alpine lakes - Wiley Online Library

Jul 4, 2013 - Pearson correlation analysis) and also using a principal component analysis. ... times accounting for up to 56% (August) and 59% (June).
343KB taille 16 téléchargements 218 vues
RESEARCH ARTICLE

Temporal dynamics and structure of picocyanobacteria and cyanomyoviruses in two large and deep peri-alpine lakes phan Jacquet1 Xu Zhong1, Lyria Berdjeb1,2 & Ste 1

INRA, UMR CARRTEL, Thonon-les-Bains cx, France; and 2Institut des Sciences de la mer, Rimouski, QC, Canada

Correspondence: Stephan Jacquet, INRA, UMR CARRTEL, 75 Avenue de Corzent, 74203 Thonon-les-Bains cx, France. Tel.: +33 4 5026 7812; fax: +33 4 5026 0760; e-mail: [email protected] Received 25 January 2013; revised 5 June 2013; accepted 5 June 2013. Final version published online 4 July 2013. DOI: 10.1111/1574-6941.12166 Editor: Angela Sessitsch

MICROBIOLOGY ECOLOGY

Keywords lakes; picocyanobacteria; cyanomyovirus; structure; richness.

Abstract We conducted a 1-year survey of the surface waters of two deep peri-alpine lakes, and investigated the abundances and community structure of picocyanobacteria and co-occurring cyanomyophages. Picocyanobacterial abundances ranged between 4.5 9 104 and 1.6 9 105 cells mL 1 in Lake Annecy vs. 2.2 9 103 and 1.6 9 105 cells mL 1 in Lake Bourget. Cyanomyoviruses ranged between 2.8 9 103 and 3.7 9 105 copies of g20 mL 1 in Lake Annecy vs. between 9.4 9 103 and 9.4 9 105 copies of g20 mL 1 in Lake Bourget. The structures of picocyanobacteria and cyanomyoviruses differed in the two lakes, and a more pronounced dynamic pattern with greater seasonality was observed in Lake Bourget. At the annual scale, there was no relationship between cyanomyovirus and picocyanobacterial abundances or structures, but we could observe that abundances of the two communities covaried in spring in Lake Bourget. We showed that (i) the changes of picocyanobacteria and cyanomyoviruses were caused by the combined effect of several environmental and biological factors the importance of which differed over time and between the lakes, and (ii) the viral control of the picocyanobacterial community was probably relatively weak at the scale of the investigation.

Introduction The picophytoplankton, that is, small photosynthetic cells ranging between 0.2 and 2-3 lm in size, has been recognized as a key component of aquatic ecosystems during the last 30 years (Li, 1994; Partensky et al., 1999; Callieri, 2010). These cells can dominate the photosynthetic biomass, and can contribute to a significant percentage of the primary production, in both the oceans (Mann, 2003; Buitenhuis et al., 2012) and lakes (Callieri, 2010). Among the picophytoplanktonic prokaryotes, members of the picocyanobacteria (e.g. Prochlorococcus and Synechococcus) are abundant, diverse, and widespread worldwide in both marine and fresh waters (Partensky et al., 1999; Stockner et al., 2000; Coleman et al., 2006; Ahlgren & Rocap, 2012). In natural environments, Synechococcus spp. generally display clear seasonal and dynamic patterns in terms of abundance (Weisse, 1993; M€ uhling et al., 2005; Sandaa & Larsen, 2006; Tai & Palenik, 2009; Wang et al., 2011) and diversity (Palenik, 1994; M€ uhling et al., 2005; Callieri et al., 2012; Gilbert et al., 2012). Such patterns and ª 2013 Federation of European Microbiological Societies Published by John Wiley & Sons Ltd. All rights reserved

population succession have been shown to be controlled by changes in a variety of abiotic environmental variables (such as the nutrient supply, light level, and temperature), and biological factors such as heterotrophic nanoflagellate (HNF) grazing and viral lysis (Agawin et al., 1998; Baudoux et al., 2008; Callieri et al., 2012; Tsai et al., 2012). Viruses are a cause of mortality in phytoplankton, acting on population mortality, population succession, and community structuring, and they constitute a major driving force behind biogeochemical cycles in aquatic ecosystems (Fuhrman, 1999; Wilhelm & Suttle, 1999; Suttle, 2007). Typically, viral lysis could account for approximately 6–25% of phytoplankton biomass reduction in the marine environment (Suttle, 2005). So far, only one method has been developed to quantify simultaneously the impacts of both viral lysis and grazing on phytoplankton populations; this is the modified dilution method (Evans et al., 2003) assay that seeks to divide phytoplankton mortality into virus- vs. grazing-induced fractions in order to determine viral lysis rates in FEMS Microbiol Ecol 86 (2013) 312–326

313

Freshwater cyanomyophages and picocyanobacteria

natural phytoplankton. The method involves creating a gradient of both grazing and viral lysis by diluting with different proportions of grazer- and virus-free filtrate, and assessing the subsequent impact on phytoplankton growth rates. Although Weinbauer et al. (2011) suggested that the modified dilution method may not be suited to assess virus-induced mortality of the picocyanobacteria, a few studies have reported some mortality estimates for picocyanobacteria using this approach for both marine and fresh waters (Baudoux et al., 2007, 2008; Personnic et al., 2009b; Tsai et al., 2012). On average, picocyanobacterial mortality due to viral lysis was found to be < 10%, but it could locally reach up to 32% (Baudoux et al., 2008). Such a percentage is likely to have harmful consequences on the dynamics of the food webs in terms of organic matter transfer to higher trophic levels, but positive effects on nutrient recycling for bacteria and small phytoplankters (Wilhelm & Suttle, 1999). Some studies also showed that co-occurring cyanophages may modulate the picocyanobacterial community in temperate estuaries (Wang et al., 2011), lowlatitude temperate and tropical seawater (M€ uhling et al., 2005), and high-latitude coastal seawater (Sandaa & Larsen, 2006). Such aspects of cyanophage ecology constitute a key issue, but so far, only a limited number of studies have dealt with picocyanobacteria–cyanophage interactions, and this is particularly true for fresh waters, in particular in large and deep temperate lakes (Matteson et al., 2011). So far, all known cyanophages isolated from aquatic environments have been double-stranded DNA viruses and have been assigned to three families (Suttle, 2000; Mann, 2003; Liu et al., 2008; Sullivan et al., 2010; Dreher et al., 2011; Huang et al., 2012): the Myoviridae, the Podoviridae, and the Siphoviridae. The great majority of phages infecting marine picocyanobacteria (both Synechococcus and Prochlorococcus) to have been isolated reportedly belong to the Myoviridae family (Lu et al., 2001; Mann, 2003; Sullivan et al., 2008, 2010; Huang et al., 2012). In freshwater, a new phage infecting Synechococcus (e.g. S-CRM01) that was also a member of the Myoviridae was recently isolated by Dreher et al. (2011). The portal-protein-encoding gene g20 has been the molecular marker most frequently used to assess cyanomyovirus diversity (Fuller et al., 1998; Zhong et al., 2002; Frederickson et al., 2003; Dorigo et al., 2004; Short & Suttle, 2005), and also in three studies, their abundance (Sandaa & Larsen, 2006; Matteson et al., 2011, 2013). In French peri-alpine lakes (i.e. Lakes Annecy, Bourget, and Geneva), the phycoerythrin-rich (PE-rich) Synechococcus spp. community dominates the picophytoplankton (Duhamel et al., 2006; Personnic et al., 2009b; FEMS Microbiol Ecol 86 (2013) 312–326

S. Jacquet, unpublished data). In this study, we investigated both the abundance and community structure of picocyanobacteria and co-occurring cyanomyophages using flow cytometry, quantitative PCR (qPCR), and PCR-denaturing gradient gel electrophoresis (DGGE) in two large and deep European peri-alpine lakes (Lake Annecy, which is oligotrophic, and Lake Bourget, which is oligomesotrophic, Jacquet et al., 2012). Carried out in 2011, this 1-year study was intended to find out (i) whether the picocyanobacteria and cyanomyophage abundances and diversities were different in these two ecosystems, (ii) what are the main factors regulating the dynamics and structure of these communities, and (iii) whether cyanomyophages could be an important regulating factor of the picocyanobacterial community.

Materials and methods Sample collection and processing

Water samples were collected once or twice each month between January and November 2011 at reference stations in Lakes Annecy (GL) and Bourget (point B), corresponding to the deepest part of each lake. We obtained 14 samples from Lake Annecy and 18 from Lake Bourget. A sample of at least 21 L, integrating the water column from the surface down to a depth of 20 m, was collected using an electric pump and appropriate tubing, and the water was stored in a polycarbonate flask placed in darkness and at 4 °C until the filtration steps (performed only a few hours after sampling). To study cyanomyovirus abundance and richness, 20 L of this water was sequentially filtered through a 60-lm mesh, then through 142-mm-diameter, 1-lm pore-size filters (Millipore, Bedford, MA), and the < 1 lm filtered water was concentrated to a final volume of 200–250 mL, using a 30 000 molecular weight cutoff, spiral-wound Millipore ultrafiltration cartridge (regenerated cellulose, PLTK Prep/scale TFF, 1 ft2; Millipore). The < 1 lm fraction thus obtained was further filtered through 47-mm-diameter, 0.45-lm pore-size filters (Millipore) to retain all viruses, and to remove any remaining bacteria. This was checked using flow cytometry, as were all the steps in order to determine particle loss. Ten milliliter aliquots of viral concentrate (VC) was then stored at 20 °C for PCR-DGGE and qPCR applications (see below). To analyze the picocyanobacterial richness, a sample of 500 mL of the water was first filtered through a 47-mm-diameter, 5-lm pore-size filter, and then through a 0.22-lm pore-size filter. Immediately after filtration, the filters were stored at 80 °C. Each filter was subsequently used to isolate and purify DNA using the PureLink Genomic DNA Isolation Kit ª 2013 Federation of European Microbiological Societies Published by John Wiley & Sons Ltd. All rights reserved

X. Zhong et al.

314

(Invitrogen). This DNA was used as PCR template for picocyanobacterial richness investigation. Environmental parameters

Nutrient concentrations (total nitrogen, N-NH4 , N-NO3 , SiO2, P-PO4 , and total phosphorus) were measured at each station and date, according to the standard French AFNOR protocols (details are available at http://www. afnor.org). A conductivity-temperature depth measuring device (CTD Seabird SAB 19 Seacat profiler) and a chlorophyll fluorescence fluoroprobe (BBE Moaldenke, Germany) were used to obtain vertical profiles for water temperature, pH, conductivity, dissolved oxygen, and chlorophyll a concentrations. Using FCM to count picocyanobacteria and viruses

Autotrophic small eukaryotes, picocyanobacteria, heterotrophic bacteria, and viruses (from ambient water and/or following filtration/concentration steps, see below) were counted using a FACSCalibur flow cytometer (Becton Dickinson) equipped with an air-cooled laser providing 15 mW at 488 nm. Viruses were fixed with glutaraldehyde (0.5% final concentration, grade I; Merck) for 30 min in the dark, then diluted in 0.02 lm filtered TE buffer (0.1 mM Tris-HCL and 1 mM EDTA, pH 8), and incubated with SYBR Green I (at a final 5 9 10 5 dilution of the commercial stock solution; Molecular Probes), for 5 min at ambient temperature, followed by 10 min at 75 °C, and then another 5 min at room temperature, prior to FCM analysis (based on Brussaard, 2004 and modified from Personnic et al., 2009a). Based on their DNA fluorescence, we could discriminate two groups of viruses, designated virus-like particles (VLPs) 1 and 2 (Personnic et al., 2009a, b). Heterotrophic bacterial counts were also performed on fixed samples, but the samples were then diluted in a 0.02 lm filtered sample of the lake water, and incubated with SYBR Green I (10 4 dilution of the commercial stock solution) for 15 min. For photosynthetic cells (i.e. the picocyanobacteria and small eukaryotes), neither fixative nor fluorochrome was used. Analysis was thus carried out on fresh samples to which we added a suspension of 1-lm beads (Molecular probes), used as a size standard. Note that here the picocyanobacteria are represented by PE-rich Synechococcus spp (Personnic et al., 2009a; S. Jacquet, unpublished data). The flow cytometer list mode files obtained were then transferred and analyzed on a PC using the custom-designed software CYTOWIN (Vaulot, 1989). More details about the FCM analysis and data processing can be obtained elsewhere (Marie et al., 1999, 2000). ª 2013 Federation of European Microbiological Societies Published by John Wiley & Sons Ltd. All rights reserved

Bacteriophage counting by transmission electron microscopy

Bacteriophages (myoviruses, podoviruses, and siphoviruses), identified on the basis of their morphological structures, were counted in fixed samples with glutaraldehyde (0.25% final concentration, grade I; Merck) using transmission electron microscopy (TEM) as described elsewhere (Pradeep & Sime-Ngando, 2010). Using qPCR to count cyanomyovirus

qPCR was carried out on the VC to count the cyanomyoviruses using two primer sets: CPS1/2 (Fuller et al., 1998) and its nondegenerate primer set CPS4/5 (Wilson et al., 1999; without the 40-nucleotides GC-clamp at 5′ of CPS4). The 20 lL qPCR reaction contained 19 QuantiTect SYBR Green PCR Master Mix (Qiagen), 0.5 lM of each primer, and 2 lL of VC. The thermal cycling was conducted in a Rotor-Gene 3000 (Corbett Research) with the following program: 5 min denaturation at 95 °C, followed by 45 cycles of denaturation at 95 °C for 30 s, annealing for 30 s, and extension at 72 °C for 30 s. After testing various different conditions, the best annealing temperatures for the CPS1/2 and CPS4/5 primers were found to be 53 and 48 °C, respectively. Nine 10-fold serial diluted standards (ranging from 9 9 10 1 to 9 9 107 molecules) were run in duplicate along with two no-template control reactions containing 2 lL of nuclease-free water. The plasmid standards contained a cloned g20 gene (CPS1/2 or CPS4/5) amplified from environmental samples, which was linearized by a restriction digest of EcoR I, purified by the illustraTM GFXTM PCR DNA and Gel Band Purification Kit (GE healthcare), and quantified using a NanoDrop ND-1000 spectrophotometer (Thermo Scientific). The (correct) size of the amplicon (i.e. 165 bp) was checked using gel electrophoresis, and the melting curves also confirmed that the fluorescence signal corresponded to a single-sized DNA fragment. The qPCR amplification efficiency was 1.01 and 1.003 for cloned amplicons of CPS1/2 and CPS4/5 (with r > 0.99, n = 9), respectively. The calculation of concentrations was based on the assumption that cyanomyoviruses contained only one g20 gene copy per particle, following Sandaa & Larsen, (2006). PCR amplification and DGGE

To obtain most of the sequences (including the rarest and without interference from the GC-clamp) from the environment, prior to the DGGE analysis, a two-stage PCR was performed as proposed by Short & Suttle (2002). This consisted of a first-stage PCR on VC as template and with FEMS Microbiol Ecol 86 (2013) 312–326

315

Freshwater cyanomyophages and picocyanobacteria

the normal primer set (i.e. without the GC-clamp). The two-stage PCR was then performed on the product of the first-stage PCR using the GC-clamp-containing primer set (e.g. CPS1.1-GC and CPS 8.1), which was found to offer better DGGE resolution. For the PCRs, a DNA Thermal Cycler T-Professional (Biometra) was used to amplify the portal-protein-encoded gene g20 using the primer set CPS1.1/8.1 (Sullivan et al., 2008), and the picocyanobacteria-specific 16S rRNA gene using the primer set CYA359F/CYA781R(b) (Boutte et al., 2006). Briefly, for all primer sets, 25 lL of the reaction mix contained 19 PCR buffer, 4 mM MgCl2, 200 lM of each dNTP, 0.4 lM of each primer, 0.5 U of Platinium Taq DNA polymerase (Invitrogen), and 1 lL of VC. The typical program for the first-stage PCR was 15 min virion-lysing and denaturation at 95 °C, followed by 34 cycles of denaturation at 95 °C for 30 s, annealing for 30 s, extension at 72 °C for 30 s, and a final extension at 72 °C for 5 min. The program for second-stage PCR was 5 min denaturation at 95 °C, followed by 24 cycles of denaturation at 95 °C for 30 s, annealing for 30 s, extension at 72 °C for 30 s, and a final extension at 72 °C for 5 min. After various optimization tests, the annealing temperatures we used were 46 and 54 °C for CPS1.1/8.1 and CYA359F/CYA781R(b), respectively. The DGGE was conducted in 6% polyacrylamide gels with optimized linear denaturing gradient (100% denaturant is defined as 7 M urea and 40% deionized formamide). The linear denaturing gradient was at 25–50% and 45–70% for amplicons of CPS1.1/8.1 and CYA359F/CYA781R(b), respectively. The 200 ng DNA for all samples (with the exception of the January and February samples, which contained only 100 ng for g20 amplicons), was loaded from the second-stage PCR products into wells with 5 lL of 59 loading buffer [12.5% Ficoll, 25 mM Tris, 5 mM EDTA (pH 8.0), 0.5% SDS, 0.1% (w/v) xylene cyanol, and 0.1% (w/v) bromophenol blue]. Electrophoresis was carried out for 16 h in 19 TAE buffer (pH 7.4; 40 mM Tris-base, 20 mM sodium acetate, 1 mM EDTA) at 120 V, and a constant temperature of 60 °C using the CBS-DGGE 2000 system (C.B.S. Scientific Co., Inc.). Gels were stained in a 29 SYBR Green I (Molecular Probes, Invitrogen) solution for 45 min and were visualized on a UV transilluminator (Tex-35M; Bioblock Scientific), and photographed with GelDoc (BioRad). DGGE banding pattern analysis

The DGGE banding patterns were analyzed by considering both the presence and relative abundance of bands using the GelCompare II software package (Applied Maths, Kortrijk, Belgium), as described in Berdjeb et al. (2011). FEMS Microbiol Ecol 86 (2013) 312–326

Statistical analyses

The potential relationships among variables were investigated in each lake by linear pairwise correlations (i.e. Pearson correlation analysis) and also using a principal component analysis. The statistics were performed using XLSTATTM. Comparative analysis of DGGE fingerprints was carried out with the PRIMER 6 software (PRIMER-E, Ltd., UK). Hierarchical agglomerative clustering of Bray– Curtis similarities was performed, using the group average method in PRIMER software, to determine the relationships among sample profiles as representative of the cyanomyophage or picocyanobacterial community structure of each sample site. To test the null hypothesis that there was no significant difference between the groups discriminated according to the agglomerative clustering analysis, we conducted an analysis of similarities with the subroutine ANOSIM of PRIMER as detailed in Berdjeb et al. (2011). Mantel tests, applying Spearman’s rank correlation coefficient as the test statistic and 999 permutations, were used to test the null hypothesis of ‘no relationship between matrices’ in order to see whether the similarity matrix of the viral community was related to the matrices of the picocyanobacteria communities. To investigate the relationships between the community structures of cyanomyophages or picocyanobacteria and the measured environmental variables, a canonical correlation analysis (CCA) was performed using the software package XLSTAT-ADA. CCA generates an ordination plot that show the main pattern of variation in community structure as accounted for by the environmental variables measured. Various variables were then submitted to the forward selection procedure, in which the statistical significance of the term was tested by the unrestricted Monte Carlo permutation test (999 permutations). Explanatory variables with P-values > 0.05 were excluded from further analyses. The ordination axes obtained (based on community structure data) are linear combinations of environmental variables that best explain the community structure data for cyanomyophages or picocyanobacteria.

Results Environmental and biological characteristics

Table 1 reports the minimum, maximum, and mean values of the different variables monitored during this study. For both lakes, water temperature (Temp) in surface waters began to increase in late February, reaching its maximum in September, and then decreased until November. The average light level also began to increase in February, and peaked in March, June, and August. The ª 2013 Federation of European Microbiological Societies Published by John Wiley & Sons Ltd. All rights reserved

X. Zhong et al.

4.5 144.4 1.4 2 0.9 65 67 2.2 891 0.06 6.92 3.35 2.2 20.9 105 14 12 9 17.4 (September) 484.9 (June) 15.4 (April) 11 (August) 4 (September) 397 (March) 243 (March) 8 (October) 3503 (March) 0.32 (August) 33.9 (October) 15.6 (March) 10 (September) 85 (September) 367 (September) 54 (November) 46 (April) 49 (July) 5.2 (February) 29.7 (May) 9.5 (January) 4 (May) 1 (October) 167 (November) 10 (November) 1 (January) 907 (November) 0.11 (March) 5.76 (July) 4.45 (June) 2.75 (June) 4.3 (June) 2.79 (January) 10 (May) 10 (November) 16 (November)

12.6 233.3 11.5 5.9 2.5 284 111 4.6 2532 0.24 18.1 9.17 5.14 41.4 89.5 24 32 31

Max. (month) Min. (month)

4 122.7 1.6 5 2 109 96 16 762 3.1 7.6 4.36 5.04 13.6 208 11 3 13 5.5 (February) 12.6 (January) 7 (September) 7 (July) 2 (October) 333 (November) 175 (November) 2 (January) 790 (September) 1.47 (January) 13.8 (February) 0.22 (May) 3.5 (January) 20.6 (February) 9.37 (March) 19 (June) 9 (July–August) 18 (November) Temp (°C) Light (lE) Dissolved oxygen (mg L 1) Ptot (lg L 1) P-PO4 (lg L 1) Ntot (lg L 1) N-NO3 (lg L 1) N-NH4 (lg L 1) SiO2 (lg L 1) Chla (lg L 1) Hbact (105 cells mL 1) Syn (104 cells mL 1) VLP1 (107 part. mL 1) VLP2 (105 part. mL 1) Cyanomyo (103 part. mL 1) Myo (% phage) Podo (% phage) Sipho (% phage)

ª 2013 Federation of European Microbiological Societies Published by John Wiley & Sons Ltd. All rights reserved

Min., minimum; Max., Maximum; SD, standard deviation.

SD

Cyanomyovirus and picocyanobacterial abundances

12.5 227 9.5 12 5 537 318 16 1737 4.17 26.4 4.14 8.88 44.5 108 37 13 38

Mean Min. (month) Variables

Max. (month)

concentration of total nitrogen (Ntot) and N-NO3 were found to decrease from March. In Lake Bourget, the total phosphorus (Ptot) and P-PO4 decreased after peaking in early April and February respectively, while the opposite trend was observed in Lake Annecy as they started to increase in late summer. N-NH4 increased considerably in spring in Lake Bourget, while it was relatively stable in Lake Annecy. The average concentration of P and N nutrients was about 1.5 to 8 times higher in Lake Bourget than in Lake Annecy. The concentration of chlorophyll a (Chla) increased between early March and early April, and then remained at a low level in Lake Bourget and somewhat higher in Lake Annecy. In Lake Bourget, the TEM analysis revealed that the myoviruses (Myo) and siphoviruses (Sipho) were two important groups, sometimes accounting for up to 56% (August) and 59% (June) of the bacteriophages, respectively (Table 1). It is noteworthy, however, that in Lake Annecy, the podoviruses (Podo) could dominate the bacteriophage community on some occasions (e.g. in April with 46%).

16.9 (September) 404.6 (June) 12.8 (April) 20 (May) 10 (February) 758 (March) 510 (March) 57 (April) 3030 (February) 13.67 (April) 45 (April) 16 (August) 27.5 (May) 66.9 (May) 942 (September) 56 (August) 18 (April) 59 (June)

Lake Annecy Lake Bourget

Table 1. Environmental and biological variables, integrated over depths of 0–20 m, characterizing Lakes Bourget and Annecy from January to November 2011

Mean

SD

316

We quantified the cyanomyoviruses (Cyanomyo) using CPS1/2 and CPS4/5 primers. When using CPS1/2, concentrations were clearly lower than with CPS4/5 and cyanomyovirus abundances were on average about nineand four times lower in Lake Annecy and Lake Bourget, respectively. However, whatever lake and primer we used, a similar dynamic pattern was observed, with three major peaks in late May and June, and early September, respectively, so that a fairly close correlation of cyanomyovirus between CPS1/2 and CPS4/5 was reported for each lake (i.e. r = 0.87 for Lake Bourget, and r = 0.72 for Lake Annecy, at P < 0.01, Fig. 1). Estimated using CPS4/5, the average density of cyanomyoviruses was 1.2 times lower in Lake Annecy (9 9 104 copies of g20 mL 1) than in Lake Bourget (1.1 9 105 copies mL 1). The maximum cyanomyovirus densities found were 3.7 9 105 copies mL 1 in Lake Annecy and 9.4 9 105 copies mL 1 in Lake Bourget, which were both recorded in early September when the water temperature was also the highest. On average, picocyanobacterial abundance in Lake Annecy (9.2 9 104 cells mL 1) was > two times higher than that of Lake Bourget (4.1 9 104 cells mL 1; Fig. 2); however, the maximum cell density was recorded in Lake Bourget with 1.6 9 105 cells mL 1 in August (Table 1). In Lake Annecy, the picocyanobacterial abundance fluctuated between 4.5 and 15.6 9 104 cells mL 1 throughout the year. In contrast, in Lake Bourget, two distinct seasonal distributions of the picocyanobacterial abundance were observed in January–May and June–November. The FEMS Microbiol Ecol 86 (2013) 312–326

317

Freshwater cyanomyophages and picocyanobacteria

Copies of g20 ml–1 (qPCR CPS4/5)

1e+6

Community structures of picocyanobacteria and cyanomyoviruses

1e+5

1:1

1e+4

1e+3 1e+2

Lake bourget 1e+3

1e+4

1e+5

1e+6

Copies of g20 ml–1 (qPCR CPS1/2)

Copies of g20 ml–1 (qPCR CPS4/5)

1e+6

1e+5

1:1

1e+4

1e+3 1e+2

Lake annecy 1e+3

1e+4

1e+5

Copies of g20 ml–1 (qPCR CPS1/2) Fig. 1. Relationships between CPS1/2 and CPS4/5 obtained on several occasions during the 2011 survey of Lakes Bourget and Annecy.

picocyanobacteria displayed a low cell density (103) from January to early May, followed by a sharp increase in late May; the concentration then fluctuated around 104 until November. When we compared the abundances of the picocyanobacteria, VLPs, and cyanomyoviruses at the time scale of our investigation, we could not detect any positive or negative relationships between the different groups. In spite of slight covariation between cyanomyovirus and picocyanobacterial abundances in Lake Bourget in spring (from March to May), there was no significant relationship over the whole year (r = 0.01, n = 34).

FEMS Microbiol Ecol 86 (2013) 312–326

The community structures of picocyanobacteria and cyanomyoviruses were investigated using PCR-DGGE by targeting the picocyanobacteria-specific 16S rRNA (16S-pico) and g20 genes, respectively. In Lake Bourget, the number of DGGE bands for g20 ranged from 7 (February) to 17 (November), and 2.4% of bands were present in all the samples (see Supporting Information, Fig. S1). The number of DGGE bands for 16S-pico ranged from 12 (May) to 30 (June), and 1.8% of bands were present in all the samples. In Lake Annecy, the number of g20 DGGE bands was very similar to that in Lake Bourget, ranging from 7 (February) to 16 (November), but 18.5% of bands were present in all the samples. The number of 16S-pico DGGE bands ranged from 17 (January and April) to 27 (July), and 13.7% of bands were present in all the samples. For both lakes, we obtained a total of 52 DGGE bands of g20, 16 being common to both (Fig. S1). We also obtained a total of 77 DGGE bands of 16S-pico, with 31 common to both. Only one band for g20 and 16S-pico persisted in both lakes in all the samples, the others varying over time and differing in the two lakes. DGGE banding pattern analysis (Fig. 3) revealed that the picocyanobacterial structure of Lake Bourget clustered to form four distinct groups (ANOSIM, r = 0.93, P < 0.001), consisting of samples from winter to spring, early summer, summer to autumn, and early May, respectively. In Lake Annecy, the samples taken in spring, early summer, and late summer to winter were grouped together (ANOSIM, R = 0.84, P = 0.02). For the cyanomyoviruses, four clusters were discriminated (ANOSIM, R = 0.93, P < 0.1) in Lake Bourget containing samples from winter to spring, summer, from late summer to early autumn, and autumn, respectively. In Lake Annecy, there were only two distinct structures, one observed from January to September, and the other from October to November. A Mantel test indicated that the overall pattern in the community structure of cyanomyoviruses was not significantly correlated with changes in the structure of picocyanobacteria in either Lake Annecy (r = 0.052, P = 0.590) or Lake Bourget (r = 0.031, P = 0.73). Community structures of picocyanobacteria and cyanomyophages in relation to biotic and abiotic variables

The potential relationships between all variables were tested firstly by linear pairwise correlations (i.e. Pearson correlation analysis). As shown in Table 2, the abundance of cyanomyophages (CPS4/5) was not significantly related to any other variables in Lake Bourget or Lake Annecy at

ª 2013 Federation of European Microbiological Societies Published by John Wiley & Sons Ltd. All rights reserved

X. Zhong et al.

318

Bourget

1e+6

Annecy

Cells ml–1 or Part ml–1

(a)

(b)

1e+5

1e+4

1e+3 Jan

Picocyano CPS45 Mar

May

Jul

Sept

Nov Jan

(c)

Mar

May

Jul

Sept

Nov

(d)

Part ml–1

1e+8

1e+7

1e+6

VLP1 VLP2 Jan

Mar

May

Jul

Sept

Nov Jan

Mar

May

a threshold of P < 0.05. The abundance of picocyanobacteria in Lake Bourget was positively related to temperature (P < 0.01) and negatively related to NO3 (P < 0.01), Ntot, PO4, and SiO2 (P < 0.05). None of these relationships was detected for the picocyanobacteria in Lake Annecy. The output of the CCA revealed that more than 79% of the variance in the temporal dynamic structure of the cyanomyophages could be explained by 11 variables in Lake Bourget (Fig. 4), whereas 52% of this variance was explained by only 5 variables in Lake Annecy. Some variables (Syn, CPS45, Podo, Ptot, and PO4) seemed to be common in both lakes, and explained the temporal dynamic structure of the cyanomyovirus community, whereas some others seemed to be more specific to Lake Bourget (VLP, Hbact, NH4, Cyano, Myo, and Chla; Fig. 4). With regard to the temporal dynamics of the picocyanobacterial structure, CCA analysis indicated that the percentage of variance explained was almost the same, with 58.9% and 60.4% for Lakes Bourget and Annecy, respectively. However, eight variables (Syn, Myo, VLP, PO4, SiO2, Ntot, Ptot, and NH4) significantly explained the temporal dynamics of picocyanobacterial structure in Lake Bourget, compared with only six variables in Lake Annecy (Syn, PO4, VLP, SiO2, HBact, and NO3). ª 2013 Federation of European Microbiological Societies Published by John Wiley & Sons Ltd. All rights reserved

Jul

Sept

Nov

Fig. 2. Dynamics of cyanomyophages (estimated by qPCR using CPS4/5 primers) and Synechococcus spp. (FCM) in Lake Bourget (a) and Lake Annecy (b). Dynamics of VLP groups 1 and 2 (VLP1 or V1 and VLP2 or V2, following Personnic et al., 2009a) in Lake Bourget (c) and Lake Annecy (d).

Discussion The dynamics and structure of the cyanomyoviruses

One objective of this study was to analyze the temporal dynamics of the abundances of cyanomyovirus in perialpine lakes using real-time qPCR. To the best of our knowledge, only one study has previously reported these abundances in a lacustrine ecosystem. By employing the primer set CPS1/2, Matteson et al. (2011) reported for Lake Erie at various stations, depths (1–20 m), and seasons (February and August) cyanomyovirus concentrations ranging from 1.3 9 105 to 4.3 9 106 copies of g20 mL 1. When comparing estimates obtained with the same primer set, cyanomyovirus abundances of Lakes Annecy (2.77 9 102–3.13 9 104 g20 copies mL 1) and Bourget (4.15 9 103–1.03 9 105 g20 copies mL 1) were lower than those of Lake Erie, regardless of the depth, the location, or the period investigated. This finding may be explained by the higher Synechococcus host concentration reported in Lake Erie (Matteson et al., 2011). By comparing CPS1/2 and CPS4/5, we always found (i) a significant and positive correlation between abundances estimated by the two primer sets and (ii) that the FEMS Microbiol Ecol 86 (2013) 312–326

319

Freshwater cyanomyophages and picocyanobacteria

Bourget_g20

2-Mar 1-Mar 2-May Feb Jan 2-Apr 1-Apr 1-Jun Nov 2-Aug 1-Aug Oct 2-Sep 1-Sep Jul 2-Jun 1-May

Sampling

Sampling

Bourget_16S

0

0.2

0.4

0.6

0.8

Nov Oct 2-Sep 1-Sep 2-Aug 1-May Jan 2-May 2-Apr 2-Mar Feb 1-Apr 1-Mar 1-Aug 2-Jun 1-Jun Jul 0

1

0.2

Dissimilarity

Sampling

Sampling 0.4

0.6

0.8

1

0.8

1

Annecy_g20

2-Jul 2-Jun Sep Aug Oct Jan Mar Feb Nov 2-May 1-May Apr 1-Jul 1-Jun 0.2

0.6

Dissimilarity

Annecy_16S

0

0.4

0.8

1

Dissimilarity

2-Jun Jan 1-Jun 1-May 2-May 2-Jul Mar Feb Apr 1-Jul Sep Aug Nov Oct 0

0.2

0.4

0.6

Dissimilarity

Fig. 3. Clustering analysis of DGGE fingerprinting patterns of picocyanobacteria (targeted picocyanobacteria-specific 16S rRNA gene) and cyanomyovirus (targeted g20 gene) for Lakes Annecy and Bourget. Dendrograms were obtained by UPGMA clustering the Bray–Curtis dissimilarity values. May-01 and May-02 represent the 1st and 2nd samples of May, respectively. The letters a–d correspond to the labeled groups of samples discriminated at 60% dissimilarity.

nondegenerated primer set CPS4/5 provided higher estimates of cyanomyovirus abundances than the degenerated CPS1/2. The underestimation of cyanomyovirus abundance with CPS1/2 was probably due to the fact that the GC content in degenerated primers can affect primer annealing kinetics and therefore the accuracy of the qPCR assay on natural samples (Brankatschk et al., 2012). Indeed, CPS1/2 was unable to amplify some putative cyanomyoviruses at low concentrations (typically those present in the VLPs after FCM sorting) where CPS4/5 was successful (data not shown). Therefore, we used the cyanomyovirus estimates obtained from CPS4/5 for interpreting dynamic patterns and to perform statistical analyses. The temporal variability of cyanomyovirus abundance was detected in both these peri-alpine lakes, but it differed FEMS Microbiol Ecol 86 (2013) 312–326

in the two ecosystems at the annual scale. However, the patterns of the cyanomyovirus abundances did reveal some similarities, such as considerable viral production in May, June, and early September, for both lakes (Fig. 2). This production, which occurred at the same time in both ecosystems, was probably the result of increasing infection rates or burst sizes, which might be driven by common factors such as increases in light and/or temperature during these periods (Weinbauer, 2004; Parada et al., 2006; Thomas et al., 2011). Following the peaks of cyanomyovirus abundances, there was a sharp decline of abundances from July to August and in October, indicating that the virus decay rate could be greater than production at these periods. Decay rates for cyanophages have indeed been shown to be high in freshwater, typically in response to UVB in sunlight, but also to other ª 2013 Federation of European Microbiological Societies Published by John Wiley & Sons Ltd. All rights reserved

ª 2013 Federation of European Microbiological Societies Published by John Wiley & Sons Ltd. All rights reserved

1

Syn

1

Syn

0.21 1

Hbact

0.27 1

Hbact

0.03 0.46 1

VLP1

0.06 0.22 1

VLP1

0.22 0.53 0.83 1

VLP2

0.31 0.43 0.73 1

VLP2

0.25 0.26 0.34 0.38 1

CPS45

0.02 0.27 0.20 0.23 1

CPS45

0.30 0.51 0.61 0.70 0.21 1

Myo

0.04 0.04 0.38 0.56 0.27 1

Myo

0.03 0.50 0.46 0.52 0.15 0.73 1

Sipho

0.04 0.15 0.50 0.69 0.23 0.95 1

Sipho

0.32 0.33 0.53 0.53 0.13 0.80 0.23 1

Podo

0.12 0.38 0.40 0.30 0.06 0.04 0.29 1

Podo

0.05 0.42 0.17 0.03 0.29 0.01 0.23 0.14 1

Chla

0.41 0.54 0.17 0.47 0.38 0.24 0.29 0.19 0.42 1

0.27 0.26 0.32 0.26 0.13 0.23 0.41 0.06 0.31 1

0.33 0.67 0.34 0.32 0.37 0.38 0.46 0.20 0.74 0.38 1

0.01 0.39 0.15 0.06 0.35 0.23 0.11 0.39 0.79 0.30 0.76 1

0.20 0.29 0.00 0.19 0.12 0.51 0.19 0.54 0.56 0.36 0.20 0.62 0.04 0.77 0.25 0.16 1

Ntot

0.55 0.20 0.17 0.58 0.06 0.28 0.34 0.05 0.09 0.53 0.48 0.41 0.45 0.24 0.56 1

Ntot

0.24 0.60 0.02 0.03 0.50 0.04 0.20 0.05 0.43 0.04 0.37 0.37 0.26 0.14 0.51 1

PO4

0.50 0.47 0.53 0.82 0.18 0.38 0.44 0.24 0.06 0.68 0.86 0.37 0.85 0.42 1

PO4

0.45 0.06 0.10 0.05 0.02 0.24 0.09 0.37 0.32 0.10 0.04 0.34 0.09 0.55 1

Ptot

0.43 0.38 0.26 0.32 0.24 0.23 0.18 0.11 0.65 0.01 0.56 0.73 0.68 1

Ptot

0.33 0.22 0.27 0.36 0.15 0.78 0.39 0.84 0.37 0.06 0.06 0.67 0.19 1

SiO2

0.56 0.12 0.45 0.76 0.44 0.47 0.47 0.15 0.43 0.51 0.97 0.71 1

SiO2

0.06 0.34 0.21 0.04 0.10 0.18 0.00 0.40 0.60 0.19 0.36 0.19 1

Oxy

0.54 0.23 0.17 0.49 0.32 0.51 0.42 0.15 0.60 0.24 0.69 1

Oxy

Temp

0.66 0.30 0.43 0.76 0.41 0.47 0.47 0.14 0.30 0.65 1

Temp

Transp

Transp

Cyanos

0.24 0.53 0.02 0.01 0.07 0.26 0.13 0.27 1

Chla

Level of significance for values in italic and bold is P < 0.01, while the level of significance is P < 0.05 for values in italics only.

Syn Hbact VLP1 VLP2 CPS45 Myo Sipho Podo Chla Cyanos Transp Temp Oxy SiO2 Ptot PO4 Ntot NH4 NO3 Ph Cond Light

(b) Variables

Syn Hbact VLP1 VLP2 CPS45 Myo Sipho Podo Chla Transp Temp Oxy SiO2 Ptot PO4 Ntot NH4 NO3 Ph Cond Light

(a) Variables

Table 2. Pearson’s correlation coefficients (r) between the different variables in Lake Bourget (a) and Annecy (b)

0.05 0.07 0.06 0.09 0.25 0.15 0.07 0.35 0.22 0.32 0.30 0.50 0.16 0.39 0.28 0.05 0.12 1

NH4

0.41 0.35 0.16 0.00 0.14 0.11 0.00 0.15 0.45 0.16 0.21 0.59 0.24 0.69 0.10 0.09 1

NH4

0.20 0.09 0.29 0.35 0.29 0.63 0.16 T_0.82 0.59 0.18 0.21 0.70 0.05 0.90 0.53 0.25 0.75 0.34 1

NO3

0.68 0.18 0.33 0.76 0.34 0.49 0.45 0.01 0.46 0.53 0.91 0.82 0.87 0.52 0.68 0.46 0.22 1

NO3

0.06 0.37 0.24 0.04 0.37 0.18 0.27 0.19 0.61 0.06 0.33 0.20 0.68 0.17 0.09 0.51 0.04 0.11 0.07 1

Ph

0.07 0.60 0.11 0.06 0.10 0.16 0.08 0.12 0.47 0.26 0.01 0.68 0.12 0.60 0.22 0.08 0.701 0.18 1

Ph

0.30 0.07 0.29 0.39 0.25 0.72 0.30 0.78 0.59 0.03 0.24 0.78 0.09 0.93 0.56 0.26 0.72 0.45 0.89 0.08 1

Cond

0.44 0.11 0.26 0.53 0.31 0.12 0.15 0.05 0.05 0.54 0.68 0.54 0.67 0.52 0.59 0.59 0.27 0.60 0.18 1

Cond

0.4 0.33 0.38 0.36 0.02 0.03 0.1 0.09 0.23 0 0.27 0.49 0.12 0.34 0.64 0.54 0.06 0.41 0.19 0.1 0.39 1

Light

0.35 0.43 0.4 0.4 0.30 0.03 0.03 0.14 0.10 0.45 0.52 0.02 0.45 0.10 0.48 0.05 0.04 0.26 0.39 0.21 1

Light

320 X. Zhong et al.

FEMS Microbiol Ecol 86 (2013) 312–326

321

Freshwater cyanomyophages and picocyanobacteria

1

Jun-02

F2 (21.28 %)

N-NH4 Jul

Cyano VLP Sep-02

May-01 Chla

Cond

P-PO4

–1

Aug-02

–0.5

0

0.5

Jul-01 Feb

Jun-01 Sep-01 1

1.5

–1 –2

2

–1.5

–1

–0.5

0

1

(c)

Aug-02

Syn Oct

F2 (20.84 %)

Mar-02

Feb

Aug-01

VLP Nov

Jan

0

Sep-02 Jun-02

Sep-01 Apr-01Ntot

Jun-01

–0.5

1

1.5

2

Jun-01

N-NO3 SiO2 Feb

Jun-02

May-02 May-01 Jan

P-PO4 SiO2

(d)

0.5

Jul Myo

F2 (24.38 %)

May-02

0.5

F1 (36.56 %)

1

Mar-01

Oct

Mar

May-02

F1 (21.87 %)

0.5

Nov

April Jan

Podo

Myo Feb

–1.5

May-01

–0.5

Apr-01

Syn

Jul-02 0

CPS45

Mar-01 Mar-02

–1 –2

F2 (29.35 %)

Nov

Podo Hbact May-02

–0.5

CPS45

0.5

Jan

0

Ptot

Sep Jun-02

Syn

Aug

P-PO4

(b)

Oct

Apr-02

0.5

1

Aug-01

Jun-01

(a)

Mar

Jul-02

0 Apr

Hbact Oct

Jul-01

Aug

Syn

–0.5

Nov

Ptot

P-PO4

VLP

N-NH4 –1

–1

May-01

Sep Apr-02 –1.5 –2

–1.5 –1.5

–1

–0.5

0

0.5

1

1.5

2

F1 (31.29 %)

–2

–1.5

–1

–0.5

0

0.5

1

1.5

2

F1 (31.18 %)

Fig. 4. Canonical correspondence analysis (CCA) of the cyanomyophage and picocyanobacterial community structure from samples of Lake Bourget and Lake Annecy, using physico-chemical and biological variables. (a and b) Cyanomyoviruses in Lakes Bourget and Annecy; (c and d) picocyanobacteria in Lakes Bourget and Annecy.

factors (Liu et al., 2011; Hewson et al., 2012), and the low or nonexistent production during these periods could also have been due to unfavorable conditions (e.g. nutrient limitation for the host). However, using the linear pairwise correlations, we did not find any significant correlations between cyanomyovirus abundance and the density of their host or the other environmental parameters considered in this study (Table 2). This suggests that a large number of possible ecological mechanisms are responsible for the patterns of cyanomyovirus observed in these lakes. The PCR-DGGE analysis of the g20 cyanomyovirus structure revealed another considerable difference between the lakes. It is noteworthy that here the DNA of each representative g20 DGGE band was isolated and sequenced, revealing that 63% of the g20 sequences were grouped into FEMS Microbiol Ecol 86 (2013) 312–326

the culture-containing cluster II, as defined by Sullivan et al. (2008), the other 37% being grouped into different environmental-sequences-only clusters, the details of which (i.e. the phylogenetic analysis) are provided elsewhere (X. Zhong & S. Jacquet, in revision). Two-thirds of the DGGE bands varied along the course of the year, and only 2% persisted throughout the year. Four shifts in the g20 cyanomyovirus structure were recorded in Lake Bourget against only two in Lake Annecy (Fig. 3), which suggests a high dynamism of the g20 cyanomyovirus community in Lake Bourget. This result was confirmed from the CCA analysis, which revealed that the number of environmental variables implicated in g20 cyanomyovirus structuring in Lake Bourget was double that in Lake Annecy. We do not know of any comparative study for fresh waters that we could refer to in order to see which biological or ª 2013 Federation of European Microbiological Societies Published by John Wiley & Sons Ltd. All rights reserved

X. Zhong et al.

322

environmental parameters could regulate the cyanomyophage structure. Important common factors that could explain the cyanophage community structure in the two ecosystems investigated here include phosphorus concentrations, but also the abundances of picocyanobacteria and cyanomyophages, supporting the expected link between host abundance, specific viral abundance, and diversity. Picocyanobacterial dynamics and structure

As for most temperate lakes, maximum picocyanobacterial abundances were recorded in spring and summer/fall, conforming to a bimodal pattern, attributed mainly to spring/summer stratification and relatively low deep mixing in fall (Weisse, 1993; Stockner et al., 2000). In the oligotrophic Lake Annecy, the picocyanobacterial abundance was on average higher than in the oligo-mesotrophic Lake Bourget, but the highest cell densities and greater yearround variability were observed in the latter. This was only expected for oligo- and mesotrophic lakes, and it has been explained elsewhere (Stockner et al., 2000; Callieri, 2010). In Lake Bourget, the picocyanobacterial abundance was positively related to water temperature (r = 0.66, P < 0.01), and this physical factor is known to be important in the regulation of this community (Agawin et al., 1998; Callieri, 2010). P and N (except for NH4) were found to be negatively related to picocyanobacterial abundance, which also suggests the importance of resource availability in the dynamics of this community, in Lake Bourget. Surprisingly, in Lake Annecy, no significant correlation was found between picocyanobacterial abundance and the different abiotic parameters considered in this study (e.g. nutrients), suggesting a weak bottom-up control on picocyanobacteria abundance in this lake, at least at the annual scale. It seems, however, that NO3 could have played an important role in the picocyanobacterial dynamics and community structuring from January to June (Fig. 4). This finding is interesting, because the relationship between picocyanobacteria and the N source is not well understood in freshwaters, and the capacity of the picocyanobacteria to use either NH4 or NO3 and/or to switch from one resource to another, in relation to their diversity, could be considered. Indeed, it has been reported that NH4 is the preferred N source of Synechococcus in culture but that, when depleted, some strains can also take up NO3 (Bird & Wyman, 2003). It is noteworthy here that grazing by HNFs (Pernthaler et al., 1996; Callieri et al., 2002; Becker et al., 2007; Tsai et al., 2012), ciliates (Simek et al., 1995; Callieri et al., 2002), rotifers (Izaguirrel et al., 2003), or microcrustaceans (Weisse, 1993) as well as viral lysis (Personnic et al., 2009a, b) not considered in this study could be important processes controlling these picocyanobacteria. ª 2013 Federation of European Microbiological Societies Published by John Wiley & Sons Ltd. All rights reserved

In both lakes, the community structure of picocyanobacteria could remain similar for only 15 days or up to 4 months (Fig. 3). The various shifts operated at a short timescale (i.e. 15 days) and were monitored between May and June in Lake Bourget, and between June and July in Lake Annecy. To the best of our knowledge, there are only a limited number of studies that looked at picocyanobacterial community structure in lakes (see Callieri, 2010 for a review), and this is even truer for large and deep alpine lakes. Callieri et al. (2012) recently reported that both pH and temperature had a significant influence on the picocyanobacterial structure in Lake Maggiore. Our statistical analysis did not pick out the importance of these factors in this study, which revealed the rather complex influence of biotic and abiotic parameters, which could explain around 60% of the dynamics of the community structure of picocyanobacteria, in both lakes. This finding reminded us of the importance of inorganic nutrients that have already been found to explain the structure of both the Bacteria and the Archaea in these lakes (Berdjeb et al., 2011, 2013). As an example, NH4 was characterized by a tremendous enrichment beginning in March and ending in June, and peaking in late April (data not shown), a phenomenon known to be associated with zooplankton excretion at this time of year (Jacquet et al., 2011), and this nutrient was clearly implicated in picocyanobacterial community structuring in Lake Bourget, but not in Lake Annecy. In the end, we did not obtain clear evidence that the cyanomyoviruses could explain the dynamics of abundance and the community structure of picocyanobacteria when data were considered over the whole year, despite the fact that the community structure of the cyanomyoviruses could also display a seasonal dynamic pattern, as discussed above. It is noteworthy that only a few studies have shown that cyanophage infection can exert a major influence on picocyanobacterial succession (see below). Cyanomyovirus–picocyanobacteria relationships

Whereas some previous studies reported covariation between Synechococcus and cyanophage abundances (Sandaa & Larsen, 2006; Wang et al., 2011), suggesting tight coupling between the two communities, at the timescale of our investigation, there was no correlation between the abundances of cyanomyoviruses and picocyanobacteria in either lake. A similar lack of relationship was also reported by Matteson et al. (2011), who concluded that, as these host–virus interactions are likely to occur across large temporal and spatial scales, their low-frequency sampling strategy probably hindered their ability to tease these relationships apart. In most cases, FEMS Microbiol Ecol 86 (2013) 312–326

323

Freshwater cyanomyophages and picocyanobacteria

cyanomyovirus abundances did not correlate with the abundance of the picocyanobacteria, suggesting a weak relationship and low viral impact on the temporal dynamic of picocyanobacterial abundances (Fig. 2). Such weak viral impact on Synechococcus has also been previously reported in several studies (Waterbury & Valois, 1993; Garza & Suttle, 1998; Baudoux et al., 2007; Personnic et al., 2009a), and several explanations have been provided. Firstly, a certain virus–host abundance threshold could be required to trigger viral lysis, as has already been proposed by Jacquet et al. (2002) and Personnic et al. (2009b). Secondly, the dominant part of the Synechococcus community could also consist of cells that are resistant to viral attack (Waterbury & Valois, 1993; Lennon et al., 2007; Lennon & Martiny, 2008), and finally, unfavorable conditions, such as oligotrophic conditions or nutrient depletion in surface waters, in summer for instance, may have initiated the switch from a lytic to lysogenic lifestyle (Wilson & Mann, 1997). Lysogeny is an effective strategy for maintaining the phage population when host abundance is too low or metabolically inactive to maintain the viral population (Mann, 2003), and such behavior has also been identified in freshwater cyanophages (Dillon & Parry, 2008) and in the bacteriophages of Lake Bourget (Thomas et al., 2011). In Lake Bourget, we were not able to demonstrate any direct causality between changes in the picocyanobacterial structure and virus-induced mortality, because we did not perform any laboratory experiments, but our results did suggest, however, that a ‘kill-the-winner’ pattern could be involved in determining the abundance or decline of the most competitive Synechococcus OTUs, as reported by M€ uhling et al. (2005) in a study carried out in the Gulf of Aqaba (Red Sea). The DGGE profiles for the picocyanobacteria did indeed reveal changes in the fluorescence intensity of some bands (Fig. S2), which was characterized by an increase in March, a marked decrease in late April culminating in a disappearance in early May, before being slightly restored in late May. For the cyanomyovirus group, two bands were greatly enriched, and three bands emerged concomitantly with viral lysis. These cyanomyovirus bands seemed to disappear in June, when the NH4 level decreased and picocyanobacteria displayed lower abundance. These results, plus the production of NH4 when both the picocyanobacteria and cyanomyoviruses covaried, suggest that NH4 might be a key factor involved in the tight interaction between cyanomyoviruses and picocyanobacteria during spring (Shelford et al., 2012). We think that NH4 could promote rapid growth of NH4-dependent picocyanobacteria that soon reach a certain level of concentration, triggering viral lysis (Thingstad & Lignell, 1997; M€ uhling et al.,

FEMS Microbiol Ecol 86 (2013) 312–326

2005; Personnic et al., 2009a). This hypothesis requires experimental evidence to confirm it.

Conclusions French peri-alpine Lakes are characterized by the temporal dynamics of cyanomyovirus abundances and community structures. At the annual scale, no clear relationship was found with the picocyanobacteria, suggesting at first sight that cyanomyovirus had only a slight impact on the temporal dynamics of the picocyanobacteria. Nevertheless, during some periods of the year (e.g. in spring), a synchronous dynamic pattern was observed between some cyanomyoviruses and picocyanobacteria in the mesotrophic Lake Bourget, suggesting that specific interactions occur between these communities. Our difficulty in demonstrating a clear and obvious interaction between these two communities could be due to our relatively low-frequency sampling strategy, which could have obscured relationships between picocyanobacteria and cyanophages occurring at shorter time and spatial scales (i.e. over a timescale of weeks and at specific depths; Marston & Sallee, 2003). Furthermore, the complexity of the parameters implicated in the temporal patterns of both cyanomyoviruses and picocyanobacteria, as revealed by statistical analysis, could make it trickier to identify the relationship between them in the lakes we were studying. Higher frequency of sampling, at various depths, will be required to detect unambiguously more subtle relationships between the abundances and genetic composition of viral and picocyanobacterial assemblages in these ecosystems. Also, before we can draw any conclusions about the role that cyanophages play in the picocyanobacteria assemblage, we would have to look at host–virus interactions at a finer scale. These viruses can be very hostspecific, and it is possible that they have less impact on the overall host abundance, than on the strain succession, richness, and/or intraspecies diversity of their host(s). Lastly, it would be helpful to consider the role of cyano-podoviruses and siphoviruses, which may play a significant role in controlling the picocyanobacterial assemblage.

Acknowledgements We would like to thank Jean-Christophe Hustache, Pascal Chifflet, and Pascal Perney for technical assistance and help with sampling. Yves Desdevises and Laure Guillou are greatly acknowledged for their reading, and criticism of a former version of the manuscript. X.Z. was granted by Region Rh^ one-Alpes. Monika Ghosh is gratefully acknowledged for improving the English of the manuscript. Anonymous reviewers helped us to build this article.

ª 2013 Federation of European Microbiological Societies Published by John Wiley & Sons Ltd. All rights reserved

324

Statement By analyzing freshwater cyanomyophage community structure and abundance in the surface waters of two large and deep peri-alpine lakes over a complete year, this study suggests the potential importance of these viruses in controlling picocyanobacterial dynamics and diversity at some periods of the year and highlights the importance of the temporal-scale resolution to unveil phage–host relationships.

References Agawin NSR, Duarte CM & Agusti S (1998) Growth and abundance of Synechococcus sp. in a Mediterranean Bay: seasonally and relationship with temperature. Mar Ecol Prog Ser 170: 45–53. Ahlgren NA & Rocap G (2012) Diversity and distribution of marine Synechococcus: multiple gene phylogenies for consensus classification and development of qPCR assays for sensitive measurement of clades in the ocean. Front Microbiol 3: 213. Baudoux AC, Veldhuis MJW, Witte HJ & Brussaard CPD (2007) Viruses as mortality agents of picophytoplankton in the deep chlorophyll maximum layer during IRONAGES III. Limnol Oceanogr 52: 2519–2529. Baudoux AC, Veldhuis MJW, Noordeloos AAM, Noort G & Brussaard CPD (2008) Estimates of virus- vs. grazing induced mortality of picophytoplankton in the North Sea during summer. Aquat Microb Ecol 52: 69–82. Becker S, Richl P & Ernst A (2007) Seasonal and habitat related distribution pattern of Synechococcus genotypes in Lake Constance. FEMS Microbiol Ecol 62: 64–77. Berdjeb L, Ghiglione JF & Jacquet S (2011) Bottom-up vs. topdown factors regulating the bacterial community structure in two peri-alpine lakes. Appl Environ Microbiol 77: 3591–3599. Berdjeb L, Pollet T, Chardon C & Jacquet S (2013) Spatiotemporal changes in the structure of archaeal communities in two deep freshwater lakes. FEMS Microbiol Ecol DOI:10.1111/1574-6941.12154. Bird C & Wyman M (2003) Nitrate/nitrite assimilation system of the marine cyanobacterium Synechococcus sp. strain WH8103: effect of nitrogen source and availability on gene expression. Appl Environ Microbiol 69: 7009–7018. Boutte C, Grubisic S, Balthasart P & Wilmotte A (2006) Testing of primers for the study of cyanobacterial molecular diversity by DGGE. J Microbiol Methods 65: 542–550. Brankatschk R, Bodenhausen N, Zeyer J & B€ urgmann H (2012) Simple absolute quantification method correcting for quantitative PCR efficiency variations for microbial community samples. Appl Environ Microbiol 78: 4481–4489. Brussaard CPD (2004) Optimization of procedures for counting viruses by flow cytometry. Appl Environ Microbiol 70: 1506–1513.

ª 2013 Federation of European Microbiological Societies Published by John Wiley & Sons Ltd. All rights reserved

X. Zhong et al.

Buitenhuis ET, Li WKW, Vaulot D et al. (2012) Picophytoplankton biomass distribution in the global ocean. Earth Syst Sci Data 5: 221–242. Callieri C (2010) Single cells and micro-colonies of freshwater picocyanobacteria: a common ecology. J Limnol 69: 257– 277. Callieri C, Karjalainen SM & Passoni S (2002) Grazing by ciliates and heterotrophic nanoflagellates on picocyanobacteria in Lago Maggiore, Italy. J Plankton Res 24: 785–796. Callieri C, Caravati E, Corno G & Bertoni R (2012) Picocyanobacterial community structure and space-time dynamics in the subalpine Lake Maggiore (N. Italy). J Limnol 71: 95–103. Coleman ML, Sullivan MB, Steglich C, DeLong EF & Chisholm SW (2006) Genomic Islands and the ecology and evolution of Prochlorococcus. Science 311: 1768–1770. Dillon A & Parry JD (2008) Characterization of temperate cyanophages active against freshwater phycocyanin-rich Synechococcus species. Freshw Biol 43: 1253–1261. Dorigo U, Jacquet S & Humbert JF (2004) Cyanophage diversity, inferred from g20 gene analyses, in the largest natural lake in France, Lake Bourget. Appl Environ Microbiol 70: 1017–1022. Dreher T, Brown N, Bozarth SC et al. (2011) A freshwater cyanophage whose genome indicates close relationships to photosynthetic marine cyanomyophages. Environ Microbiol 13: 1858–1874. Duhamel S, Domaizon-Pialat I, Personnic S & Jacquet S (2006) Assessing the microbial community dynamics and the role of bacteriophages in bacterial mortality in Lake Geneva. J Water Sci 19: 115–126. 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. Frederickson CM, Short SM & Suttle CA (2003) The physical environment affects cyanophage communities in British Columbia inlets. Microb Ecol 46: 348–357. Fuhrman JA (1999) Marine viruses and their biogeochemical and ecological effects. Nature 399: 541–548. Fuller NJ, Wilson WH, Joint IR & Mann NH (1998) Occurrence of a sequence in marine cyanophages similar to that of T4 g20 and its application to PCR-based detection and quantification techniques. Appl Environ Microbiol 64: 2051–2060. Garza DR & Suttle CA (1998) The effect of cyanophages on the mortality of Synechococcus spp. and selection for UV resistant viral communities. Microb Ecol 36: 281–292. Gilbert JA, Steele J, Caporaso J et al. (2012) Defining seasonal marine microbial community dynamics. ISME J 6: 298–308. Hewson I, Barbosa JG, Brown JM, Donelan RP, Eaglesham JB, Eggleston EM & LaBarre BA (2012) Temporal dynamics and decay of putatively allochtonous and autochtonous viral genotypes in contrasting freshwater lakes. Appl Environ Microbiol 78: 6583–6591.

FEMS Microbiol Ecol 86 (2013) 312–326

Freshwater cyanomyophages and picocyanobacteria

Huang S, Wang K, Jiao N & Chen F (2012) Genome sequences of siphoviruses infecting marine Synechococcus unveil a diverse cyanophage group and extensive phage-host genetic exchanges. Environ Microbiol 14: 540–558. Izaguirrel I, Allende L & Marinone MC (2003) Comparative study of the planktonic communities of three lakes of contrasting trophic status at Hope Bay (Antarctic Peninsula). J Plankton Res 25: 1079–1097. Jacquet S, Heldal M, Iglesias-Rodriguez D, Larsen A, Wilson W & Bratbak G (2002) Flow cytometric analysis of an Emiliana huxleyi bloom terminated by viral infection. Aquat Microb Ecol 27: 111–124. Jacquet S, Rimet F & Girel C et al. (2011) Suivi scientifique de la qualite des eaux du lac du Bourget pour l’annee 2010. Report INRA-CISALB. Jacquet S, Domaizon I & Anneville O (2012) Evolution de parametres cles indicateurs de la qualite des eaux et du fonctionnement ecologique des grands lacs peri-alpins (Leman, Annecy, Bourget): Etude comparative de trajectoires de restauration post-eutrophisation. Arch Sci 65: 191–208. Lennon JT & Martiny JBH (2008) Rapid evolution buffers ecosystem impacts of viruses in a microbial food web. Ecol Lett 11: 1178–1188. Lennon JT, Khatana SAM, Marston MF & Martiny JBH (2007) Is there a cost of virus resistance in marine cyanobacteria? ISME J 1: 300–312. Li WKW (1994) Primary productivity of prochlorophytes, cyanobacteria and eukaryotic ultraphytoplankton: measurements from flow cytometric sorting. Limnol Oceanogr 39: 169–175. Liu X, Kong S, Shi M, Fu L, Gao Y & An C (2008) Genomic analysis of freshwater cyanophage Pf-WMP3 Infecting cyanobacterium Phormidium foveolarum: the conserved elements for a phage. Microb Ecol 56: 671–680. Liu W, Cheng K, Zhao Y, Wu Y, Cai C, Liu SY & Jin XC (2011) Cyanophage decay and its causes in eutrophic freshwater. J Food Agric Environ 9: 963–966. Lu J, Chen F & Hodson RE (2001) Distribution, isolation, host specificity, and diversity of cyanophages infecting marine Synechococcus spp. in river estuaries. Appl Environ Microbiol 67: 3285–3290. Mann NH (2003) Phages of the marine cyanobacterial picophytoplankton. FEMS Microbiol Rev 27: 17–34. Marie D, Brussaard CPD, Thyrhaug R, Bratbak G & Vaulot D (1999) Enumeration of marine viruses in culture and natural samples by flow cytometry. Appl Environ Microbiol 65: 45–52. Marie D, Partensky F, Simon N, Guillou L & Vaulot D (2000) Flow cytometry analysis of marine picoplankton. Living Colors: Protocols in Flow Cytometry and Cell Sorting (Diamond RA & DeMaggio S, eds), pp. 421–454. Springer Verlag, New York. Marston MF & Sallee JL (2003) Genetic diversity and temporal variation in the cyanophage community infecting marine Synechococcus species in Rhode Island’s coastal waters. Appl Environ Microbiol 69: 4639–4647.

FEMS Microbiol Ecol 86 (2013) 312–326

325

Matteson AR, Loar SN, Bourbonniere RA & Wilhelm SW (2011) Molecular enumeration of an ecologically important cyanophage in a Laurentian Great Lake. Appl Environ Microbiol 77: 6772–6779. Matteson AR, Rowe JM, Ponsero AJ, Pimentel TM, Boyd PW & Wilhelm SW (2013) High abundances of cyanomyoviruses in marine ecosystems demonstrate ecological relevance. FEMS Microbiol Ecol 84: 223–234. M€ uhling M, Fuller NJ, Millard A et al. (2005) Genetic diversity of marine Synechococcus and co-occurring cyanophage communities: evidence for viral control of phytoplankton. Environ Microbiol 7: 499–508. Palenik B (1994) Cyanobacterial community structure as seen from RNA polymerase gene sequence analysis. Appl Environ Microbiol 60: 3212–3219. Parada V, Herndl GJ & Weinbauer MG (2006) Viral burst size of heterotrophic prokaryotes in aquatic systems. J Mar Biol Assoc UK 86: 613–621. Partensky F, Blanchot J & Vaulot D (1999) Differential distribution and ecology of Prochlorococcus and Synechococcus: a review. Marine Cyanobacteria (Charpy L & Larkum AWD, eds). Bull Inst Oceanogr, Monaco, no Special 19: 431–449. Pernthaler J, Simek K, Sattler B, Schwarzenbacher A, Bobkova J & Psenner R (1996) Short-term changes of protozoan control on autotrophic picoplankton in an oligomesotrophic lake. J Plankton Res 18: 443–462. Personnic S, Domaizon I, Dorigo U, Berdjeb L & Jacquet S (2009a) Seasonal and spatial variability of virio, bacterioand picophytoplanktonic abundances in three peri-alpine lakes. Hydrobiologia 627: 99–111. Personnic S, Domaizon I, Sime-Ngando T & Jacquet S (2009b) Seasonal variations of microbial abundances and virus- vs flagellate-induced mortality of picoplankton in three perialpine lakes. J Plankton Res 31: 1161–1177. Pradeep RAS & Sime-Ngando T (2010) Resources drive tradeoff between viral lifestyles in the plankton: evidence from freshwater microbial microcosms. Environ Microbiol 12: 467–479. Sandaa RA & Larsen A (2006) Seasonal variations in virus-host populations in Norwegian coastal waters: focusing on the cyanophage community infecting marine Synechococcus spp. Appl Environ Microbiol 72: 4610–4618. Shelford EJ, Middelboe M, Møller EF & Suttle CA (2012) Virus-driven nitrogen cycling enhances phytoplankton growth. Aquat Microb Ecol 66: 41–46. Short SM & Suttle CA (2002) Sequence analysis of marine virus communities reveals that groups of related algal viruses are widely distributed in nature. Appl Environ Microbiol 68: 1290–1296. Short CM & Suttle CA (2005) Nearly identical bacteriophage structural gene sequences are widely distributed in both marine and freshwater environments. Appl Environ Microbiol 71: 480–486. Simek K, Bobkova J, Macek M, Nedoma J & Psenner R (1995) Ciliate grazing on picoplankton in a eutrophic reservoir

ª 2013 Federation of European Microbiological Societies Published by John Wiley & Sons Ltd. All rights reserved

X. Zhong et al.

326

during the summer phytoplankton maximum: a study at the species and community level. Limnol Oceanogr 40: 1077– 1090. Stockner J, Callieri C & Cronberg G (2000) Picoplankton and other non-bloom-forming cyanobacteria in lakes. The Ecology of Cyanobacteria: Their Diversity in Time and Space (Whitton BA, & Potts M, eds), pp. 195–231. Kluwer Academic Publishers, Dordrecht. Sullivan MB, Coleman ML, Quinlivan V et al. (2008) Portal protein diversity and phage ecology. Environ Microbiol 10: 2810–2823. Sullivan MB, Huang KH, Ignacio-Espinoza JC et al. (2010) Genomic analysis of oceanic cyanobacterial myoviruses compared with T4-like myoviruses from diverse hosts and environments. Environ Microbiol 12: 3035–3056. Suttle CA (2000) Ecological, evolutionary, and geochemical consequences of viral infection of cyanobacteria and eukaryotic algae. Viral Ecology (Hurst C, ed.), pp. 247–296. Academic Press, San Diego, CA. Suttle CA (2005) Viruses in the sea. Nature 437: 356–361. Suttle CA (2007) Marine viruses-major players in the global ecosystem. Nat Rev Microbiol 5: 801–812. Tai V & Palenik B (2009) Temporal variation of Synechococcus clades at a coastal Pacific Ocean monitoring site. ISME J 3: 903–915. Thingstad TF & Lignell R (1997) Theoretical models for the control of bacterial growth rate, abundance, diversity and carbon demand. Aquat Microb Ecol 13: 19–27. Thomas R, Berdjeb L, Sime-Ngando T & Jacquet S (2011) Viral abundance, production, decay rates and life strategies (lysogeny versus lysis) in Lake Bourget (France). Environ Microbiol 13: 616–630. Tsai AY, Gong GC, Robert WS, Chiang KP, Huang JK & Chan YF (2012) Viral lysis and nanoflagellate grazing as factors controlling diel variations of Synechococcus spp. summer abundance in coastal waters of Taiwan. Aquat Microb Ecol 66: 159–167. Vaulot D (1989) CYTOPC: processing software for flow cytometric data. Signal Noise 2: 8. Wang K, Wommack KE & Chen F (2011) Abundance and distribution of Synechococcus spp. and cyanophages in the Chesapeake Bay. Appl Environ Microbiol 77: 7459–7468.

ª 2013 Federation of European Microbiological Societies Published by John Wiley & Sons Ltd. All rights reserved

Waterbury JB & Valois FW (1993) Resistance to co-occurring phages enables marine Synechococcus communities to coexist with cyanophages abundant in seawater. Appl Environ Microbiol 59: 3393–3399. Weinbauer MG (2004) Ecology of prokaryotic viruses. FEMS Microbiol Rev 28: 127–181. Weinbauer M, Bonilla-Fidji O, Chan AM et al. (2011) Synechococcus growth in the ocean may depend on the lysis of heterotrophic bacteria. J Plankton Res 33: 1465–1476. Weisse T (1993) Dynamics of autotrophic picoplankton in marine and freshwater ecosystems. Advances in Microbial Ecology, Vol. 13 (Jones JG, ed.), pp. 327–370. Plenum Press, New York. Wilhelm SW & Suttle CA (1999) Viruses and nutrient cycles in the sea. Bioscience 49: 781–788. Wilson WH & Mann NH (1997) Lysogenic and lytic viral production in marine microbial communities. Aquat Microb Ecol 13: 95–100. Wilson WH, Fuller NJ, Joint IR & Mann NH (1999) Analysis of cyanophage diversity and population structure in a south-north transect of the Atlantic Ocean. Marine Cyanobacteria (Sharpy L & Larkum AWD, eds), pp. 209–216. Monaco Bul Inst Oceanogr, Paris. Zhong Y, Chen F, Wilhelm SW, Poorvin L & Hodson RE (2002) Phylogenetic diversity of marine cyanophage isolates and natural virus communities as revealed by sequences of viral capsid assembly protein gene g20. Appl Environ Microbiol 68: 1576–1584.

Supporting Information Additional Supporting Information may be found in the online version of this article: Fig. S1. Changes over time in the number of bands corresponding to the cyanomyovirus and picocyanobacterial communities in 2011 in Lakes Annecy and Bourget according to PCR-DGGE analyses. Fig. S2. DGGE fingerprinting for amplicons of picocyanobacteria-specific 16S-rRNA-gene from January to November in Lake Bourget.

FEMS Microbiol Ecol 86 (2013) 312–326