxicology. second ed. London: Taylor and Francis; 1995. p. 3-66.

Romani AM, Fisher H, Mille-Lindblom C, Tranvik LJ. Interactions of bacteria and fungi ondecomposing litter: differential extracellular enzyme activities. Ecology2006;87:2559–69.

Rosemond AD, Pringle CM, Ramirez A, Paul MJ, Meyer JL. Landscape variation inphosphorus concentration and effects on detritus-based tropical streams. LimnolOceanogr 2002;47:278–89.

Royer TV, Minshall GW. Effects of nutrient enrichment and leaf quality on thebreakdown of leaves in a hardwater stream. Freshw Biol 2001;46:603–10.

Sampaio A, Cortes R, Leão C. Invertebrate andmicrobial colonisation in native and exoticleaf litter species in a mountain stream. Int Rev Hydrobiol 2001;86:527–40.

Sridhar KR, Bärlocher F. Initial colonization, nutrient supply, and fungal activity onleaves decaying in streams. Appl Environ Microbiol 2000;66:1114–9.

Sridhar KR, Krauss G, Bärlocher F, Raviraja NS, Wennrich R, Baumbach R, et al.Decompostion of alder leaves in two heavy metal-polluted streams in centralGermany. Aquat Microb Ecol 2001;26:73–80.

Sridhar KR, Bärlocher F, Krauss G-J, Krauss G. Response of aquatic hyphomycetecommunities to changes in heavy metal exposure. Int Rev Hydrobiol 2005;90:21–32.

Suberkropp K. Microorganisms and organic matter decomposition. In: Naiman RJ, BilbyRE, editors. River ecology and management: lessons from the Pacific coastalecoregion. New York, USA: Springer; 1998. p. 120–43.

Suberkropp K, Chauvet E. Regulation of leaf breakdown by fungi in streams: influencesof water chemistry. Ecology 1995;76:1433–45.

Triska FJ, Sedell JR. Decomposition of four species of leaf litter in response to nitratemanipulation. Ecology 1976;57:783–92.

Wang W-X, Dei RCH. Effects of major nutrient additions on metal uptake inphytoplankton. Environ Pollut 2001;111:233–40.

White TJ, Bruns T, Lee S, Taylor JW. Amplification and direct sequencing of fungalribosomal RNA genes for phylogenetics. In: Innis MA, Gelfand DH, Sninsky JJ, WhiteTJ, editors. PCR protocols: a guide to methods and applications. New York, USA:Academic Press; 1990. p. 315–22.

Zar JH. Biostatistical analysis. third ed. Englewood Cliffs: Prentice-Hall; 1996.