RESEARCH ARTICLE
Community composition of lacustrine small eukaryotes in hypereutrophic conditions in relation to top-down and bottom-up factors ´ Cecile Lepe`re1,2, Isabelle Domaizon2 & Didier Debroas1 1
Universite´ Blaise Pascal, Laboratoire de Biologie des Protistes, Aubie`re, France; and 2Universite´ de Savoie, Laboratoire CARRTEL, Le Bourget du Lac, France
Correspondence: Didier Debroas, Universite´ Blaise Pascal, Laboratoire de Biologie des Protistes, UMR CNRS 6023, 63177 Aubie`re, France. Tel.: 133 473 407837; fax: 133 473407837; e-mail:
[email protected] Received 8 November 2006; revised 4 May 2007; accepted 4 May 2007. First published online 26 July 2007. DOI:10.1111/j.1574-6941.2007.00359.x Editor: Patricia Sobecky Keywords small eukaryotes; lake; 18S rRNA; T-RFLP; cloning–sequencing; mesocosm experiment.
Abstract Small eukaryotes (0.2–5 mm) in hyper-eutrophic conditions were described using terminal restriction fragment length polymorphism and cloning–sequencing, and were related to environmental variables both by an experimental approach and by a temporal field study. In situ analysis showed marked temporal variations in the dominant terminal restriction fragments (T-RFs), which were related to environmental variables such as nutrient concentrations and metazooplankton composition. To monitor the responses of the small-eukaryote community to top-down (absence or presence of planktivorous fish) and bottom-up (low or high nitrogen and phosphorus addition) effects, a cross-classified design mesocosm experiment was used. Depending on the type of treatment, we recorded changes in the diversity of T-RFs, as well as modifications in phylogenetic composition. Centroheliozoa and Cryptophyta were found in all types of treatment, whereas Chlorophyta were specific to enclosures receiving high nutrient loadings, and were associated either with LKM11 and ‘environmental sequences’. Cercozoa and Fungi were not detected in enclosures receiving high nutrient loadings and fishes. Our results showed that resources and top-down factors are both clearly involved in shaping the structure of small eukaryotes, not only autotrophs but also heterotrophs, via complex interactions and trophic cascades within a microbial loop, notably in response to nutrient loading.
Introduction Small eukaryotes (0.2–;5 mm) are found in both marine and freshwater ecosystems, accounting for a significant fraction of the biomass (Stockner & Antia, 1986; Courties et al., 1994; Li, 1994; Fogg, 1995). They are likely to belong to various functional groups, and thus to play important roles as autotrophs, heterotrophs or mixotrophs. Because of its importance, the diversity and distribution of the smalleukaryote community have recently received considerable attention with newly developed molecular techniques. Studies have generally focused on marine food webs, however, and consequently the diversity, distribution and natural abundance of the freshwater small-eukaryote taxa are largely unknown. Only two recent studies have dealt with small-eukaryote diversity in lakes, and they reported a large diversity of rRNA sequences within this community (o 5 mm in size) (Lefranc et al., 2005; Richards et al., 2005). Furthermore, some seasonal changes in small-eukaryote FEMS Microbiol Ecol 61 (2007) 483–495
diversity structure were recently observed during a study conducted over two consecutive years in an oligomesotrophic lake (Lepe`re et al., 2006). The modification in small-eukaryote community composition (SECC) through time raises the question of how regulatory factors, such as grazing (top-down) and resources (bottom-up), exert pressure on this community. Within metazooplankton, all the main groups (cladocerans, copepods and rotifers) are able to have a significant predation effect on small eukaryotic cell abundance (Banse, 1992). Among protists, nanoflagellates and small ciliates (Scuticociliates, small Oligotrichidea) may also be potential grazers of picoplankton (Simek et al., 1995; Reckermann & Veldhuis, 1997; Samuelsson & Andersson, 2003). Moreover, several studies have shown that, in addition to grazing, viruses are a further important cause of mortality among prokaryotes and eukaryotes (Suttle et al., 1990; Weinbauer & H¨ofle, 1998; Jacquet et al., 2005). As regards bottom-up regulation, the picophytoplankton is dependent on the c 2007 Federation of European Microbiological Societies Journal compilation Published by Blackwell Publishing Ltd. No claim to original French government works
484
availability of nutriment, and one of the advantages possessed by these small cells is that they are characterized by a high surface/volume ratio with a large surface for exchange, which favours nutrient uptake. Although the mortality factors are now identified as potential driving forces in controlling food webs, their specific effects and, especially, their relative importance in comparison to bottom-up pressure are still a matter for discussion. A review of several parallel experiments testing the effects of fish addition and nutrient loading on shallow lakes (from Finland to southern Spain) showed that nutrient effects were about three times as important as the fish effect (in terms of their impact on primary producers and metazooplankton) (Moss et al., 2004). However, most of the studies dealing with top-down vs. bottom-up factors considered mainly metazooplankton and phytoplankton communities (nano- to micro-algae), and the picoplanktonic or small-eukaryote component was rarely taken into account. Clearly, there is now a need to enhance our understanding of the role of regulatory factors on the physiological, taxonomic and functional diversity of microbial communities. In order to reveal the community composition of the small-eukaryotes and to visualize its possible changes in relation to environmental variables, we investigated both the phylogenetic diversity of small eukaryotes (size o 5 mm) from an hyper-eutrophic freshwater system and the effects of resource and predation manipulations on this community. A cross-classified-design mesocosm experiment combined with several in situ samplings was used. The SECC was determined using terminal restriction length polymorphism (T-RFLP) analysis and cloning–sequencing of PCR-amplified 18S rRNA gene fragments. The potential modifications in SECC were considered in relation to the other planktonic communities (prokaryotes, viruses, nanoflagellates, ciliates, nano- and micro-phytoplankton, metazooplankton) in order to reveal direct and indirect regulatory links.
Materials and methods Study site Villerest reservoir is a 40 km-long artificial reservoir with a maximum depth of 40 m. Its relevant morphometric characteristics are described in Bonnet & Poulin (2002). We recorded the physical, chemical and biological characteristics of Villerest reservoir from April to July and from the mesocosm experiment.
In situ sampling Samples from Villerest reservoir were collected bimonthly in 1999 at 1 m depth with a Van Dorn bottle at the sampling station referred to as ‘Cordelle’, which is located in the middle of the reservoir, near the experimental facilities. c 2007 Federation of European Microbiological Societies Journal compilation Published by Blackwell Publishing Ltd. No claim to original French government works
C. Lepe`re et al.
Water was pumped from this sampling station to fill the mesocosms.
Mesocosm experiment The experiment was conducted at the Villerest mesocosm facility, which consisted of 12 white fiberglass tanks, each 2.2 m high, 1.8 m in diameter and containing 5500 L (Domaizon & D´evaux, 1999). During the experiment, to prevent stratification, tanks were mixed for 3 h each day with an airlift mixer system. Water was pumped into the tanks on 18 June using a submerged pump, and the time for complete filling of each tank was o 15 min. The experimental design of the study, a 2 2 factorial design (with or without fish low or high nutrient addition), resulted in four treatment combinations: (1) low nutrient addition and fishless, termed ‘n’; (2) low nutrient addition and the presence of fish, termed ‘nF’; (3) high nutrient addition and fishless, termed ‘N’; and (4) high nutrient addition and the presence of fish, termed ‘NF’. Each treatment combination had three replicates, necessitating 12 experimental tanks. The treatment combinations were randomly assigned to tanks. The treatment combination with low nutrient addition received 10 mgP L1 and 60 mgN L1, whereas the treatment combination with high nutrient addition received 50 mgP L1 and 300 mgN L1. Nutrients were added in the form of H3PO4 and NaNO3 every 3 days throughout the 21 days of the experiment. The tanks with high nutrient addition maintained nutrient concentrations similar to those measured in the reservoir. In the fish enclosures we used 1-year-old roach (01) (Rutilus rutilus L.), planktivorous fishes that are able strongly to structure the metazooplankton assemblage. The mean ( SD) individual weight and length of roach were 21.3 2.9 g and 11.7 0.5 cm, and fishes were stocked at a density of five roach per tank (20 g m3), based on the results of a previous mesocosm experiment that determined the density at which roach had a significant effect on metazooplankton assemblages (Domaizon & De´ vaux, 1999). On the basis of experience gained from experimental manipulations made in comparable conditions (Domaizon & D´evaux, 1999), samples were first collected after 3 days (21 June), the latent period for the stabilization of communities in tanks, and then at 3-day intervals during the 21 days of the experiment.
Main biotic and abiotic variables The following parameters were analysed both for in situ and for mesocosm samples. The water temperature and dissolved oxygen were determined with a multiparameter probe (YSI GRANT 3800). Phosphate (PO4-P), ammonium (NH4-N), nitrate (NO3-N) and nitrite (NO2-N) were analysed in the laboratory using standard methods (American FEMS Microbiol Ecol 61 (2007) 483–495
485
Lacustrine small eukaryotes in hyper-eutrophic conditions
Public Health Association, 1992). For determining total prokaryotic abundance, 1–6 mL samples (fixed in formaldehyde solution, final concentration 2%) were filtered onto 0.2-mm black polycarbonate filters (25 mm, Millipore) and stained with 1 mg L1 (final concentration) of 4,6-diamidino-2-phenylindole (DAPI) (Porter & Feig, 1980). Water samples for chlorophyll estimation were filtered through Whatman GF/C filters and frozen immediately. Chlorophyll was extracted with 90% acetone, and the absorbance was measured at 665 and 750 nm before and after acidification, according to Lorenzen’s (1967) modified method. Phytoplankton was counted using Uterm¨ohl’s (1958) method with a Leitz-type inverted microscope (Wild M40). Ciliates were fixed in mercuric chloride solution (final concentration 2.5%) (Staley & Gosink, 1999) and counted using Uterm¨ohl’s (1958) method. Flagellates were preserved in a solution of glutaraldehyde (final concentration 1%). Subsamples (30–40 mL) for counting were stained with primulin (Caron, 1983) and collected onto black 0.8 mm pore size Nucleopore filters. In the mesocoms, pigmented and heterotrophic nanoflagellates (PNFs and HNFs) were separated into two fractions, small eukaryotes (o 5 mm) and large eukaryotes (4 5 mm), by measuring cell lengths. Metazooplankton was sampled with vertical tows of 60 mm-mesh plankton net, preserved in 5% sucrose formalin (Prepas, 1978), and counted under a binocular microscope (Wild M3Z) in a Dolfus chamber. For the in situ study, metazooplankton was sampled from the whole water column (0–40 m) in order to integrate dial migrations of some species.
Molecular analysis For the study of SECC we chose a snapshot analysis at t = 21 days. The greatest difference between treatments were obtained at t = 21 days. Indeed, for both metazooplankton and chlorophyll a, which are the two groups directly affected by the presence of fish and the addition of nutrients, the greatest differences between treatments were observed at t = 21 days (Fig. 1).
Nucleic acid extraction Water samples were prefiltered through 5 mm pore-size polycarbonate prefilters (Millipore) at a very low vacuum to prevent cell damage (pressure o 20 mbar). Nucleic acid extraction was performed as described previously by Lefranc et al. (2005). Nucleic acids were extracted with chloroform–isoamyl alcohol (24 : 1); the aqueous phase containing nucleic acids was kept and purified by adding phenol–chloroform–isoamyl alcohol (25 : 24 : 1). After adding isopropanol (0.6 volume), the nucleic acids were precipitated at FEMS Microbiol Ecol 61 (2007) 483–495
Fig. 1. Temporal changes in (a) zooplankton abundances and (b) chlorophyll a concentrations in the four types of treatment (mean value obtained for triplicates).
20 1C for 12 h. After centrifugation, the DNA pellet was ethanol-rinsed and resuspended in 50 mL of TE buffer.
T-RFLP analysis T-RFLP analysis was conducted on in situ and mesocosm samples. 18S rRNA genes were amplified using fluorescently labelled forward primer Ek-1f-FAM (CTGGTTGATCCTTGCCAG) (6-carboxylfluorescein) (labelled at the 5 0 -end with fluorescent sequencing dye; MWG Biotech, Germany) and Ek-1520r (CYGCAGGTTCACCTAC) (Lo`pez-Garcia et al., 2001). From six to eight PCR products were pooled and purified using a Qiaquick PCR purification kit (Qiagen), visualized on 1% agarose gels and quantified (DNA quantification kit; Sigma). Enzymatic digestions were performed by incubating separately 100 ng of PCR products with 20 U of MspI and RsaI (Sigma) at 37 1C overnight. The samples were desalted with Microcon columns (Amicon, Millipore). The terminal restriction fragments (T-RFs) were separated on an automated sequencer (ABI 3700). T-RF sizes between 48 and 684 bp with a peak area 4 50 fluorescence units were determined using GENESCAN analytical software (Dunbar et al., 2001; Boucher et al, 2006). Samples were analysed in triplicates, and a peak was kept if it occurred in at least two profiles. The relative abundance of T-RFs was determined by calculating the ratio between the peak area of each peak and the total peak area of all peaks within one sample. Ratios were converted into percentages, and the results are displayed as histograms. c 2007 Federation of European Microbiological Societies Journal compilation Published by Blackwell Publishing Ltd. No claim to original French government works
486
T-RFs obtained from enzymatic digestion by MspI and RsaI allowed us to track changes in SECC for each of the four treatments and for the in situ study. For the mesocosms, digestion by MspI led to a higher number of T-RFs (mean: 42 T-RFs for the four treatments) than did digestion by RsaI (36 T-RFs). For the natural sampling, on average 138 T-RFs were found with MspI and 30 with RsaI (four dates tested). As MspI is more discriminative in terms of diversity, only data obtained with this enzyme are presented. Eukaryotic rRNA genetic library Eukaryotic rRNA genetic libraries were constructed on mesocosm samples. Environmental DNA extracted from one replicate of each treatment (n, N, nF, NF) was used to construct the 18S rRNA gene clone library. The eukaryotespecific primers Ek-1f and Ek-1520r were used in PCR amplifications. The clone library was constructed using a TOPO TA Cloning Kit (Invitrogen) with the PCR vector 2.1, according to the manufacturer’s instructions. About 60 clones were randomly picked from different plates. The presence of the 18S rRNA gene insert in positive colonies was checked by PCR amplification using flanking vector primers (M13f and M13r). Fingerprint analysis was conducted according to Lefranc et al. (2005). At least one clone of each operational taxonomic unit (OTU) was selected and extracted with a QIAprep Spin Miniprep Kit (Qiagen). Euk-1F was used for partial sequencing. Sequencing reactions were performed using MWG (http:// www.mwg-biotech.com). Phylogenetic analysis To determine the first phylogenetic affiliation, each sequence was compared with sequences available in databases from the National Center for Biotechnology Information using BLAST (Altschul et al., 1997). The sequences were aligned with 20 427 sequences of an ARB database (ssu_jan-03.ARB) using the latter’s automatic alignment tool (www.arb-home.de) (Ludwig et al., 2004). The resulting alignments were checked and corrected manually, considering the secondary structure of the rRNA molecule. Sequences were inserted into an optimized tree according to the maximum parsimony criteria without allowing any changes to the existing tree topology (ARB software). The resulting tree was pruned to retain the closest relatives, sequences representative of eukaryotic evolution and our clones. The sequences were screened for potential chimeric structures using Chimera check from Ribosomal Database Project II and by constructing alternative phylogenetic trees using 350-bp pieces from the 5 0 and 3 0 ends of the sequenced DNA fragments. Two chimeras were removed from the dataset. Nucleotide sequences determined in this study have been deposited c 2007 Federation of European Microbiological Societies Journal compilation Published by Blackwell Publishing Ltd. No claim to original French government works
C. Lepe`re et al.
in the genbank database under accession numbers DQ409084–DQ409135. Good’s coverage was used as a coverage index. It is a nonparametric estimator of the proportion of phylotypes in a library of infinite size that would be represented in a smaller library. As defined by Good (1953), the coverage C is calculated as C ¼ 1 n1 =N; where n1 is the number of phylotypes appearing only once in a library, and N is the library size. The relative distribution of OTUs in the library was also used to calculate an independent estimator of the total number of OTUs, namely the SChao1 estimator (Kemp & Aller, 2003).
Statistical analysis Univariate analysis We tested the homogeneity of the main biological variables (prokaryotes, metazooplankton, phytoplankton, flagellates, and small eukaryotes) in tanks at the initial time of the experiment using an ANOVA. To test the effects of nutrient enrichment and fish presence on the various parameters, we used a two-way ANOVA. Equality of the variances and normality of the residuals were tested by Bartlett and Shapiro–Wilk tests. Multivariate analysis The presence–absence matrix of T-RFs from the mesocosm experiments was analysed by correspondence analysis (COA). To explain the variation of in situ SECC, measured by T-RFLP and expressed by percentage area (4 2%), in relation to environmental data, canonical correspondence analysis (CCA) was used (Ter Braak, 1986). We tested the following explanatory variables: NH4-N, NO2-N, NO3-N, PO4-P, total phosphorus, temperature, dissolved oxygen, prokaryotic abundance, and chlorophyll a, HNF (heterotrophic nanoflagellates), and metazooplankton (cladocerans, copepods and rotifers) abundances. These statistics were computed with R software using the ADE package for the COA, and the VEGAN package (http:// cc.oulu.fi/jarioksa/softhelp/vegan.html) for the CCA and related methods (http://cran.r-project.org/).
Results Environmental variables Temporal in situ study The physical, chemical and biological characteristics at 1 m depth in Villerest reservoir were monitored from April to FEMS Microbiol Ecol 61 (2007) 483–495
487
Lacustrine small eukaryotes in hyper-eutrophic conditions
Table 1. Mean, minimum and maximum values recorded for environmental variables at 1 m depth in Villerest reservoir Parameter
Value for 1 m depth 6
1
Bacteria (10 cells mL ) Chl a (mg L1) Zooplankton (ind L1) Cladocerans (ind L1) Copepods (ind L1) Rotifers (ind L1) Phytoplankton (106 cells L1) PNF (103 cells mL1) HNF (103 cells mL1) NO3-N (mg N L1) NO2-N (mg N L1) NH4-N (mg N L1) PO4-P (mg P L1) Temperature ( 1C) Oxygen (mg L1)
6.5 (3.4–10.5) 19.0 (0.2–54.3) 80.7 (11.7–205.8) 8.5 (1.7–21.7) 32.6 (4.1–100) 39.7 (2.1–84.1) 0.8 (0.1–2.4) 1.0 (0.1–4.0) 1.2 (0.1–5.8) 1.2 (0.3–3.1) 0.0 (0.0–0.01) 0.1 (0.0–0.2) 0.1 (0.005–0.1) 18.3 (9.9–26.1) 8.5 (6.5–10.0)
Variables for which samples were integrated from 0 to 40 m.
PNF, pigmented nanoflagellates; HNF, heterotrophic nanoflagellates; ind, individuals.
nutrients led to a significant increase in chlorophyll a content (treatments N and NF) (P o 0.05), and the highest concentrations were recorded on day 21 (Fig. 1b, Table 2). Four systematic groups were distinguished in the quantitative study of phytoplankton: Cyanobacteria, Chlorophyceae, Diatoms and Cryptophyceae. More precisely, in terms of biovolume, Cryptophyceae and Chlorophyceae were mostly retrieved in the N and NF treatments (P o 0.05) (Table 2), whereas Diatoms reached a maximum value in the NF treatment. Prokaryotes and ciliates Prokaryotic abundances increased significantly (P o 0.05) after the addition of nutrients by direct or indirect effects (Table 2), with the highest values being recorded in the N and NF treatments. The highest density of ciliates was recorded in the nF treatment (15 000 ind L1); in the three other treatments the average density was only 1900 ind L1. Large flagellates
July. The mean, minimum and maximum values recorded for each variable are listed in Table 1.
On day 0, the statistical analysis showed that there were no significant differences between mesocosms in terms of chemical and biological parameters and, more particularly, for the small-eukaryote community (results not shown). Thus, conditions were considered homogeneous in all mesocosms before manipulation.
Microscopic counts showed that large-sized (5–30 mm) total flagellates (pigmented and heterotrophs) had the lowest abundances in nF mesocosms (P o 0.05), where Bosmina and rotifers were the main grazers. Moreover, the decrease in pigmented flagellate abundance with fish addition was significant (P o 0.05). More precisely, Chrysidalis dominated (90% of total large flagellates) in the treatments receiving low nutrient additions (n and nF), whereas for the N and NF treatments, flagellates were divided between Cryptomonas and Chlamydomonas.
Metazooplankton
Global characteristic of the small-eukaryote community
Mesocosm experiments
Cladocerans were divided into two class sizes: small cladocerans (roughly o 500 mm), comprising Bosmina longirostris and Chydorus sphaericus; and large cladocerans (4 500 mm), comprising Ceriodaphnia quadrangula and Daphnia spp. Large cladocerans are heavily predated by fish, so that, at t = 21 days, treatments nF and NF had the lowest densities of Daphnia and Ceriodaphnia (mean = 4.7 individuals L1) (Table 2). As regards small cladocerans, significantly higher abundances were recorded when nutrient levels were moderate (treatments n and nF). The highest density of rotifers was reached in NF mesocosms (1868 ind L1). No significant statistical difference between the four types of treatment was recorded because SDs around the means were large. Chlorophyll a and phytoplankton In the n and nF mesocosms the chlorophyll a concentration was on average 136 mg L1. As expected, the addition of FEMS Microbiol Ecol 61 (2007) 483–495
In all mesocosm treatments, the mean abundance of small eukaryotes (o 5 mm) was 7.02 103 cells mL1. Among heterotrophs we distinguished various morphotypes, such as Monas-like cells (2–5 mm) and unidentified uniflagellates (2–4 mm), which represented a considerable fraction within the heterotrophic small eukaryotes (30–65% of the total abundance, according to the treatment). Changes in the genetic diversity of the small-eukaryote assemblages in situ and in mesocsom experiments under the different conditions were visualized by T-RFLP analysis of 18S rRNA gene fragments obtained after amplification of rRNA gene. T-RFs that had a percentage area 4 2% were considered as dominant. Out of 100 T-RFs (detected by MspI) obtained in situ, 49 T-RFs with a percentage area 4 2% were detected over the entire study period; in the mesocosm experiments we retained 28 T-RFs out of 131 (Fig. 2a and b). For each sampling, only a few T-RFs represented a high percentage of the total area; in the c 2007 Federation of European Microbiological Societies Journal compilation Published by Blackwell Publishing Ltd. No claim to original French government works
488
C. Lepe`re et al.
Table 2. Mean values ( SDs) recorded for variables measured in the four types of treatment at T = 21 days (n, low nutrient addition and fishless; nF, low nutrient addition and fish present; N, high nutrient addition and fishless; NF, high nutrient addition and fish present), and results of the two-way ANOVA
ANOVA
Four types of treatments Response of variables
n
N
2.23 ( 2.15) 3.46 ( 1.76) Bacteria (106 cells mL1) Large size cladocerans 22.47 ( 3.62) 46.64 ( 14.14) (ind L1) Small size cladocerans 464.83 ( 57.93) 9.67 ( 7.32) (ind L1) Copepods (ind L1) 441.94 ( 79.29) 578.46 ( 157.15) 250.65 ( 184.38) 42.51 ( 37.77) Rotifers (ind L1) Chl a (mg L1) 133.6 ( 12.27) 300.63 ( 22.88) Cyanobacteria 2.43 ( 0.20) 3.55 ( 0.04) (107 mm3 mL1) Chlorophyceae 0.85 ( 0.05) 2.38 ( 0.60) (107 mm3 mL1) Diatoms 0.02 ( 0.01) 0.1 ( 0.01) (107 mm3 mL1) Cryptophyceae 0.15 ( 0.06) 0.92 ( 1.43) (107 mm3 mL1) Total large flagellates 3.23 ( 0.26) 1.57 ( 0.14) (104 cells mL1) Large pigmented 1.32 ( 0.17) 0.31 ( 0.11) flagellates (104 cells mL1) Large heterotrophic 1.91 ( 1.13) 0.79 ( 0.32) flagellates (104 cells mL1) Ciliates (103 cells L1) 1.4 ( 0.78) 1.71 ( 1.04) Pigmented small 1.28 ( 0.98) 0 ( 0) eukaryotes (103 cells mL1) Heterotrophic small 5.45 ( 1.87) 3.89 ( 1.65) eukaryotes (103 cells mL1) T-RFs numbers 54 ( 5.19) 43 ( 3.22) Shannon index 3.42 ( 0.12) 2.65 ( 0.03)
nF
NF
results (P)
N (nutrients)
F (Fish)
FN (interaction)
1.8 ( 1.18)
3.04 ( 1.90)
0.010 (1)
NS
0.020
3.92 ( 1.74)
5.62 ( 0.92)
NS
0.010 ( )
0.153
0.001 ()
NS
0.256
NS NS o 0.001 (1) NS
NS NS NS NS
NS NS 0.257 NS
505.36 ( 85.52)
169.44 ( 9.74)
414.76 ( 209.01) 326.94 ( 90.53) 283.52 (190.41) 575.06 ( 246.73) 138.36 ( 18.66) 372.63 ( 66.60) 3.76 ( 0.92) 5.02 ( 0.43) 0.65 ( 0.16)
2.06 ( 0.96)
0.005 (1)
NS
0.980
0.03 ( 0.01)
3.65 ( 0.37)
NS
NS
NS
0.09 ( 0.08)
1.43 ( 1.06)
NS
NS
NS
0.54 ( 0.16)
1.07 ( 0.23)
NS
0.003 ()
0.010
0.35 ( 0.23)
0.34 ( 0.10)
NS
0.050 ()
0.040
0.70 ( 0.28)
1.05 ( 0.68)
NS
NS
NS
15 ( 9.72) 0 ( 0)
2.7 ( 1.08) 6.99 ( 2.23)
NS 0.050 (1) 0.010 o 0.001 (1) o 0.001 (1) o 0.001 (1)
2.13 ( 1.10)
8.86 ( 3.81)
NS
NS
NS
31 ( 3.57) 2.6 ( 0.06)
39 ( 5.08) 2.87 ( 0.10)
NS 0.040 ( )
0.003 () 0.017 ()
0.006 0.004
P-values obtained for nutrient effects (N), fish effects (F) and the interaction between the two factors are presented.1 and signs indicate the direction of the effects (positive or negative impact). Bold font corresponds to significant values. ind, individuals.
mesocosms, nine T-RFs represented on average 62% of the total area. T-RFLP results showed that 64.3% of the T-RFs found in the mesocosms at 21 days of incubation (highest contrasted conditions between treatments) were also observed in the samples from the in situ study. In the mesocosms, in order to describe the phylogenetic diversity of small eukaryotes in eutrophic conditions precisely, one clone library was constructed for each type of experimental treatment. COA showed that there was a good reproducibility between replicates (Fig. 3), and we therefore randomly chose one replicate per treatment for the cloning. c 2007 Federation of European Microbiological Societies Journal compilation Published by Blackwell Publishing Ltd. No claim to original French government works
The SChao1 index was 77.8, and the coverage value for our genetic library was 70.8%. Fifty-two OTUs were identified after the analysis of RFLP profiles (Fig. 4). Cryptophyta accounted for 18% of the total OTUs and was therefore the major group. Chlorophyta were also well represented in the library, accounting for 14%. Fungi, Ciliophora, Centroheliozoa and Cercozoa occurred in approximately the same proportion (12% of OTUs). Eight percent of OTUs were not clearly affiliated by BLAST. Phylogenetic analysis placed these sequences between Fungi and Ichthyosporea and they have been named ‘environmental sequences’. We also identified FEMS Microbiol Ecol 61 (2007) 483–495
Lacustrine small eukaryotes in hyper-eutrophic conditions
489
Fig. 3. Graphic representation of correspondence analysis treated by the presence–absence of T-RFs.
Fig. 2. Temporal variations from April to July at 1 m depth (date format: day/month) (a) and treatment variations (b) in the number and relative abundance of the various terminal restriction fragments (T-RFs) detected by T-RFLP analysis of 18S rRNA gene digestion by MspI. Only T-RFs representing more than 2% of the total area are considered. The total number of T-RFs detected for each sample is added above each histogram bar. Thus, although 119 T-RFs were detected on 24/06/99, only six T-RFs represented nearly 70% of the total area and were considered to be dominant T-RFs.
sequences belonging to Dinophyceae and Choanoflagellida (4% of OTUs each), and environmental sequences LKM11(2% of OTUs). Finally, Bicosoecida and Apicomplexa were present in low proportions (1% each). Some sequences were not inserted in a specific clade in the tree (Fig. 4), such as VNP13 and VP36, which are affiliated with Fungi and Ciliates, respectively. The majority of OTUs (67.3%) in the library were affiliated with heterotrophic organisms, and microscopic counts confirmed this dominance.
Variations of SECC with environmental variables In the mesocosm experiment, small eukaryotes had the highest abundance in the NF treatments (15.8 103 cells mL1), whereas in the nF treatment their abundance FEMS Microbiol Ecol 61 (2007) 483–495
was not more than 2.1 103 cells mL1 (P o 0.05). In contrast to the case for pigmented small eukaryotes, no significant difference was observed between treatments for heterotrophic organisms (Table 2). The number of T-RFs per sample fluctuated strongly during the in situ study (25–165), and also in enclosures. The highest number of T-RFs was recorded in n treatments, with an average of 54 T-RFs, followed by mesocosms receiving high nutrient loadings (43 T-RFs). The lowest diversity was measured in nF treatments (31 T-RFs). The ANOVA showed a significant negative impact of the presence of fish on the number of T-RFs (Table 2). Concerning the in situ analysis, marked variations in the dominant T-RFs were observed according to the sampling date (Fig 2a). CCA showed that some environmental variables were related to the variations in SECC (Fig. 5). On 7 April, the number of dominant T-RFs (18) was much higher than that observed for the other dates and accounted for more than 90% of the total area. This particular structure within the small-eukaryote community was related to phytoplankton and chlorophyll a concentrations, and consequently to the high concentrations of nitrogen nutrients (NO3-N, NO2-N) (Fig. 5), which reached their maximal values on this date. The analysis of T-RFLP profiles (Fig. 2a) and the CCA plot (Fig. 5) also highlighted a modification of structure on 19 May. On this date the dominant T-RFs accounted for only 61% of the total area, whereas on average during the study they represented about 80%. Moreover, among the 20 T-RFs, 14 were specific to this date and were not observed on any other date. This specific composition of small eukaryotes was linked mainly to high abundances of metazooplankton, HNFs and phosphorus. However, dramatic changes occurred between 24 June and 7 July, but this modification cannot be explained by the environmental parameters recorded in this study. T-RF specificity was verified by the mesocosm approach. Indeed, some T-RFs were observed specifically in mesocosms receiving a high nutrient loading (T-RFs 237 and 399). T-RF 383 was found c 2007 Federation of European Microbiological Societies Journal compilation Published by Blackwell Publishing Ltd. No claim to original French government works
490
C. Lepe`re et al.
Fig. 5. Canonical correspondence analysis plot performed on T-RFs expressed by percentage area ( 4 2%) obtained from the sampling in Villerest reservoir. PNF, pigmented nanoflagellates; HNF, heterotrophic nanoflagellates; Pt, total phosphorus
Fig. 4. Phylogenetic tree of eukaryotic small-subunit rRNA genes showing the position of environmental clones. The tree was constructed using the ARB software package. The treatments in which we found sequences are given in brackets.
only in nF enclosures, whereas T-RF 300 was specific to n mesocosms. The analysis of the presence–absence matrix of T-RFs in the different experimental treatments using COA showed a discrimination of SECC according to nutrient enrichment (axis 1) and the presence/absence of fish (axis 2) (Fig. 3). In contrast, we observed some T-RFs to be widely represented both during the in situ survey and in the different types of experimental treatment. In situ, some c 2007 Federation of European Microbiological Societies Journal compilation Published by Blackwell Publishing Ltd. No claim to original French government works
T-RFs, such as the T-RFs 397, were recorded on each date and always represented a large area, except on the two final days, on which the area represented o 2%, suggesting that the small eukaryotes corresponding to this T-RF were fairly stable, despite temporal changes in environmental parameters. Similarly, T-RFs 241, 395, 301 were present in three types of treatment during the mesocosm experiment; however, it is noteworthy that no T-RF was recorded in all types of treatment. The results of cloning–sequencing in mesocosms offer a taxonomic vision of changes in SECC. In contrast to T-RFLP analysis, at the scale of clades, cloning showed that two groups, the Centroheliozoa and the Cryptophyta, were found in all types of treatment, but differences in the relative importance of this last group could be observed. OTUs of Cryptophyta were more abundant in N and nF enclosures, where they accounted for 16% and 20%. Cercozoa and Fungi were identified in the same three treatments (but were absent in the NF treatment). Ciliophora were also present in three enclosures, but were absent in n treatments. Chlorophyta were specific to enclosures receiving high nutrient loadings (N and NF). Other groups were retrieved in only one treatment; thus, Dinophyceae, Bicosoecida and Apicomplexa (sequence VP1) were recorded in the nF treatment. Choanoflagellida were observed in treatments receiving moderate amounts of nutrients (n), whereas LKM11 groups were present in mesocosms characterized by high nutrient additions (N).
Discussion Our experimental results showed the remarkable specificity of some groups according to treatment in these eutrophic conditions. The relative abundance of small plankton generally tends to decrease with increasing nutrient load (Bell & Kalff, 2001). In these environmental conditions, however, FEMS Microbiol Ecol 61 (2007) 483–495
Lacustrine small eukaryotes in hyper-eutrophic conditions
small-eukaryote abundance was similar to abundances registered in more oligotrophic conditions, such as in marine areas (D´ıez et al., 2001) or in an oligotrohic lake (Lefranc et al., 2005). The CCA showed the links that exist between these changes and the modifications of biotic and abiotic variables in situ, mainly metazooplankton abundance and nutrient concentrations (nitrogen and phosphorus). A number of similar seasonal changes linked to environmental parameters had previously been reported during a long-term study conducted in an oligo-meso-trophic lacustrine system (Lepe`re et al., 2006). Similarly, variations in T-RFLP profiles according to treatment involved both the number of T-RFs and their relative proportions. Although a clear specificity appeared according to the treatment, some T-RFs (241, 395) were present in three of the four types of treatment and were also found in the lake-water samples. 64.3% of the T-RFs obtained in mesocosms were also found in in situ observations, suggesting that the evolution of SECC in the enclosures was not a specific development of communities arising from the mesocosm confinement, but was probably connected to the top-down and bottom-up manipulations.
Structure of the small-eukaryote community in hyper-eutrophic conditions Two groups, namely Cryptophyta and Centroheliozoa, were present in all types of treatment. The presence of Cryptophyta in the smallest planktonic fraction agrees with previous results obtained from lacustrine ecosystems (Stockner & Shortreed, 1989; Lefranc et al., 2005) and from lakes studied in the same geographic area, where Cryptophyta predominate among the pigmented organisms (Carrias et al., 1996; Thouvenot et al., 2000). The library of clones produced from the Villerest mesocosms showed that Cryptophyta dominated the other algae groups, in particular the genus Cryptomonas, which was also identified by microscopy. Phylogenetic identification also showed the presence of lineages rarely observed in this size fraction, for example Centroheliozoa. Indeed, our results revealed the recurrent presence of Centroheliozoa, which form the major part of the heliozoan phylum (Cavalier-Smith & Chao, 2003). This is in contrast to a previous study on three lakes of different trophic status (oligotrophic, mesotrophic and eutrophic), in which no Centroheliozoa sequence was found (Lefranc et al., 2005). This group is also absent (D´ıez et al., 2001; Lo`pez-Garcia et al., 2001) or rare (Yuan et al., 2004) in the marine environment. In these studies, however, cloning was conducted in situ and at a single point in time, so that possible temporal variations in SECC were not revealed and there was only a selective view of the community structure. Zimmermann et al. (1996) showed that heliozoan cell FEMS Microbiol Ecol 61 (2007) 483–495
491
numbers were significantly positively correlated with the chlorophyll a concentration close to the surface of Lake Constance. The development of heliozoans is greater during the summer period, and it is likely, therefore, that our chlorophyll-rich experimental system favoured the development of this group. Cercozoa, Fungi, Ciliates and ‘environmental sequences’ were identified in almost all treatments. Cercozoa sequences have usually been observed in studies conducted on the marine eukaryote picoplankton but never dominated the clone library (D´ıez et al., 2001; Dawson & Pace, 2002; Romari & Vaulot, 2004), in contrast to what has been observed in the lacustrine environment (Lefranc et al., 2005; Lepe`re et al., 2006). Molecular biology techniques allowed us to demonstrate the existence of this group, which is not usually identified by microscopy in such environments but is probably classified as unidentified flagellates in many studies because of the general lack of distinct morphological features of these small cells (Keeling, 2001). We also revealed the presence of several fungal species that escaped identification using traditional identification methods (Lefranc et al., 2005). Our experimental approach allowed us to confirm their recurring presence in most of our treatments (with the exception of NF). It would be interesting to investigate in more detail the functional roles of fungi, which could, among others, include the recycling and decomposition of organic matter. They are also known to be parasites of phytoplankton (Lopez-llorca & Hernandez, 1996; Ibelings et al., 2004). Our results showed that the composition of the fungal community changes depend on resources and predation levels. The presence of this group in many aquatic environments suggests that these organisms are a fundamental component of aquatic microbial ecosystems and that the ecological impact of parasitism is certainly underestimated. Among pigmented organisms, Chlorophyta are characteristic of nutrient-rich systems. A study conducted in a eutrophic lake in Antarctica showed, as in our study, that the dominant algal group was Chlorophyta, mainly represented by Chlamydomonas, (Izaguirre et al., 2001). In agreement with our study, Dinophyceae have previously been discovered in the small-eukaryote fraction (Moon-van der Staay et al., 2001; Yuan et al., 2004) in the marine environment, although our Dinophyceae sequences were not inserted in marine clusters (Groisillier et al., 2006). Their detection in this planktonic fraction may, however, be the result of filtration problems or else of the existence of unidentified small Dinophyceae (Lefranc et al., 2005). The few studies relating to the lacustrine SECC seem to show that some lineages, such as Cryptophyta, Fungi and Cercozoa, are present in all the lakes studied (Fig. 4) (Lefranc et al., 2005). For example, we can see that the Cercozoa phylogenetic group was constituted by sequences c 2007 Federation of European Microbiological Societies Journal compilation Published by Blackwell Publishing Ltd. No claim to original French government works
492
stemming from the Villerest Lake study, as well as by sequences A51 and P34.14, which arose, respectively, from Aydat Lake (eutrophic) and from Pavin Lake (oligomesotrophic) (Lefranc et al., 2005). The discovery of similar lineages of small eukaryotes in different environments (various trophic statuses, marine and lacustrine systems) indicates that the species are widespread (Staley & Gosink, 1999). However, the phylogenetic tree (Fig. 4) showed that some lineages formed particular clades and were found only in this lake; for example, Centroheliozoa and ‘Environmental sequences’ were not found in previous studies conducted in lakes (Lefranc et al., 2005; Richards et al., 2005). Furthermore, in spite of the very different environmental conditions in the mesocosms, we did not detect Perkinsozoa, previously described in some lakes (Lefranc et al., 2005). These results could therefore suggest that there is a certain specificity of the SECC in hyper-eutrophic systems. However, there is still a lack of data on freshwater SECC, and our conclusions have to be seen light of this.
Bottom-up and top-down effects on abundance and composition of small eukaryotes The prediction from classical food-chain theory (phytoplankton, metazooplankton–planktivorous fish) fitted rather well with our experimental data. Food-chain theory predicts that nutrient loading increases phytoplankton growth after selectively increasing the proportion of green algae, and decreasing the proportion of small algae and phytoplankton diversity (Moss et al., 2004). In our study, the Shannon index, calculated from T-RFLP analysis, showed that nutrient loading also led to a decrease in small-eukaryote diversity (Table 2). However, this decrease in diversity was probably not caused by nutrients alone; indeed, the results of the ANOVA showed the existence of an interaction between the effect of fish and nutrients on the small-eukaryote diversity. With increasing nutrient levels, algal biomass does not necessarily increase, as a greater proportion of the algal biomass production is removed by a growing herbivorous metazooplankton community. However, if planktivorous fish are also present, algal biomass will increase proportionately, as planktivores will keep the herbivores at a low level (Persson et al., 1988; Hansson et al., 2004). Fish tend to selectively remove the larger Cladocera (Gliwicz, 2004), and thus increase the rotifer biomass. Among the direct effects previously reported in classical food-web theory, we confirmed the significant impact of nutrients on the total phytoplankton communities, including small pigmented eukaryotes. This result confirms recent observations made during a study conducted on Lake Pavin, which showed that, among the bottom-up factors, nutrients seemed to be the main factor associated with variations in SECC in the photic zone. c 2007 Federation of European Microbiological Societies Journal compilation Published by Blackwell Publishing Ltd. No claim to original French government works
C. Lepe`re et al.
More especially, nitrogenous nutrients (NH1 4 ) and total phosphorus play a significant role (Lepe`re et al., 2006). Thus, in this study we observed, especially by cloning, a specificity in the community composition of the pigmented small eukaryotes that was directly related to nutrient levels. Some phylotypes were present at background abundances at the initial time, and increased above the detection threshold during the experiment following the manipulation of nutrients and/or fish levels. Dinophyceae were only observed in the nF treatments and are generally absent from hypereutrophic lakes, although they become more common as lake trophic status declines (Watson et al., 1997). We observed an increase in the proportion of pigmented small-eukaryote sequences (Cryptophyta and Chlorophyta) in N enclosures. The LKM11 sequence was associated with Chlorophyta in treatments with high nutrient loadings (N and NF). LKM11 seems to be associated with the decomposition of detritus composed of algae and Cyanobacteria (Van Hannen et al., 1999). LKM11 could therefore participate in the recycling of the organic matter stemming from the development of Chlorophyceae and Cyanobacteria, which represented the majority of the phytoplankton population in these treatments. In enclosures with fish (NF and nF), small eukaryotes were exposed mainly to the predation of rotifers and small cladocerans (Bosminidae, Chydoridae). It seems that the impact of predation on eukaryotic taxonomic groups varied according to the structure of potential predators. As an example, the absence of some groups affiliated to potential parasites (Cercozoa, ‘environmental sequences’, and Fungi) in the NF treatment could be associated with the high abundance of rotifers and high biomass of Diatoms. Rotifers are known preferentially to ingest particles of size between 1 and 5 mm (Ross & Nunavar, 1981), more particularly flagellates with no specific protection (Pont, 1995). Moreover, Cercozoa and some of the Fungi, affiliated with the lineage of Chytrids, could be implicated in the regulation of Diatoms (Lepe`re et al., 2006). However, the size of these organisms is well within the preferred range of food-particle size for larger Cladocera present in n and N, which could induce a decrease of fungal zoospores for example. However, Kagami et al. (2004) showed that Daphnia had a small positive effect on fungal infections by increasing the encounter rate between Fungi and host phytoplankton cells. The direct effects of nutrient addition were accompanied by cascade effects; in particular, the enclosures receiving high inputs were characterized by a significant increase in prokaryotic abundance. Others studies have reported that, in whole-lake and mesocosm experiments, prokaryotes responded strongly to the addition of nitrogen and phosphorus. We assume that a second cascading effect affects bacterivores, typically small heterotrophic eukaryotes o 5 mm (Sherr et al., 1989, 1991; Strom, 2000). Thus FEMS Microbiol Ecol 61 (2007) 483–495
493
Lacustrine small eukaryotes in hyper-eutrophic conditions
Centroheliozoa and Ciliates, which were especially represented in the high-nutrient-loading treatments, some taxa may have been favoured by the increase of bacterioplankton and eukaryotic picoplankton, which represent food resources for them (Hausmann, 2002). Viruses could also be implicated in these modifications of DOM availability by means of the release of organic matter during mass lysis (Weinbauer, 2004); however, we could not observe any relationship between these two parameters. The abundance of virus-like particles did not vary significantly from one treatment to another (data not shown), and, although viruses could be responsible for a considerable portion of the total mortality of eukaryotic phytoplankton (Suttle et al., 1990; Evans et al., 2003), we could not detect any direct link between viral abundance and SECC. This study, which represents the first phylogenetic analysis of small lacustrine eukaryotes subjected to manipulated top-down and bottom-up factors, raises new questions about the composition, regulation and role of this community. The changes that we observed in pigmented smalleukaryote abundance were expected following nutrient loading, but our results revealed that small heterotrophic eukaryotes are also significantly affected by this kind of manipulation. If the abundance of small heterotrophic eukaryotes was not significantly modified according to treatment, we clearly observed changes in the community composition of small heterotrophs. Our results showed that both resources and top-down factors are clearly involved in shaping the structure of small eukaryotes, not only for autotrophs but also heterotrophs, via complex interactions and trophic cascades within the microbial loop, notably in response to nutrient loading. These variations can lead to modifications that may impact the ecosystem functioning; as an example, among heterotrophs, a large proportion of small-eukaryote sequences are affiliated with parasitic lineages, and can participate in the control of plankton communities. It is likely that small eukaryotes, known as a particularly important planktonic component in oligotrophic systems, are probably also able to play an important role in hyper-eutrophic systems, and it is now essential to take into account precisely the structure of small eukaryotes and particularly the heterotrophic organisms in future studies and in food-web models.
Acknowledgements We would like to thank Se´ bastien Specel for automated sequencer and GENESCAN analysis, Jean-Claude Romagoux for his invaluable collaboration, and Maryline Basset for her participation in the mesocosm experiment. A. Thouvenot and D. Sargos are thanked for their help in microscopic counting. We also thank M.B. Britton for the English revision of the manuscript. FEMS Microbiol Ecol 61 (2007) 483–495
References Altschul SF, Madden TL, Sch¨affer AA, Zhang J, Zhang Z, Miller W & Lipman DJ (1997) Gapped BLAST and PSIBLAST: a new generation of protein database search programs. Nucleic Acids Res 25: 3389–3402. American Public Health Association (1992) Standards Methods for the Examination of Water and Wastewater, 18th edn. APHA, Washington, DC. Banse K (1992) Grazing, Temporal Changes of Phytoplankton Concentrations, and the Microbial Loop in the Open Sea. Plenum Press, New York, NY. pp. 409–440. Bell T & Kalff J (2001) The contribution of picophytoplankton in marine and freshwater systems of different trophic status and depth. Limnol Oceanogr 46: 1243–1248. Bonnet MP & Poulin M (2002) Numerical modelling of the planktonic succession in a nutrient-rich reservoir: environmental and physiological factors leading to Microcystis aeruginosa dominance. Eco Model 156: 93–112. Boucher D, Jardillier L & Debroas D (2006) Succession of bacterial community composition over two consecutive years in two aquatic systems: a natural lake and a lake-reservoir. FEMS Microb Ecol 55: 79–97. Caron DA (1983) Technique for enumeration of heterotrophic and phagotrophic nanoplankton, using epifluorescence microscopy, and comparison with other procedures. App Environ Microbiol 46: 491–498. Carrias JF, Amblard C & Bourdier G (1996) Protistan bacterivory in an oligomesotrophic lake: importance of attached ciliates and flagellates. Microb Ecol 31: 249–268. Cavalier-Smith T & Chao EE (2003) Molecular phylogeny of centrohelid heliozoan, a novel lineage of bikont eukaryotes that arose by ciliary loss. J Mol Evol 56: 387–396. Courties CA, Vaquer M, Trousselier J, Lautier M, Chr´etiennotDinet J, Neveux J, Machado C & Claustre H (1994) Smallest eukaryotic organism. Nature 370: 255. Dawson S & Pace N (2002) Novel kingdom-level eukaryotic diversity in anoxic environments. PNAS 99: 8324–8329. ´ D´ıez B, Pedros-Ali o´ C, Marsh TL & Massana R (2001) Application of denaturing gradient gel electrophoresis (DGGE) to study the diversity of marine picoeukaryotic assemblage and comparison of DGGE with other molecular techniques. Appl Environ Microbiol 67: 2942–2951. Domaizon I & D´evaux J (1999) Impact of moderate silver carp biomass gradient on zooplankton communities in a eutrophic reservoir. Consequences for the use of silver carp in biomanipulation. Comptes Rendus de l’Acad´emie des Sciences – Series III – Sciences de la Vie 322: 621–628. Dunbar J, Ticknor LO & Kuske CR (2001) Phylogenetic specificity and reproducibility and new method for analysis of terminal restriction fragment profiles of 16S rRNA genes from bacterial communities. Appl Environ Microbiol 67: 190–197. Evans C, Archer SD, Jacquet S & Wilson WH (2003) Direct estimates of the contribution of viral lysis and microzooplankton grazing to the decline of a Micromonas spp. population. Aquat Microb Ecol 30: 207–219.
c 2007 Federation of European Microbiological Societies Journal compilation Published by Blackwell Publishing Ltd. No claim to original French government works
494
Fogg GE (1995) Some comments on picoplankton and its importance in the pelagic ecosystem. Aquat Microb Ecol 9: 33–39. Gliwicz M (2004) Zooplankton. The Lakes Handbook (O’Sullivan PE, & Reynolds CS, eds), pp. 461–516. Blackwell Publishing, Oxford. Good IL (1953) The population frequencies of species and the estimation of population parameters. Biometrika 40: 237–264. Groisillier AR, Massana R, Valentin K, Vaulot D & Guillou L (2006) Genetic diversity and habitats of two enigmatic marine alveolate lineages. Aquat Microb Ecol 42: 277–291. Hansson LA, Gyllstrom M, Stahl-Delbanco A & Svensson M (2004) Responses to fish predation and nutrients by plankton at different levels of taxonomic resolution. Freshw Biol 49: 1538–1550. Hausmann K (2002) Food acquisition, food ingestion, and food digestion by protists. Jpn J Protozool 35: 85–95. Ibelings BW, De Bruin A, Kagami M, Rijkeboer M, Brehm GM & Van Donk E (2004) Review of host parasite interactions between freshwater phytoplankton and chytrid fungi (Chytridiomycota). J Phycol 40: 437–453. Izaguirre I, Mataloni G, Allende L & Vinocur A (2001) Summer fluctuations of microbial planktonic communities in a eutrophic lake–Cierva Point, Antarctica. J Plankton Res 23: 1095–1109. Jacquet S, Domaizon I, Personnic S, Duhamel S, Pradeep Ram AS, Heldal M & Sine-Ngando T (2005) Estimates of viralinduced vs. protozoan-induced bacterial mortality in Lake Bourget, France. Freshw Biol 50: 627–645. Kagami M, Van Donk E, De Bruin A, Rijkeboer M & Ibelings BW (2004) Daphnia can protect diatoms from fungal parasitism. Limnol Oceanogr 49: 680–685. Keeling PJ (2001) Foraminifera and Cercozoa are related in actin phylogeny: two orphans find a home? Mol Biol Evol 18: 1551–1557. Kemp PF & Aller JY (2003) Bacterial diversity in aquatic and other environments: what 16S rDNA libraries can tell us. FEMS Microb Ecol 47: 161–177. Lefranc M, Th´enot A, Lepe`re C & Debroas D (2005) Genetic diversity of small eukaryotes in lakes differing by their trophic status. Appl Environ Microbiol 71: 5935–5942. Lepe`re C, Boucher D, Jardillier L, Domaizon I & Debroas D (2006) Succession and regulation factors of small eukaryote community composition in a lacustrine ecosystem (Lake Pavin). Appl Environ Microbiol 72: 2971–2981. Li WKW (1994) Primary production of prochlorophytes, cyanobacteria, and eucaryotic ultraphytoplankton: measurements from flow cytometric sorting. Limnol Oceanogr 39: 169–175. Lo`pez-Garcia P, Rodriguez-Valera F, Pedro`s-Alio` C & Moreira D (2001) Unexpected diversity of small eukaryotes in deep-sea Antarctic plankton. Nature 409: 603–607. Lopez-llorca LV & Hernandez P (1996) Infection of the green alga Oocystis lacustris Chod with the Chytrid fungus
c 2007 Federation of European Microbiological Societies Journal compilation Published by Blackwell Publishing Ltd. No claim to original French government works
C. Lepe`re et al.
Diplochytridium deltanum (Masters) Karling. An SEM Study Micron 27: 355–358. Lorenzen CJ (1967) Determination of chlorophyll and pheopigments: spectrophotometric equations. Limnol Oceanogr 12: 343–346. Ludwig W, Strunk O, Westram R et al. (2004) ARB: a software environment for sequence data. Nucleic Acids Res 32: 1363–1371. Moon-van der Staay SY, De Wachter R & Vaulot D (2001) Oceanic 18S rDNA sequences from picoplankton reveal unsuspected eukaryotic diversity. Nature 409: 607–610. Moss B, Stephen D, Balayla DM, B´ecares E, Collings SE, Fern´andez-Al´aez C, Fern´andez-Al´aez M, Ferriol C, Garc´ıa P et al. (2004) Continental-scale patterns of nutrient and fish effects on shallow lakes: synthesis of a pan-European mesocosm experiment. Freshw Biol 49: 1633–1649. Persson L, Andersson G, Hamrin SF & Johansson L (1988) Predator regulation and primary production along the productivity gradient of temperate lake ecosystems. Complex Interactions in Lake Communities (Carpenter SR, ed), pp. 45–65. Springer-Verlag, Berlin. Pont D (1995) Le zooplancton herbivore dans les chaıˆnes alimentaires p´elagique. Limnologie g´en´erale (Masson, ed), pp. 515–540. Paris. Porter KJ & Feig YS (1980) The use of DAPI for identifying and counting aquatic microflora. Limnol Oceanogr 34: 943–948. Prepas E (1978) Sugar frosted Daphnia: an improved fixation technique for cladocera. Limnol Oceanogr 23: 557–559. Reckermann M & Veldhuis MJW (1997) Trophic interactions between picophytoplankton and micro-and nanozooplankton in the western Arabian Sea during the NE monsoon 1993. Aquat Microb Ecol 12: 263–273. Richards TA, Vepritskiy AA, Gouliamova DE & Nierzwicki-Bauer SA (2005) The molecular diversity of freshwater picoeukaryotes from an oligotrophic lake reveals diverse, distinctive and globally dispersed lineages. Environ Microbiol 7: 1413–1425. Romari K & Vaulot D (2004) Composition and temporal variability of picoeukaryote communities at a coastal site of the English Channel from 18S rDNA sequences. Limnol Oceanogr 49: 784–798. Ross PE & Nunavar M (1981) Preference for nanoplankton size fractions in lake Ontario zooplankton grazing. J Great Lakes res 7: 65–67. Samuelsson K & Andersson A (2003) Predation limitation in the pelagic microbial food web in an oligotrophic aquatic system. Aquat Microb Ecol 30: 239–250. Sherr E, Sherr B & Pedros-alio C (1989) Simultaneous measurement of rates of bacterioplankton production and protozoan bacterivory in estuarine water. Mar Ecol Prog Ser 54: 209–219. Sherr E, Sherr B, Hadas O & Berman T (1991) High abundance of pico-plankton-ingesting ciliates during late fall in lake Kinnert, Israel. J Plankton Res 13: 789–812.
FEMS Microbiol Ecol 61 (2007) 483–495
Lacustrine small eukaryotes in hyper-eutrophic conditions
Simek K, Bobkova J, Macek M & Nedoma J (1995) Ciliates grazing on picoplankton in a eutrophic reservoir during the summer phytoplankton maximum: a study at the species and community level. Limnol Oceanogr 40: 1077–1090. Staley JT & Gosink JJ (1999) Poles apart: biodiversity and biogeography of sea ice bacteria. Annu Rev Microbiol 53: 189–215. Stockner JG & Antia NJ (1986) Algal picoplankton from marine and freshwater ecosystems: a multidisciplinary perspective. Can J Fish Aquat Sci 43: 2472–2503. Stockner JG & Shortreed KS (1989) Algal picoplankton and contribution to food webs in oligotrophic British Columbia Lakes. Hydrobiologia 173: 151–166. Strom SL (2000) Bacterivory: interactions between bacteria and their grazers. Microbial Ecol Oceans, pp. 351–286. WileyLiss, New York. Suttle CA, Chan AM & Cottrell MT (1990) Infection of phytoplankton by viruses and reduction of primary productivity. Nature 347: 467–469. Ter Braak CJF (1986) Canonical correspondence analysis: a new eigenvector technique for multivariate direct gradient analysis. Ecology 67: 1167–1179. Thouvenot A, Debroas D, Richardot M, Jugnia LB & D´evaux J (2000) A study of changes between years in the structure of
FEMS Microbiol Ecol 61 (2007) 483–495
495
plankton community in a newly-flooded reservoir. Arch Hydrobiol 149: 131–152. Uterm¨ohl M (1958) Zur Vervollkommung der quantitativen phytoplankton-Methodick. Mitt Intern Verein Limnol 9: 1–38. Van Hannen EJ, Mooij W, van Agterveld MP, Gons HJ & Laanbroek HJ (1999) Detritus-dependent development of the microbial community in an experimental system: qualitative analysis by denaturing gradient gel electrophoresis. Appl Environ Microbiol 65: 2478–2484. Watson S, McCauley E & Downing JA (1997) Patterns in phytoplankton taxonomic composition across temperate lakes of differing nutrient status. Limnol Oceanogr 42: 487–495. Weinbauer MG (2004) Ecology of prokaryotic viruses. FEMS Microbiol Rev 28: 127–187. Weinbauer MG & H¨ofle MG (1998) Significance of viral lysis and flagellate grazing as factors controlling bacterioplankton production in a eutrophic lake. Appl Environ Microbiol 64: 431–438. Yuan J, Chen M, Peng S, Zhou H, Chen Y & Qu L (2004) Genetic diversity of small eukaryotes from the coastal waters of nansha islands in China. FEMS Microbiol Lett 240: 163–170. Zimmermann U, M¨uller H & Weisse T (1996) Seasonal and spatial variability of planktonic heliozoa in Lake Constance. Aquat Microb Ecol 11: 21–29.
c 2007 Federation of European Microbiological Societies Journal compilation Published by Blackwell Publishing Ltd. No claim to original French government works
