Introduction

Photosynthetic organisms have a crucial role in the marine environment and in vast areas of the oceans the majority of the primary production is performed by cells smaller than 2–3 μm (picophytoplankton) (Marañón et al., 2001). Marine photosynthetic picoeukaryotes (PPEs) are found in the ocean photic zone (Sieburth et al., 1978) at concentrations between 102 and 105 cells ml−1 (Li, 2009). PPEs contribute significantly to global biomass and primary productivity, in spite of their relatively low abundance as compared with prokaryotic picophytoplankton (Li, 1994; Worden et al., 2004). Their distribution has been mainly estimated by flow cytometry, based on their size and natural pigment fluorescence (Li et al., 1993), and by high-performance liquid chromatography analysis (Andersen et al., 1996). In the past 10 years, molecular studies (Díez et al., 2001; López-García et al., 2001; Moon-van der Staay et al., 2001) showed wide genetic diversity of picoplanktonic eukaryotes by sequencing 18S ribosomal RNA (rRNA) genes directly from natural samples. However, it was soon realized that this approach favors heterotrophic organisms such as those belonging to stramenopiles or alveolates (Vaulot et al., 2002). Recently, approaches such as plastid 16S rRNA gene phylogenies (Fuller et al., 2006) or flow cytometric sorting (Shi et al., 2009) have allowed more direct targeting of PPE diversity. However, most studies have focused on the gene (DNA) rather than transcript (RNA) level, thus estimating abundance rather than biomass or activity of the cells. In the case of eukaryotes this issue might get very complicated, as in some cases they contain between one and a few thousand copies of the rRNA gene (Prokopowich et al., 2003; Zhu et al., 2005; Massana et al., 2008). The few studies that have looked at rRNA instead of the rRNA genes (Stoeck et al., 2007; Not et al., 2009) have shown that the most abundant groups are in general not the most active ones. For example, in the North West Mediterranean Sea (Not et al., 2009), though picoplanktonic rRNA gene sequences are dominated by alveolates and radiolarians, rRNA transcripts are dominated by marine heterotrophic stramenopiles, which are known to be very active predators (Massana et al., 2009).

Compared with photosynthetic cyanobacteria that are dominated by just two genera (Prochlorococcus and Synechococcus), PPEs are more diverse with every algal class containing at least some organism of picoplanktonic size, although the number of described species remains very low (Thomsen, 1986; Vaulot et al., 2008). The most abundant PPE groups observed in the marine environment (Díez et al., 2001; Moon-van der Staay et al., 2001; Marie et al., 2006; Le Gall et al., 2008) are the Chlorophyta (Prasinophyceae), Stramenopiles (or Heterokontophyta, containing in particular diatoms but also numerous other classes such as the Pelagophyceae or the Chrysophyceae), Alveolates (Dinophyceae), Haptophyta (Prymnesiophyceae), and Cryptophyta (Cryptophyceae).

Although previous studies have focused on the survey of rRNA operons, we decided to use a functional, photosynthetic protein-coding gene. Not using rRNA genes enables focusing on photosynthetic microorganisms without the masking of high background data coming from heterotrophic microorganisms found in the same water column. We decided to focus on psbA genes and transcripts (coding for the protein D1 of photosystem-II reaction center), which proved previously to be good indicators for a variety of photosynthetic marine microbial groups: dinoflagellates on the DNA level (Zhang et al., 2000), marine picoeukarya on the DNA level (Zeidner et al., 2003; Zeidner and Béjà, 2004), and cyanobacteria on the DNA and RNA levels (Zeidner et al., 2003; Zeidner and Béjà, 2004; Sharon et al., 2007).

In this paper, we report a detailed analysis of the diversity of photosynthetically active PPEs in oligotrophic East Mediterranean waters at different seasons based on psbA transcripts.

Materials and methods

Cultures

All cultures were obtained from the Roscoff Culture Collection (http://www.sb-roscoff.fr/Phyto/RCC/; Vaulot et al., 2004) and are listed in Table 1.

Table 1 List of cultured strains for which the psbA gene was sequenced
Full size table

Sample collection

Seawater samples were collected during four cruises (March, May, and October 2006 and January 2007) on board the R/V Mediterranean Explorer. Two locations were sampled: near-shore station (Tb200) and open sea station (Tb1000). Station Tb200 is located 20 km from the coast (32°09′N, 34°34′E) at ca. 200 m bottom depth and station Tb1000 is located 51 km offshore (32°10′N, 34°14′E) at ca. 1000 m bottom depth (Figure 1). The highest concentration of photosynthetic biomass (based on Chl a concentrations and variable fluorescence) was encountered at both stations during the winter mixing period. At that time of the year, the deep chlorophyll maximum (DCM) layer (∼110–150 m) was not developed at Tb200, showing uniform Chl a distribution (∼0.2–0.3 μg Chl a l−1), whereas at Tb1000 it was widely spread between 50 and 130 m (∼0.4 μg Chl a l−1). During the stratified period (June–September) Chl a concentrations peaked to 0.45 μg l−1 at the DCM. Additional description of these stations can be found in Bar-Zeev et al. (2008). Surface water samples were collected at both stations, whereas samples from the DCM were collected only at Tb1000. Twenty liters were pre-filtered through a 3 μm polycarbonate (GE Water & Process Technologies filter, Trevose, PA, USA) (samples from January 07) or a Whatman GF/A glass-fiber filter (1.6 μm nominal pore size) in all other samples, and the filtrate was collected on a 0.2 μm Sterivex filter (Millipore, Bedford, MA, USA) using a peristaltic pump (Cole Parmer Masterflex 5, channel hardware, Vernon Hills, IL, USA). Although the January samples were collected using a bigger-pore-size pre-filter, we see no significant bias compared with the other samples with regard to photosynthetic picoeukaryal proportions on the RNA level. After collection, the Sterivex filters were filled with 1 ml of lysis buffer (Massana et al., 1997) and stored at −80 °C.

Figure 1
figure 1

Map of station locations. Samples were taken at a near-shore station (Tb200), located 20 km from the coast (32°09′N, 34°34′E) at ca. 200 m bottom depth and at an open sea station (Tb1000), located 51 km offshore (32°10′N, 34°14′E) at ca. 1000 m bottom depth.

Full size image

Flow cytometry

Samples of 1.8 ml were taken directly from the Niskin bottles and were fixed immediately at room temperature with 23 μl of 25% glutaraldehyde (Sigma G-5882, St Louis, MO, USA) for 20 min after which they were frozen in liquid nitrogen. In the laboratory, the samples were kept at −80 °C until analysis. Samples were thawed at 37 °C and analyzed on a FACScan (Becton Dickinson, Franklin Lakes, NJ, USA) using excitation with an argon laser (488 nm), for either 10–15 min or until 10 000 cells were counted. Discrimination was based on the following parameters: forward and side scatter both related to cell size and pigment orange and red fluorescences for phycoerythrin and chlorophyll (585 and 630 nm, respectively). Beads (0.93 μm, Polysciences, Warrington, PA, USA) served as a standard.

Carbon biomass estimations

Carbon biomass was calculated according to Verity et al. (1992). Cell carbon content was calculated as carbon (pg)=0.433 × (biovolume)0.863 using the following biovolume parameters for Prochlorococcus (0.6)3 μm3, Synechococcus (1)3 μm3, and picoeukarya (2.5)3 μm3. Carbon biomass=carbon(pg) × cell concentration (cell ml−1).

Nucleic acid extraction and reverse transcription

RNA and DNA were extracted from the samples according to Man-Aharonovich et al. (2007). Extracted DNA was stored frozen at −20 °C. The RNA was treated with RNase-Free DNase I (Ambion, Austin, TX, USA) for 30 min at 37 °C to remove DNA. DNase was inactivated by heat denaturation at 75 °C for 10 min and samples were stored at −80 °C until further use. For RT–PCR, total RNA (100–300 ng) was reverse transcribed with psbA degenerate reverse primer psbA-2R (Wang and Chen, 2008) using Bio-RT (Bio-Lab, Ipswich, MA, USA) according to the manufacturer's instructions. Reaction mixtures were incubated at 37 °C for 1 h.

Amplification of psbA genes

PsbA gene fragments (∼750 bp) were amplified by PCR from cDNA and genomic DNA using the degenerated PCR primers designed by Wang and Chen (2008) that target the conserved YPIWEA and HNFPLD regions. PCR was performed in a total volume of 25 μl containing 10 ng of template DNA/cDNA, 2.5 μl of 10 X OptiBuffer, 2 μl of dNTP, 1.3 μl of 50 mM MgCl2, 1 μl of 25 μM psbA-1F (TAYCCNATYTGGGAAGC), 1 μl of 25 μM psbA-2R (TCRAGDGGGAARTTRTG), and 1.2 U of BIO-X-ACT (Bioline, London, UK). Amplification conditions comprised steps at 95 °C for 2 min, and 30 cycles at 94 °C for 1 min, 55 °C for 1 min, and 68 °C for 1 min followed by one step of 7 min at 68 °C. We performed two tests for presence of contaminating DNA in the RNA sample: (1) PCR of the RNA samples without the reverse transcription step and (2) treating the RNA samples with RNase and subjecting them to RT–PCR. To test the reagents for DNA contamination, PCR reactions without template were performed. All tests gave the expected negative results.

Cloning of psbA genes, library construction, and sequencing

PCR products were cloned using the Qiagen PCR cloning kit according to the manufacturer's instruction. Clones were randomly picked up into 20 96-well plates. Each 96-well plate represented different date (March, May, and October 2006 and January 2007), different station and depth (Tb200 surface, Tb1000 surface, and Tb1000 DCM), and different source (DNA and RNA). All plates were sequenced at the MPI for molecular genetics in Berlin.

Rarefaction analysis and community structure analysis

All 1205 psbA sequences, which had previously been predicted to be either of bacteriophage, bacterial, or eukaryotic origin (Tzahor et al., 2009), were used to infer phylogenies using PhyML (Guindon and Gascuel, 2003). The data were analyzed both as DNA and (translated) amino acid sequences. To estimate total sequence diversity and efficiency of sampling, maximum likelihood (ML) trees were broken down into a distance matrix (using MATLAB), which served as input for the rarefaction analysis and the ACE non-parametric diversity estimator all implemented in the DOTUR (Schloss and Handelsman, 2005) and SONS (Schloss and Handelsman, 2006) packages. These ML distances differ from the more commonly used Hamming distances or percent sequence identity used in such analyses; however, when operational taxonomic units are calculated as sequence clusters at predetermined similarity cutoffs, ML distances provide a higher probability that operational taxonomic unit boundaries coincide with true phylogenetic clusters. To ensure comparison with more traditional distance measures, the range of percent sequence similarity corresponding to a specific ML-based distance was computed from several randomly selected sequences.

GenBank deposition

psbA sequences from RCC cultures were deposited in GenBank under accession #s EU851954-EU851972; environmental psbA sequences were deposited in GenBank under accession #s EU940373-EU940692.

Results and discussion

psbA clone libraries

We amplified by PCR psbA genes and transcripts directly from DNA and RNA (cDNA) extracted simultaneously from the same samples (derived from a coastal station (Tb200) and a pelagic station (Tb1000) in the Eastern Mediterranean Sea) from mixed picoplankton assemblages using recently designed psbA primers (Wang and Chen, 2008). These primers amplify psbA from eukaryotes, cyanobacteria, and cyanophages. A total of 1205 randomly picked clones containing psbA inserts were sequenced; 618 were derived from DNA and 587 from RNA. We first differentiated psbA sequences of eukaryotic and bacterial origin. For rapid classification of the cyanobacterial and cyanophage psbA genes, we used the approach developed by Tzahor et al. (2009), which uses genomic signature and position-specific codon. This method successfully classified different psbA fragments into seven taxonomic groups (Synechococcus, HL-Prochlorococcus, LL-Prochlorococcus, Synechococcus-like Myovirus, Synechococcus-like Podovirus, Prochlorococcus-like Myovirus, and Prochlorococcus-like Podovirus), not including eukarya. Unexpectedly, though completely absent in the DNA extracts, eukaryotic psbA transcripts dominated surface water samples and could reach up to 88% of the total psbA RNA (Figure 2). This is not observed with the DCM samples, in which the eukaryotic psbA transcripts were found to be between 3% and 24% of the total psbA RNA.

Figure 2
figure 2

PPE abundance versus carbon biomass estimate. Relative psbA clone abundance (DNA and RNA), cell abundance (flow cytometry), and estimated carbon biomass, based on Verity et al. (1992), of PPE (black) compared with cyanobacteria and cyanophages (white) retrieved from station Tb1000 and Tb200 in the Eastern Mediterranean Sea.

Full size image

Comparison of the observed sequence diversity with that estimated by the ACE non-parametric richness estimator (Supplementary Figure S1) suggests that a large fraction of the taxonomic diversity was sampled. Using an operational taxonomic unit definition of ∼95–97% amino acid sequence similarity (or 0.02 ML-based distance), corresponding roughly to the order/class level (Supplementary Table S2), more than half of the diversity appears to be accounted for. Changing this cutoff to 80–85% amino acid identity (or 0.16 ML-based distance), corresponding to a level above the division, sampling was effectively saturated.

Photosynthetic picoplankton abundance and carbon biomass estimates

Flow cytometry analysis was used to provide information on abundance, cell size, and pigment content of the major photosynthetic picoplankton groups (cyanobacteria and PPEs) in our samples. Indeed, picophytoplankton cell counts (Supplementary Table S1) showed that Prochlorococcus and Synechococcus were numerically dominating at all stations and depths (Figure 2). In most cases, PPEs represented between 1 and 11% of the total cell counts. When converting abundance to carbon biomass (Verity et al., 1992), the relative part of PPEs increased, especially in surface waters (Figure 2). PPEs represented about 60% of the photosynthetic picoplankton carbon biomass in surface water at the pelagic station (Tb1000) in May and October 2006, and between 15 and 30% at the coastal station (Tb200). At the DCM depth, the PPEs carbon biomass fraction was lower (3–24% of the total). These observations agree with previous observations that although cyanobacteria are more abundant, PPEs sometimes dominate with respect to chlorophyll, carbon, and primary production, because of their higher carbon content (Li, 1994, 1995; Worden et al., 2004).

Diversity of active PPEs

Most of PPEs psbA sequences were clustered with the red algal lineage (94%), the rest being grouped with green algae (Figure 3). To better assign the different psbA transcript sequences retrieved, we included in the phylogenetic tree new psbA gene sequences corresponding to photosynthetic eukaryotes available in cultures (Table 1), trying to target phylogenetic groups containing picoeukaryotic species.

Figure 3
figure 3

Phylogenetic relationships of Eastern Mediterranean picoeukaryotic psbA transcripts. An ML tree of eastern Mediterranean PPE and cultures psbA sequences (308 and 38 sequences, respectively). Only sequences from cultures are labeled whereas sequences from this study were left unlabeled for clarity. Coloring of the innermost circle indicates phyla affiliation whereas month, depth, and station attributes of each environmental sequence are marked on the three outer circles. Bootstrap analysis of 100 replicates was conducted, and gray circles indicate values above 50%. Scale bar represents 0.1 substitutions/site for each unit of branch length. Colored stars and black arrows indicate branching of major phylogenetic clades.

Full size image

Haptophyta constitute the most numerous group in the psbA sequences (Figure 4), all of which seem to belong to the class Prymnesiophyceae and nearly all originating from surface waters. In addition, Haptophyta are also present in the sample with the lowest frequency of PPE psbA transcripts (3%, DCM of Station Tb1000 in May 2006). The high contribution of this group that we observe in surface waters is consistent with the dominance of the diagnostic pigment 19′hexanoylfucoxanthin in open oceanic waters where picoplankton is most important (Liu et al., 2009) as well as in the picoplanktonic fraction itself (Moon-van der Staay et al., 2001; Not et al., 2008). However, despite the dominance of this carotenoid in the small size fractions, very few truly picoplanktonic Prymnesiophyceae species have been described (Vaulot et al., 2008) and this class is in general quite under-represented in 18S clone libraries (Moon-van der Staay et al., 2001; Vaulot et al., 2008). However, several recent studies have shown that this under-representation is probably artefactual. First, the use of probes targeting 16S plastid rRNA has shown that Prymnesiophyceae can dominate the eukaryotic pico or ultra-phytoplankton fraction in environments as diverse as the Indian Ocean, the Mediterranean Sea, or the Pacific Ocean (Fuller et al., 2006; McDonald et al., 2007; Lepère et al., 2009). Second, Liu et al. (2009) using primers targeting the LSU rRNA gene showed that a very large number of haptophyte sequences could be recovered from the picoplankton size fraction in a variety of oceanic waters. Third, Shi et al. (2009) obtained a large number of Haptophyta 18S rRNA gene sequences from flow cytometry sorted PPE populations from the South East Pacific, including some from a potentially new class intermediate between Prymnesiophyceae and Pavlophyceae. In this study, a large group of sequences were closely related to Emiliania huxleyi (Figure 3), an ubiquitous species that not only makes large scale blooms in mid- and high-latitude nutrient-rich waters but is also often isolated from oligotrophic waters (Le Gall et al., 2008). The present data suggest that Haptophyta are also dominant members of the PPE community in the low-nutrient waters of the East Mediterranean Sea (Figure 4). In addition, two sequences were somewhat related to Haptophyta but fell outside its radiation and could belong a novel phylogenetic group (Figure 3).

Figure 4
figure 4

Taxonomic composition of each sample based on RNA-derived psbA sequences. Samples were taken on four different months, at two stations, TB200 (coastal) and TB1000 (Pelagic). At each station, samples were taken from the surface and, at TB1000, also from the DCM. Colors correspond to the coloring of phylogenetic affiliation in Figure 2.

Full size image

The second most abundant group of sequences was related to the Pelagophyceae. This is consistent with the carotenoid 19′butanoyloxyfucoxanthin, which is characteristics of this class, being important (after 19′hexanoylfucoxanthin) in oligotrophic waters (Not et al., 2008). Surprisingly, these sequences diverged quite significantly (only 95–98% identity) from Pelagomonas calceolata, a picoplankton species (Andersen et al., 1993) that is very often isolated from oceanic waters (Le Gall et al., 2008) and whose sequences found in 18S rRNA gene clone libraries are generally highly conserved (Moon-van der Staay et al., 2001; Shi et al., 2009). It is therefore possible that these psbA sequences belong to a new class, not yet brought in culture.

Some pbsA sequences also fell into the closely related class of Dictyochophyceae of which a picoplanktonic species, Florenciella parvula, has been described recently (Eikrem et al., 2004).

Another large group of sequences belonged to Chrysophyceae (Figure 3). This is quite interesting because though Chrysophyceae are abundant in fresh water, most marine Chrysophyceae described to date are heterotrophic such as those belonging to the genus Paraphysomonas and they make only small contributions to picoplankton 18S rRNA gene clone libraries (Vaulot et al., 2008). However, many plastid 16S rRNA gene sequences related to Chrysophyceae have been recovered from the open ocean (Fuller et al., 2006; McDonald et al., 2007) and 16S rRNA probes hybridized on PCR products suggest that Chrysophyceae are important contributors in oligotrophic waters (Fuller et al., 2006; Lepère et al., 2009). More recently, Chrysophyceae 18S sequences have also been recovered from flow cytometry sorted PPE populations in the SE Pacific Ocean (Shi et al., 2009). However, the nature of these cells still escapes us as no marine photosynthetic Chrysophyceae cultures of picoplanktonic size have been isolated and characterized until now.

Quite a few Cryptophyceae sequences were also recovered. Sequences of the 18S rRNA gene from this group are abundant in fresh waters (Lepère et al., 2006) but in marine systems usually only found in coastal waters and absent in open ocean samples (Vaulot et al., 2008). Here, they were found in all near shore samples and offshore only in surface and not at the DCM (Figure 4). One possibility is that coastal populations may have been transported offshore, the other being that the ecology of Cryptophyceae is different in Mediterranean Sea waters and that this group is present in blue waters.

Finally, a few sequences were affiliated to Mamiellales, an order from the Prasinophyceae that contains three genera Micromonas, Bathycoccus, and Ostreococcus that can dominate PPEs in coastal waters, for example in the English Channel or in the Chile upwelling (Not et al., 2004; Shi et al., 2009), but are also found sporadically in open ocean waters, especially for the genus Ostreococcus (Marie et al., 2006). In this study, Micromonas psbA sequences were only found near shore whereas those related to Ostreococcus and Bathycoccus were found offshore.

Conclusions

The analysis of different genes (nuclear 18S rRNA, plastid 16S rRNA, and now psbA) begins to offer a better view of the most abundant and most active groups within the PPE community. The present data based on psbA transcripts bring two major conclusions:

  1. 1)

    Eukaryotes account for a much higher fraction of psbA transcripts than of psbA genes, and therefore probably contribute significantly to primary production.

  2. 2)

    Prymnesiophyceae, Pelagophyceae, Chrysophyceae, and Cryptophyceae appear as the most active PPEs in Eastern Mediterranean waters. This confirms previous estimates in the other regions of the Mediterranean Sea based on pigment signatures and on environmental plastid 16S rRNA gene sequences (McDonald et al., 2007). Nonetheless, many of the corresponding species remain to be isolated and described.