SEARCH

SEARCH BY CITATION

Keywords:

  • river;
  • periphyton;
  • community composition;
  • spatial variability

Abstract

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

Spatial variability in the microbial community composition of river biofilms was investigated in a small river using two spatial scales: one monitored the upstream–downstream pesticide contamination gradient, referred to as the ‘between-section variability’, and the other monitored a 100-m longitudinal transect (eight sampling sites per section) within each sampling section, referred to as the ‘within-section variability’. Periphyton samples were collected in spring and winter on artificial substrates placed in the main channel of the river. Denaturing gradient gel electrophoresis (DGGE) was used to assess the prokaryotic and eukaryotic community richness and diversity, and HPLC pigment analysis to assess the global taxonomic composition of the photoautotrophic community. In order to try to reduce the biological variability due to differences in flow velocity and in light conditions within each sampling section, and consequently to take into account only the changes due to water chemistry, nine plates (three per sampling section) subjected to similar physical conditions were chosen, and the results for these plates were compared with those obtained for all 24 plates. As shown by DGGE and by HPLC analyses, using these three substrate plates exposed to similar environmental conditions did indeed reduce the within-section variability and maximize the between-section variability. This sampling strategy also improved the evaluation of the impact of pollutants on the periphytic communities, measured using short-term sensitivity testing.


Introduction

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

Temporal variability and species in the composition of river periphyton is now well documented for both autotrophic (e.g. Stevenson, 1983; Johnson et al., 1997; Wellnitz & Rader, 2003) and heterotrophic communities (e.g. Jackson et al., 2001; Lyautey et al., 2005a). On the other hand, as for pelagic bacterial communities in lentic ecosystems (Dorigo et al., 2006), very few papers have dealt with the spatial variability of these communities at different scales, or with the sampling strategies that can be used to evaluate this parameter correctly.

This question of the spatial variability in the composition of benthic communities is particularly important in the overall context of evaluating the impact of pollutants in rivers. Indeed, one of the major difficulties in such studies is that of distinguishing the effects of pollutants on the structure and the composition of the communities from those of natural environmental factors and processes (Acuna et al., 2004; Ylla et al., 2007).

It has been definitely established that both physicochemical factors and biological factors influence the functioning and composition of periphytic communities in river ecosystems (e.g. Guasch et al., 2002). Temperature, light intensity and current velocity seem to be the main physical factors driving the biofilm structure and function (Biggs et al., 1998; Guasch et al., 1998; Guasch & Sabater, 1998; McCabe & Cyr, 2006), particularly in small rivers, but their impacts occur at different spatial scales. Differences in light intensity and in current velocity are observed at a small spatial scale (a few centimeters), whereas differences in water temperature generally occur at a larger scale (km).

The emergence of molecular methods in the past 10 years, particularly of fingerprinting methods such as denaturing gradient gel electrophoresis (DGGE), automated ribosomal spacer analysis or terminal-restriction fragment length polymorphism (T-RFLP), has now made it easier to determine and compare the species composition of periphytic communities (Amann et al., 1995; Dorigo et al., 2002; Szabo et al., 2008). With regard to bacterial diversity in biofilms, three recent papers by Lyautey et al. (2003, 2005a, b) have demonstrated the usefulness of 16S rRNA gene DGGE analysis for characterizing and comparing the composition of such communities in rivers. At the same time, the development of the pollution-induced community-tolerance (PICT) approach now makes it possible to estimate and compare the ability of natural communities to tolerate pollutants (Blanck et al., 1988; Bérard & Benninghoff, 2001; Dorigo et al., 2003; Schmitt-Jansen & Altenburger, 2005) within a complex ecosystem subjected to multiple stressors (pesticides, nutrients etc.). A combination of these approaches provides direct evidence of the impact of pollutants on periphytic communities, as illustrated, for example, in the recent paper by Dorigo et al. (2007).

The goals of the work reported here were (1) to evaluate, at different spatial scales, the variability of the composition of periphytic communities in a small river characterized by an upstream–downstream pollution gradient (Gouy & Nivon, 2006) and (2) to propose a field method suitable for distinguishing between the overall toxic effects of pesticides, and the natural variability associated with local environmental factors. With these objectives, we studied a small river flowing through a vineyard watershed, and characterized by an increasing upstream–downstream pesticide gradient (Dorigo et al., 2007), during two seasons (in spring, while pesticides were being applied, and in winter, when no pesticide was being used). A first sampling section was defined within the unpolluted upstream area, a second in the polluted middle area and finally a third section in the downstream sector of this river. Within each of the three sampling sections, eight artificial substrates were provided for periphyton colonization. The diversity of the bacterial and eukaryotic periphytic communities was estimated by 16S and 18S rRNA gene DGGE, respectively. In addition, characterization of the photoautotrophic communities (microalgae, including cyanobacteria) was performed by HPLC analysis of the photosynthetic pigments. Finally, we used PICT to estimate the short-term ability of these communities to tolerate a pollutant, diuron, that is regularly detected in this river.

Materials and methods

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

Description of the site investigated

This study was carried out on the Morcille river (latitude 46.150; longitude 4.600), which is a 7-km-long and shallow river (Strahler order 1), located within the Beaujolais vineyard area (Dorigo et al., 2007). In this region characterized by considerable wine-producing activity (vineyards occupy 70% of the catchment area), the routine application of pesticides to the vineyards from the end of March to the end of July contributes to an observed pesticide contamination gradient from the upstream sector of the river down to the mouth of the river (Gouy & Nivon, 2006; Montuelle et al., 2006). Three sampling sections with different levels of pesticide contamination were chosen: J (St Joseph: draining 0.2 km2 of vineyards), V (Versauds: draining 2.27 km2 of vineyards) and E (St Ennemond: draining 6.2 km2 of vineyards). J is located the furthest upstream, and E the furthest downstream.

Periphyton field sampling strategy

Periphyton was collected from artificial substrates in order to reduce the variability associated with natural substrates (Lowe et al., 1996; Cattaneo et al., 1997; Sabater et al., 1998) by controlling both the colonization surface of the substrate and the maturity of the biofilm. We used small frosted glass disks (1.5 cm2 surface area), which were glued onto 18 × 22 cm Plexiglas plates (4 mm thick) using a silicon sealant (Dorigo & Leboulanger, 2001). In each of the three sampling sections, eight Plexiglas plates holding c. 120 glass disks each were placed horizontally above the riverbed in the main lotic channel of the river, at equidistant points across a 100-m longitudinal transect (i.e. 24 plates overall). The decision to work only on the main lotic channel was motivated by the fact that we wanted to avoid the bank effects and sand deposits that can occur during low-water periods in lentic areas over the course of an experiment lasting 2 months. Plates were anchored to the riverbed using stainless steel stakes. At each sampling site, after 2 months of biofilm colonization, the plates were removed, placed in individual plastic bags filled with river water and transported in cool-boxes to the laboratory within 2 h. For the HPLC and DGGE analyses, the glass substrates were immediately frozen and stored at −80 °C before further processing. The whole sampling strategy was repeated twice, the first campaign lasting from mid-April to mid-June for the spring survey, and the second from February to March for the winter survey, in order to investigate contrasting biological and physico-chemical conditions (pesticides, water level, temperature, light, weather, etc.). In spring, two Plexiglas plates at the E sampling site were lost during a flood event.

Water sampling and collection of environmental data

The physical parameters, such as the water current, light intensity and temperature corresponding to each Plexiglas plate, were measured once a week at 10:00 hours throughout the 2-month incubation period. Alongside these analyses, 2 L water samples were collected in polyethylene bottles at each sampling section (J, V and E) and used to assess conventional chemical parameters, such as DOC, NO2, NO3, NH4+ and PO43−, conductivity and pH following the standard procedures and protocols (AFNOR, 1982). Additionally, at the end of the biofilm colonization period, water samples for analysis of organic pesticides were collected from each sampling section using clean, dark glass bottles. Three hundred and thirty three pesticides and degradation products were tested using standardized protocols. This was performed using a multidetection system of analysis comprising gas (GC/MS) and liquid chromatography (HPLC/MS/MS), and was carried out by the ‘Laboratoire Départemental d'Analyses de la Drôme’ (http://www.lda26.com/).

Periphyton DNA collection and extraction

In the laboratory, each biofilm sample was collected by scraping six glass disks per Plexiglas plate using a hard-bristled tooth brush (Patil & Anil, 2005). The material collected was suspended in 2 mL of 0.2-μm filtered river water, which had been kept at –80 °C before use. Biofilm suspensions were centrifuged at 14 000 g for 30 min, and the supernatant was discarded. Nucleic acid extraction was performed on the biofilm pellets according to Massana et al. (1997), with some modifications. Briefly, the pellets were exposed to 750 μL of the lysis buffer, and were processed as described in Dorigo et al. (2007). Purified DNA was precipitated and resuspended in 50 μL of TE (10 mM Tris, 1 mM EDTA, pH 8). The integrity of the total DNA was checked by agarose gel electrophoresis, and the nucleic acid concentration was determined using A260 nm.

Bacterial and eukaryotic DNA analyses by DGGE

PCR amplification of eukaryotic 18S rRNA gene fragments, and bacterial 16S rRNA gene fragments, and their DGGE analysis were performed according to Tlili et al. (2008). After migration, separated PCR products were stained for 45 min in the dark with SYBRGold (Molecular probes), visualized on a UV transilluminator (Claravision) and photographed (Scion Corporation camera). Digital images were then saved for further analysis using microsoft photo editor software, and finally processed as described in Dorigo et al. (2006). As it was impossible to load all 22 (spring) or all 24 (winter) samples on the same DGGE gel (maximum capacity of 20), and due to the difficulty of comparing samples run on different DGGE gels (low reproducibility for same PCR products on different DGGE gels), we had to adopt a testing strategy that would allow us to compare the community composition of these samples. Firstly, to compare the community composition of bacteria (or eukaryotes) in spring, all samples from J and V were run on one gel, and all those from E on a second gel. Samples giving the same profiles on both gels were identified, and in each group, a single sample was chosen to represent the rest of the group, and run on a third gel (the recapitulative gel). The same strategy was adopted for the winter samples. For each recapitulative gel (eukaryotic and bacterial), two tables (with samples as rows and DGGE bands as columns) were constructed: the first including all sampling sites, the second including only three selected sampling sites per section. The sites were selected as having experienced similar physical conditions throughout the colonization period. The presence or absence of a nucleic acid band at a given height in each lane was scored as 1 or 0, respectively. Matrices were used to perform correspondence analysis (COA) using ade-4 software (Thioulouse et al., 1997). We also performed a molecular analysis of variance (amova) using arlequin software (Excoffier et al., 2005). For the purposes of this analysis, we considered that the presence/absence of the different bands found in DGGE gels was equivalent to the presence/absence of bands in the RFLP approach. amova makes it possible to determine the relative percentage of variations in the DGGE band patterns of bacterial and eukaryotic communities between the three sampling sections (J, V and E) and within them, and the fixation index values (Fst) were estimated.

Pigment analysis by HPLC

For each biofilm sample, one glass disk per Plexiglas plate was selected and processed for HPLC pigment analysis as described in Dorigo et al. (2007). Each pigment was identified from its retention time and absorption spectrum, using Diode-Array Detection (DAD) according to SCOR (Scientific Committee on Ocean Research) methods (Jeffrey et al., 1997). Chlorophyll a was selected as the indicator of the total periphyton biomass (Bonin & Travers, 1992), which was expressed as μg chl a cm−2. A quantitative method was used, which results from a calculation model based on published ratios (rw), for monocultures (Wilhelm et al., 1991; Dorigo et al., 2003). The ratio was used as the 100% value in the calculation of the percentage contribution of each algal group (a.g.), where A is the area of the peak and d.p. the diagnostic pigment.

  • image

A table was constructed (with samples as rows and pigments as columns) by taking into account the relative abundance of each pigment in a given sample (expressed as a percentage of the sum of the area of all the pigments in a sample). Diatoms, cyanophytes and green algae were identified from their specific pigment signatures (Wilhelm et al., 1991).

Short-term evaluation of diuron tolerance and PICT assessment of phototrophic communities

The effects of diuron on periphyton were assessed using 14C photosynthetic assimilation as the endpoint (Guasch et al., 1998). A stock solution containing 100 μM diuron (MW 233) (Sigma high-grade standard 99.5%) was prepared in water, and stored at −20 °C before use. A semi-logarithmic series of concentrations was freshly prepared by serial dilution of the stock solution in 0.2-μm filtered river water. Final test concentrations in the test vessels ranged from 0 to 10 μM (one blank and nine increasing concentrations of diuron). The measurements of photosynthesis activity by 14C-carbonate incorporation were carried out as described in Dorigo & Leboulanger (2001). Data were fitted into a logistic equation (Seguin et al., 2001) using the least squares method, and used to plot a dose–response curve and determine photosynthetic EC50 values for each sampling area and period.

Results

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

Environmental conditions

When all the plates were taken into consideration, the median temperature ranged from 8.25 to 12.4 °C in spring, and from 0.7 to 3.7 °C in winter, with an increasing gradient from up- to downstream in spring, and the reverse in winter. During each season, the temperature was constant within each of the three river sections sampled (Table 1). Light conditions were mainly determined by the riparian canopy growth, and to a minor extent by the height of water above the substrates. The median light intensity measured under water at the level of the plates, ranged from 17.1 to 931 μmol photons m−2 s−1 in spring and from 54.2 to 475 μmol photons m−2 s−1 in winter. Significant between-section differences were only found in winter, with the highest intensity values recorded at J. The variation within the sections was particularly marked for this parameter (62–148% in spring and 23–94% of variation in winter, Table 1). The median water current ranged from 0.01 to 0.4 m s−1 in spring and from 0.05 to 0.78 m s−1 in winter, with no significant difference between the three river sections. The within-section variability ranged from 7% to 119% in spring and from 23% to 94% in winter (Table 1).

Table 1.   Average values of temperature, water current and light intensity recorded in spring and in winter for the three sampling sections: J (St Joseph), V (Versauds) and E (St Ennemond)
 T (°C)Water current (m s−1)Light intensity (μmol m−2 s−1)
  1. The values reported are the averages, and the percentage variation is shown in parentheses. The upper line in each row corresponds to the value found when all the plates were analyzed, and the lower line to that when only three per sampling section were analyzed.

  2. In bold font, we have highlighted the values where a significant reduction in the percentage variation was obtained when only three plates per sampling section were taken into account.

J (spring)9.0 (7.2)0.14 (81.1)267.53 (114.5)
9.1 (0)0.16 (31.3)83.4 (102.8)
V (spring)10.7 (0)0.12 (44.8)116.8 (148.1)
10.7 (0)0.09 (6.7)49.2 (42.7)
E (spring)11.7 (2.8)0.09 (79.8)72.8 (61.7)
11.9 (3.9)0.07 (55.1)82.5 (50.7)
J (winter)3.3 (18.7)0.15 (59.2)267.5 (94.2)
3.3 (10.3)0.26 (23.4)96.6 (68.6)
V (winter)2.4 (0)0.12 (32.2)116.8 (23.3)
2.4 (0)0.33 (12.7)118.6 (31.1)
E (winter)2.0 (33.1)0.09 (54.3)72.8 (28.3)
2.4 (0)0.38 (25.1)77.3 (28.2)

By considering the whole physical dataset for each season, we selected nine plates covering all sampling sections, with three per sampling section, and that shared similar physical water conditions, in order to analyze the influence of other parameters (chemical parameters) on the composition and structure of the biofilms (Fig. 1a: spring; Fig. 1b: winter). As the temperature was the same throughout each sampling section, the two main physical parameters recorded were water current and light intensity. Finally, more attention was paid to selected plates that had been exposed to similar water current values, because the light intensity values varied more rapidly during colonization. As a result, by considering three plates within each sampling section, which shared similar physical conditions, the within-section variability of water current and irradiance was reduced in spring and winter, respectively.

image

Figure 1.  Median light intensities (μmol m−2 s−1) vs. median water current (m s−1) during the (a) spring and (b) winter campaigns measured above each Plexiglas sampling plate. Plates originating from the sampling sections J, V and E are indicated by squares, triangles and circles, respectively. A subset of nine plates (three per sampling sections; open symbols) that shared similar conditions of light intensities and water current conditions was chosen among all the sampling plates.

Download figure to PowerPoint

The results for chemical parameters were described in Dorigo et al. (2007). Briefly, all chemical parameters, except pH and NO3, increased gradually from up- to downstream in both spring and in winter. Fifteen pesticides and 10 of their degradation products were detected and quantified in spring and in winter, in at least one sampling section. The sum of pesticide concentrations increased as expected from up- to downstream during both seasons, and the highest contamination was found at V and E in spring (up to 3.1 μg L−1 of pesticides, mainly herbicides) and at E in winter (up to 4.2 μg L−1, mainly herbicides).

DGGE analyses

Bacterial community

COA was performed using the presence/absence of bands within each of the samples in spring and in winter. In spring, the projection of all 22 sampling sites in the first plane defined by axes 1 (32% of variance) and 2 (20% of variance) showed a contrast on the first axis between samples taken from the pristine sampling section, J, and those from the two polluted sections (Fig. 2a). In winter, the projection of 24 sampling sites in the first plane (65% of variance), showed a contrast on the first axis between samples taken from section E, and those from the other sections (Fig. 2b). After choosing three homogeneous sampling sites per river section, the dispersion of the sampling points was reduced in each of the sections studied, and the three sampling sections were more clearly distinguished during both the spring and the winter campaigns (Fig. 2c and d, respectively).

image

Figure 2.  COA of bacterial DGGE bands of all sampling sites (a, b) in spring and (c, d) in winter taking into account all the sampling plates (a, c) or only three plates/sampling section displaying similar physical conditions (b, d). J, St Joseph; V, Versauds; E, St Ennemond.

Download figure to PowerPoint

These results were confirmed by the amova analysis. As reported in Fig. 3, it was found in both spring and in winter that choosing three plates sharing common hydrological and light intensity conditions per sampling section considerably reduced the within sampling section variance (E, J and V). In the same manner, higher Fst values (Fst corresponding to ‘between variability’ in this analysis) were observed after making this choice, which implies better differentiation between the three river sections. When three or eight plates per sampling station were considered, the between-section variation was almost higher than the within-section variation.

image

Figure 3.  Estimate of the relative percentage of variations in the DGGE band patterns of eukaryotic and bacterial communities, between the three sampling sections (J, V and E) and within them as estimated by amova using arlequin (Excoffier et al., 2005). Variations were calculated either for all the plates, i.e. all 24 (eight per sampling section) or the 16 remaining plates (after some plates had been lost during a flood event), or for the nine selected plates (three per sampling section) on the basis of light intensities and water current conditions.

Download figure to PowerPoint

Eukaryotic community

As for the prokaryotic communities, COA was performed using the presence/absence of bands within each of the samples in spring and in winter. The first two axes of each COA accounted for 47% (Fig. 4a) and 65% (Fig. 4b) of the variability in spring and in winter, respectively. In spring, the first axis distinguished between the pristine station J, on the one hand, and V and E on the other, whereas in winter J and V were very similar to each other, but differed from E. After choosing three homogeneous sampling sites per river section, the distinction between the three sampling sections was clearer both in winter (Fig. 4c and d, respectively) and in spring. COA analyses showed a general reduction of the variation of each sampling section when three out of eight plates were considered, even though E still displayed some dispersion of its sampling sites.

image

Figure 4.  COA of eukaryote DGGE bands at all sampling sites (a, b) in spring and (c, d) in winter taking all the sampling plates (a, c) or only three plates/sampling section exposed to similar physical conditions (b, d) into account. J, St Joseph; V, Versauds; E, St Ennemond.

Download figure to PowerPoint

The amova analysis confirmed the decrease in the within-section eukaryotic community variance after reducing the number of plates in spring (Fig. 3). On the other hand, no decrease in the within-section variance was found in winter. However, regardless of whether eight or three plates per sampling station were taken into consideration, the between-section variation was always higher than the within-section variation in both spring and in winter.

HPLC pigment analyses

Considering all plates per sampling section, HPLC analyses revealed 20 different pigments, eight of which cannot be identified from their measured absorbance spectra. The number of pigments/sample was very variable within a given river section. In winter, the average number of pigments decreased from up- to downstream (Kruskal–Wallis, P=0.028), whereas in spring no overall trend was observed. Chlorophyll a ranged from 0.1 to 2.8 μg cm−2 in spring (highest on average at V), and from 0.1 to 4.3 μg cm−2 in winter (highest on average at J), accounting on average for 54% of the total pigment in all periphytic samples in both spring and winter. The other dominant pigments in spring and in winter (mean >5%) were fucoxanthin and chl c, whereas violaxanthin was only abundant in winter. Fucoxanthin, the major xanthophyll in diatoms, was found to contribute up to 24% of total pigments in spring, and 25% in winter, revealing the dominance of these microalgae in phototrophic communities.

Using COA analysis, in spring, when all the plates were considered (Fig. 5b), it appeared that the pristine sampling section J was clearly distinguished from the more contaminated sections V and E (Fig. 5a), whereas in winter, the most contaminated section, E, was distinct from the less contaminated sections, V and J. By taking into account only three samples per section, a decrease in the within-section variability was obtained, as well as better discrimination between the three river sections, both in spring and in winter (Fig. 5c and d, respectively).

image

Figure 5.  COA of the percentage contribution of each pigment at all sampling sites (a, b) in spring and (c, d) in winter taking all the sampling plates (a, c) or only three plates/sampling section exposed to similar physical conditions (b, d) into account. J, St Joseph; V, Versauds; E, St Ennemond.

Download figure to PowerPoint

PICT measurements

In spring, on the basis of eight plates or only three plates per sampling section, the EC50 values, and thus community tolerance, displayed the same increase from up- to downstream, with values ranging from 5.3 to 63 μg L−1 and from 14 to 59 μg L−1 of diuron, respectively. The EC50 value for section J was significantly lower (Mann–Whitney U-test, P≤0.05) than that for sections V and E, with no statistical difference for either three or eight plates (Fig. 6). The confidence limits of the EC50 values were reduced for sections J and V, but not for station E when only three plates were taken in account.

image

Figure 6.  Sensitivity (expressed as EC50 in μg L−1 estimated from primary productivity measurements for periphyton samples over a range of diuron concentrations) from three sampling sections (J, V and E) on the River Morcille, either taking all the plates (gray bars) or three plates per sampling section (white bars) into consideration. Means±confidence limits (n=3 or n=8) of the EC50 values are shown. Confidence limits were calculated from SDs of replicate EC50 for spring.

Download figure to PowerPoint

Discussion

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

In the context of assessing the impact of pollutants on the diversity and the functioning of microbial communities in natural ecosystems, one of the most critical points remains the distinction between the relative effects of selective pressure resulting from pollution, and those that result from natural environmental factors and processes. Indeed, major natural variations can be found in the structure and function of nonimpacted communities when different ecosystems, or different sampling stations within the same ecosystem, are compared. In order to allow for these natural variations, multisampling strategies are generally used in order to estimate both the ‘within’ and the ‘between’ diversity, and thus to determine the real impact of a pollutant on the communities. However, multisampling strategies are expensive and time consuming, and are likely to generate very variable results, which are difficult to interpret. In this general context, we tested the effectiveness of a ‘two-level’ sampling strategy for evaluating the impact of pollutants on the composition of microbial communities in a small river. Initially eight sampling sites were studied within each sampling section with the intention of obtaining the best possible assessment of both the within- and between-section variability. We then selected three out of the sampling sites for each sampling section, which shared the same environmental conditions, with the intention of minimizing the within-section variability and maximizing the between-section variability.

Overall, our results showed that selecting sampling sites with the same environmental conditions reduces the within-section variability and increases the between-section variability. More precisely, it appeared that when all eight plates were taken into account for each sampling section, local environmental conditions led to considerable variability in the composition of the periphytic communities, as shown for example by COA and amova. Despite this variability within each sampling section, COA did distinguish between the three sampling sections, and hence did reveal the impact of pollutants on the periphytic communities. Selecting just three plates for each sampling section dramatically reduced the within-section variability, revealing the impact of light intensity and current velocity on the composition of periphytic communities, in the main stream of the river. It was no surprise to find that light and current had an impact on the composition of periphytic communities, as this indicated that a large proportion of the biomass in these periphytic communities was constituted by microalgae. Indeed, it had been demonstrated previously that water current influences various characteristics of biofilms within lotic ecosystems, such as their structure and composition (Stevenson, 1996), their biomass (Biggs, 1996; Lehtola et al., 2006) and the different life forms (mucilaginous, stalks/short filamentous or long filamentous diatoms) of benthic microalgae (Biggs et al., 1998), and it also affects the interaction between algae and grazers (Poff & Baker, 1997), and the effect of pollutants (Sabater et al., 2002). It has also been demonstrated that water current may stimulate nutrient assimilation (Lock & John, 1979; Stevenson & Glover, 1993), as well as photosynthesis and respiration (Whitford & Schumacher, 1961; Pfeifer & McDiffett, 1975). Perturbations linked to flood events not only control biomass but may also modify the specific composition of periphyton communities (Biggs & Thomsen, 1995; Van der Grinten, 2004), as a result of suspended sediment scour (Francoeur & Biggs, 2006). At lower water current levels (0.065–0.23 m s−1), equivalent to those reported during our spring campaign, Battin et al. (2003) highlighted changes in the architecture and function of biofilms. More precisely, they found that a decrease in the density and the thickness of the biofilm layer, and increases in chlorophyll a concentration and in bacterial density were linked to an increase in the water current velocity.

The availability of light depends not only on the climatic conditions but also on the stream width and depth, and on the riparian vegetation. Light intensity controls photosynthesis, and several authors have demonstrated that this process is promoted in the 30–400 μmol m−2 s−1 range, and inhibited above 500 μmol m−2 s−1 (Steinman & McIntire, 1987; Guasch & Sabater, 1995; Roberts et al., 2004). In our study, light intensity was higher in winter than in spring (less shade, as the riparian trees are deciduous), and varied between 17.1 and 931 μmol photons m−2 s−1. This parameter varied considerably in each sampling section, and depended mainly on the presence or absence of riparian canopy, but also on the meteorological conditions. Light intensity may influence the composition of the periphytic communities, for example, diatoms overrun other microalgal groups under shaded conditions (Mosisch et al., 2001), but also the functioning of river biofilms through its impact on photosynthesis, and thus on primary production (Dodds et al., 1996). It has also been pointed out that changes in light conditions may lead to changes in the algal pigment composition, in resistance to contaminants (Guasch & Sabater, 1995, 1998), and in the composition of algal communities (Van der Grinten, 2004).

While there was a decrease in the within-section variability of the prokaryote communities during both sampling seasons, for the eukaryote communities we only found a reduction of the within-section variability in spring (Fig. 3). It also emerged that the reduction in the within-section variability, when three plates out of the eight were chosen for each sampling section, was greater for prokaryote communities than for eukaryote ones. These two findings suggest that the two environmental factors (water current and light intensity) used to select the plates are likely to have a greater impact on the prokaryotic communities than on the eukaryotic ones. This outcome was clearly unexpected, because light, for example, would be expected to have a greater impact on photosynthetic than on heterotrophic microorganisms, unless the prokaryotic communities were dominated by cyanobacteria, which, like microalgae, are photosynthetic microorganisms. However, this hypothesis has to be rejected, because HPLC analyses did not identify the presence of a high cyanobacterial biomass. One possible explanation for the lack of reduction of the within-section variability in eukaryotic communities after choosing three plates in winter could be that environmental conditions in winter are less favorable for the development of eukaryotic communities, which could be expected to lead to marked differences between the sites. However, regardless of season, there was no significant difference in algal biomass as indicated by determining the chlorophyll a (Dorigo et al., 2007), which allows us to reject this hypothesis.

Application of the PICT concept allowed us to confirm that the differences found between the sampling sections in the composition of periphytic communities were mainly related to contamination by pollutants (Dorigo et al., 2007). Photosynthetic communities that had grown on the most contaminated areas exhibited the greatest tolerance of further contamination during short-term laboratory tests. The choice of three out of the eight plates for each sampling section did not significantly change the estimated EC50 values. Moreover, a reduction in the SD value was found for two sections (J and V), whereas no change was found for the third station (E). This finding is very interesting, because it has been demonstrated previously that community tolerance to a given pesticide may be influenced by environmental conditions, including the concentrations of nutrients such as phosphorus and nitrogen (Guasch et al., 1998, 2004), temperature (Bérard et al., 1999), light conditions (Guasch & Sabater, 1998) or water current (Sabater et al., 2002). Thus, the lack of any significant change in EC50 values after selecting three plates for each sampling section means (1) that reducing the within-section variability in the composition of the periphytic communities has not resulted in the selection of species that are more or less sensitive to diuron, and (2) that the impact of diuron did not depend on the natural variability of the physical parameters (light intensity and water current) in the sampling sections, at least in the range measured.

In conclusion, our study has revealed that choosing sampling sites sharing similar environmental conditions of light and current velocity allowed to reduce the within-section variability, and hence made it possible to obtain a better picture of the between-section variability, which is associated with the pollution gradient. From a practical point of view, this sampling strategy provides an unequivocal relationship between pollution exposure and biofilm response. However, this strategy provides a somewhat restricted view of the natural variability in the composition of periphytic communities at different spatial scales, because only the main stream of the river was taken into account and not, for example, the lentic areas. Thus, depending on the objectives of the study, two different sampling strategies (extensive sampling vs. targeted sampling) can be chosen. If the goal is to perform ecological studies of the functioning of a river, an extensive sampling strategy that takes almost all of the microhabitats into account in the river is required. On the other hand, in the context of ecotoxicological studies performed to assess the impact of pollutants, a more targeted sampling strategy that makes it possible to compare different sections in a river appears to be more appropriate than an extensive sampling strategy.

Acknowledgements

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

The authors would like to thank Annette Bérard for fruitful discussion and B. Motte and B. Volat for their support during field sampling and help in laboratory. They would also like to thank the Water Chemistry Laboratory of the Cemagref station in Lyon for pesticide and nutrient analysis. Monika Ghosh is acknowledged for revising the English version of the manuscript. The comments of the anonymous reviewers were greatly appreciated. The project was partially funded by the Région Rhône-Alpes (‘Appel d'offre Développement Durable’, contract no. 05 01880201), by the ECCO-Ecodyn Research Program (Program Périphyteau 2003–2006), and was supported by the LTER Zone Atelier du Bassin du Rhône (ZABR).

References

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References
  • Acuna V, Giorgi A, Munoz I, Uelingher U & Sabater S (2004) Flow extremes and benthic organic matter shape the metabolism of a headwater Mediterranean stream. Freshwater Biol 49: 960971.
  • AFNOR (1982) Essais des Eaux: Norme Française. NFT 90-023, AFNOR.
  • Amann RI, Ludwig W & Schleifer K-H (1995) Phylogenetic identification and in situ detection of individual microbial cells without cultivation. Microbiol Rev 59: 143169.
  • Battin TJ, Kaplan LA, Newbold DJ, Cheng X & Hansen C (2003) Effects of current velocity on the nascent architecture of stream microbial biofilms. Appl Environ Microb 69: 54435452.
  • Bérard A & Benninghoff C (2001) Pollution-induced community tolerance (PICT) and seasonal variations in the sensitivity of phytoplankton to atrazine in microcosms. Chemosphere 45: 427437.
  • Bérard A, Leboulanger C & Pelte T (1999) Tolerance of Oscillatoria limnetica Lemmermann to atrazine in natural phytoplankton populations and in pure culture: influence of season and temperature. Arch Environ Con Tox 37: 472479.
  • Biggs BJF (1996) Patterns in benthic algae of streams. Algal Ecology (StevensonRJ, BothwellML & LoweRL, eds), pp. 3156. Academic Press, San Diego.
  • Biggs BJF & Thomsen HA (1995) Disturbance of stream periphyton by perturbations in shear-stress – time to structural failure and differences in community resistance. J Phycol 31: 233241.
  • Biggs BJF, Goring DG & Nikora VI (1998) Subsidy and stress responses of stream periphyton to gradients in water velocity as a function of community growthform. J Phycol 34: 598560.
  • Blanck H, Wänkberg S-Å & Molander S (1988) Pollution-induced community tolerance – a new ecotoxicological tool. Functional Testing of Aquatic Biota for Estimating Hazards of Chemicals, Vol. 988 (CairsJrJ & PrattJR, eds), pp. 219230. ASTM STP, Philadelphia.
  • Bonin DJ & Travers M (1992) Examen critique des méthodes d'estimation de la biomasse et de l'activité des microorganismes dans les systèmes aquatiques. Mar Life 2: 129.
  • Cattaneo A, Kerimian T, Roberge M & Marty J (1997) Periphyton distribution and abundance on substrata of different size along a gradient of steam trophy. Hydrobiology 354: 101110.
  • Dodds WK, Hutson RE, Eichem AC, Evans MA, Gudder DA, Fritz KM & Gray L (1996) The relationship of floods, drying, flow and light to primary production and producer biomass in a prairie stream. Hydrobiology 333: 151159.
  • Dorigo U & Leboulanger C (2001) A PAM fluorescence-based method for assessing the effects of photosystem II herbicides on freshwater periphyton. J Appl Phycol 13: 509515.
  • Dorigo U, Bérard A & Humbert J-F (2002) Comparison of eukaryotic phytobenthic community composition in a polluted river by partial 18S rRNA gene cloning and sequencing. Microb Ecol 4: 372380.
  • Dorigo U, Bourrain X, Bérard A & Leboulanger C (2003) Seasonal changes in river microalgae sensitivity to atrazine and isoproturon, along a contamination gradient. Sci Total Environ 318: 101114.
  • Dorigo U, Fontvieille D & Humbert J-F (2006) Spatial variability in the abundance and composition of the free-living bacterioplankton community in the pelagic zone of Lake Bourget (France). FEMS Microbiol Ecol 58: 109119.
  • Dorigo U, Leboulanger C, Bérard A, Bouchez A, Humbert J-F & Montuelle B (2007) Lotic biofilm community structure and pesticide tolerance along a contamination gradient in a vineyard area. Aquat Microb Ecol 50: 91102.
  • Excoffier L, Laval G & Schneider S (2005) Arlequin ver. 3.0: an integrated software package for population genetics data analysis. Evol Bioinformatics Online 1: 750.
  • Francoeur SN & Biggs BJF (2006) Short-term effects of elevated velocity and sediment abrasion on benthic algal communities. Hydrobiology 561: 5969.
  • Gouy V & Nivon C (2006) Protection des eaux en Beaujolais viticole: caractérisation et suivi de la qualité des eaux sur le bassin versant de la Morcille sur la période 2001–2005. Rapport final d'étude, Cemagref, Lyon.
  • Guasch H & Sabater S (1995) Seasonal-variations in photosynthesis-irradiance responses by biofilms in mediterranean streams. J Phycol 31: 727735.
  • Guasch H & Sabater S (1998) Light history influences the sensitivity to atrazine in periphytic algae. J Phycol 34: 233241.
  • Guasch H, Ivorra N, Lehmann V, Paulsson M, Real M & Sabater S (1998) Community composition and sensitivity of periphyton to atrazine in flowing waters: the role of environmental factors. J Appl Phycol 10: 203213.
  • Guasch H, Paulsson M & Sabater S (2002) Effect of copper on algal communities from oligotrophic calcalreous streams. J Phycol 38: 241248.
  • Guasch H, Navarro E, Serra A & Sabater S (2004) Phosphate limitation influences the sensitivity to copper in periphytic algae. Freshwater Biol 49: 463473.
  • Jackson CR, Churchill PF & Roden EE (2001) Successional changes in bacterial assemblage structure during epilithic biofilm development. Ecology 82: 555556.
  • JeffreySW, MantouraRFC & WrightSW (eds) (1997) Phytoplankton Pigments in Oceanography. UNESCO, Paris.
  • Johnson RE, Tuchman NC & Peterson CG (1997) Changes in vertical microdistribution of diatoms within a developing periphyton mat. J N Am Benthol Soc 16: 503519.
  • Lehtola MJ, Laxandera M, Miettinena IT, Hirvonenc A, Vartiainenb T & Martikainenc PJ (2006) The effects of changing water flow velocity on the formation of biofilms and water quality in pilot distribution system consisting of copper or polyethylene pipes. Water Res 40: 21512160.
  • Lock MA & John PH (1979) The effects of flow patterns on uptake of phosphorus by river periphyton. Limnol Oceanogr 24: 376383.
  • Lowe RL, Guckert JB, Belanger SE, Davidson DH & Johnson DW (1996) An evaluation of periphyton community structure and function on tile and cobble substrata in experimental stream mesocosms. Hydrobiology 328: 135146.
  • Lyautey E, Teissier S, Charcosset JY, Rols JL & Garabetian F (2003) Bacterial diversity of epilithic biofilm assemblages of an anthropised river section, assessed by DGGE analysis of a 16 rDNA fragment. Aquat Microb Ecol 33: 217224.
  • Lyautey E, Jackson CR, Cayrou J, Rols J-L & Garabetian F (2005a) Bacterial community succession in natural river biofilm assemblages. Microb Ecol 50: 589601.
  • Lyautey E, Lacoste B, Ten-Hage L, Rols J-L & Garabetian F (2005b) Analysis of bacterial diversity in river biofilms using 16S rDNA PCR-DGGE: methodological settings and fingerprints interpretation. Water Res 39: 380388.
  • Massana R, Murray AE, Preston CM & DeLong EF (1997) Vertical distribution and phylogenetic characterization of marine planktonic Archaea in the Santa Barbara Channel. Appl Environ Microb 63: 5056.
  • McCabe SK & Cyr H (2006) Environmental variability influences the structure of benthic algal communities in an oligotrophic lake. Oikos 115: 197206.
  • Montuelle B, Gouy B, Roger MC et al. (2006) Evaluation de gains biologique et écologique associés à une réduction d'intrants polluants en milieu aquatique. Rapport final Cemagref–CPER 2003–2006, Région Rhône Alpes, 55p.
  • Mosisch TD, Bunn SE & Davies PM (2001) The relative importance of shading and nutrients on algal production in subtropical streams. Freshwater Biol 46: 12691278.
  • Patil JS & Anil AC (2005) Quantification of diatoms in biofilms: standardisation of methods. Biofouling 21: 181188.
  • Pfeifer RF & McDiffett WF (1975) Some factors affecting primary productivity of stream riffle communities. Arch Hydrobiol Suppl 75: 306317.
  • Poff NL & Baker NK (1997) Habitat heterogeneity and algal-grazers interactions in streams: explorations with a spatially explicit model. J N Am Benthol Soc 16: 263276.
  • Roberts S, Sabater S & Beardall J (2004) Benthic microalgal colonization in streams of differing riparian cover and light availability. J Phycol 40: 10041012.
  • Sabater S, Gregory SV & Sedell JR (1998) Community dynamics and metabolism of benthic algae colonizing wood and rock substrata in a forest stream. J Phycol 34: 561567.
  • Sabater S, Navarro E & Guasch H (2002) Effects of copper on algal communities at different current velocities. J Appl Phycol 14: 391398.
  • Schmitt-Jansen M & Altenburger R (2005) Predicting and observing responses of algal communities to photosystem II-herbicide exposure using pollution-induced community tolerance and species-sensitivity distributions. Environ Toxicol Chem 24: 304312.
  • Seguin F, Leboulanger C, Rimet F, Druart JC & Bérard A (2001) Effects of atrazine and nicosulfuron on phytoplankton in systems of increasing complexicity. Arch Environ Con Tox 40: 198208.
  • Steinman AD & McIntire CD (1987) Effects of irradiance on the community structure and biomass of algal assemblages in laboratory streams. Can J Fish Aquat Sci 44: 16401648.
  • Stevenson RJ (1983) Effects of current and conditions simulating autogenically changing microhabitats on benthic diatom immigration. Ecology 64: 15141524.
  • Stevenson RJ (1996) An introduction to algal ecology in freshwater benthic habitats. Algal Ecology (StevensonRJ, BothwellML & LoweRL, eds), pp. 330. Academic Press, San Diego.
  • Stevenson RJ & Glover R (1993) Effects of algal density and current on ion transport through periphyton communities. Limnol Oceanogr 38: 12761281.
  • Szabo K, Makk J, Kiss KT, Eiler A, Acs E, Toth B, Kiss AR & Bertilsson S (2008) Sequential colonization by river periphyton analysed by microscopy and molecular fingerprinting. Freshwater Biol 53: 13591371.
  • Thioulouse J, Chessel D, Dolédec D & Olivier JM (1997) Ade-4: a mulitvariate analysis and graphical display software. Stat Comput 7: 7583.
  • Tlili A, Dorigo U, Montuelle B, Margoum C, Carluer N, Gouy V, Bouchez A & Bérard A (2008) Responses of chronically contaminated biofilms to short pulses of diuron. An experimental study simulating flooding events in a small river running through a vineyard watershed. Aquat Toxicol 87: 252263.
  • Van der Grinten E (2004) Dynamic species interactions in phototrophic biofilms. PhD Thesis, Department of Aquatic Ecology and Ecotoxicology, FNWI, Université d'Amsterdam.
  • Wellnitz T & Rader RB (2003) Mechanisms influencing community composition and succession in mountain stream periphyton: interactions between scouring history, grazing and irradiance. J N Am Benthol Soc 22: 528541.
  • Whitford LA & Schumacher GJ (1961) Effects of current on mineral uptake and respiration by a fresh-water alga. Limnol Oceanogr 6: 423425.
  • Wilhelm C, Rudolph I & Renner W (1991) A quantitative method based on HPLC-aided pigment analysis to monitor structure and dynamics of the phytoplankton assemblage-A study from Lake Meerfelder Maar (Eifel, Germany). Arch Hydrobiol 123: 2135.
  • Ylla I, Romani AM & Sabater S (2007) Differential effects of nutrients and light on the primary production of stream algae and mosses. Fundam Appl Limnol 170: 110.