Links between Geographic Location, Environmental Factors, and. Microbial Community Composition in Sediments of the Eastern. Mediterranean Sea.
Links between Geographic Location, Environmental Factors, and Microbial Community Composition in Sediments of the Eastern Mediterranean Sea P.N. Polymenakou1,3, S. Bertilsson2, A. Tselepides1 and E.G. Stephanou3 (1) Hellenic Center for Marine Research, Gournes Pediados, 71003, Heraklion, Crete, Greece (2) Department of Limnology, Evolutionary Biology Center, Norbyv. 20, SE-752 36 Uppsala University, Uppsala, Sweden (3) Environmental Chemical Processes Laboratory, Department of Chemistry, University of Crete, 71409, Heraklion, Greece Received: 21 December 2003 / Accepted: 7 April 2004 / Online publication: 7 July 2005
content and sediment chlorophyll a were important in this regard.
Abstract
The bacterial community composition of marine surface sediments originating from various regions of the Eastern Mediterranean Sea (12 sampling sites) was compared by parallel use of three fingerprinting methods: analysis of 16S rRNA gene fragment heterogeneity by denaturing gradient electrophoresis (DGGE), terminal restriction fragment length polymorphism (T-RFLP), and analysis of phospholipid-linked fatty acid composition (PLFA). Sampling sites were located at variable depths (30–2860 m; water column depth above the sediments) and the sediments differed greatly also in their degree of petroleum contamination (0.4–18 lg g)1), organic carbon (0.38–1.5%), and chlorophyll a content (0.01–7.7 lg g)1). Despite a high degree of correlation between the three different community fingerprint methods, some major differences were observed. DGGE banding patterns showed a significant separation of sediment communities from the northern, more productive waters of the Thermaikos Gulf and the oligotrophic waters of the Cretan, S. Ionian, and Levantine Sea. T-RFLP analysis clearly separated the communities of deep sediments (>1494 m depth) from their shallow (108 cells mL)1 [18, 19, 58]) relative to the overlying water column and their key function in mediating and regulating the transformation and speciation of major bioactive elements (e.g., carbon, nitrogen, phosphorus, oxygen, and sulfur) in these environments. Sediment bacteria are also key players in the degradation of organic pollutants [1, 9, 28, 37, 60, 61] and represent a major reservoir of genetic variability with a local diversity equal to soil systems [56]. Despite their importance, our knowledge of the bacteria that inhabit sediments is very limited and based primarily on highly selective cultivation studies [10, 54]. In order to study the regional variability and to some extent also the influence of environmental parameters on bacterial community composition in marine sediments, we sampled multiple stations from a variety of regions and depths in the eastern basin of the Mediterranean Sea (Thermaikos Gulf, Cretan Sea, South Ionian Sea, Levantine Sea). The microbial community composition in the sediments was analyzed by three different cultureindependent techniques: analysis of 16S rRNA gene fragment heterogeneity by denaturing gradient gel electrophoresis (DGGE) [42–44] and terminal restriction fragment length polymorphism (T-RFLP) [20, 39, 40], as well as analysis of phospholipid-linked fatty acid composition (PLFA) [63]. Cultivation-based techniques are less suited for assessing the composition of bacterial
Springer Science+Business Media, Inc. 2005
367
368
P.N. POLYMENAKOU
communities in natural environments since the dominant fraction of extant microbiota (>99%) escape cultivation [46, 53]. The eastern basin of the Mediterranean Sea is considered to be one of the most oligotrophic regions in the world and is characterized by an overall nutrient deficit [26, 58]. This area is also characterized by a complex bottom geomorphology [27, 45] and dynamic hydrographic and meteorological conditions [6, 32, 55, 58]. The Thermaikos Gulf is mesotrophic and the transport and deposition of particulate matter of riverine origin mainly takes place along the western part of the continental shelf [27]. In contrast, the Cretan Sea is an oligotrophic and very dynamic region from a hydrological and physiographic point of view. The Cretan Basin consists of a narrow continental shelf (1.5 slope) followed by a steep slope (2–4) and a rather flat deeper area with depths >1500 m [14]. The Cretan basin is influenced by the cold and oxygen-depleted Trans-Mediterranean water. This water also has a low salinity but is on the other hand quite rich in nutrients flowing in from the Levantine Sea [6, 55]. The South Ionian Sea exchange water with the Levantine basin through the Cretan passage and is also linked to the South Adriatic Sea. Hence the Ionian Sea is the transition basin for the spreading of the deep thermohaline water mass from its source in the southern Adriatic Sea to the Levantine Sea [38]. These features affect the distribution of nutrients and consequently the organic matter production in the euphotic zone which eventually reaches the ocean sea floor to fuel benthic communities [57]. Finally, the sampled regions are exposed to varying degree of anthropogenic influences (e.g., petroleum contamination [23]), and these features combined are likely to affect not only the chemical composition of the surface sediments, but also the associated microbial communities. There is a general lack of information regarding microbial community composition in sediments of the Eastern Mediterranean Sea, with the only available data retrieved from the extreme hydrothermal vent area near Milos Island (Greece) [12, 49, 50] and the late Pleistocene organic-rich sediments (sapropels) southeast of Crete (Greece) [17]. The available information on geographical variation in sediment microbiota globally is also limited and mostly restricted to 16S rDNA clone library analysis from different deep-sea environments such as the Japan Trench [33] or from a comparative study between a sediment lake and a coastal marine basin [11]. The present study represents the first geographical survey of microbial community structure in sediments of the poorly characterized Eastern Mediterranean Sea as well as one of the first studies of geographical variation in sediment bacterial communities in general.
ET AL.:
MICROBIAL COMMUNITIES OF THE EASTERN MEDITERRANEAN SEA
Figure 1. Geographic locations of sediment sampling stations in the Eastern Mediterranean Sea. Therm: Thermaikos Gulf; Creta: Cretan Sea; S. Ionian: Ionian Sea; Levan: Levantine Sea.
Materials and Methods Sediment Collection and Biogeochemical CharacterizaSediment samples were collected from four diftion.
ferent regions of the Eastern Mediterranean Sea (Fig. 1). A Bowers and Connelly Multiple-corer (8 cores, i.d. 9.0 cm) [7] was used to collect undisturbed sediment samples from the Thermaikos Gulf in February 2002 and from the Levantine and Cretan seas in May and July 2002, respectively. Cores from the South Ionian Sea were collected in November 2001, using an USNEL Boxcorer. All sampling was carried out onboard the R/Vs Aegaeo and Philia. Mixed surface sediment (0–1 cm) of each sample was used for all analyses. Subsamples for analysis of phospholipids and petroleum hydrocarbons were sealed in aluminum foil and stored at )20C until further analysis took place, whereas samples for DNA analysis were collected and stored frozen in sterile plastic cups. Sediment redox potential was measured using calibrated combined redox electrodes (Mettler Toledo, Switzerland). Sediment chlorophyll a concentration was determined fluorometrically [36, 64] using a Turner TD)700 fluorometer. Total organic carbon in sediments was analyzed according to Hedges and Stern [25], using a PerkinElmer CHN 2400 analyzer. All measurements were earned out in triplicate. n-Alkanes and petroleum hydrocarbons (petroleum H/C; assessed as the sum of nalkanes and unresolved complex mixture—UCM) were Soxhlet extracted from the sediments and fractionated as described by Gogou et al. [22]. Compound identification was performed by GC-MS analysis with authentic standard compounds. 1-Chlorohexadecane was used as internal standard for quantification. The GC-MS analysis was carried out on a Hewlett-Packard 6890 GC equipped with a mass-selective detector [22]. The sediments were
P.N. POLYMENAKOU
ET AL.:
MICROBIAL COMMUNITIES OF THE EASTERN MEDITERRANEAN SEA
compared by principal component analysis (PCA) using analyzed environmental parameters (chlorophyll a, organic carbon content, total phospholipids, petroleum hydrocarbons, and n-alkanes). PCA was carried out using the PRIMER 5.2.2 software package (Plymouth Routines In Multivariate Ecological Research). In addition, the analyzed environmental parameters were transformed into a sediment similarity matrix and a dendrogram was produced using neighbor-joining cluster analysis included in the same software. Extraction of Nucleic Acids. Total DNA was extracted from sediments using the FastDNA-Spin Kit for Soil (Q-BIOgene, Carlsbad, CA). Aliquots of 0.5 g of fresh sediment were distributed to individual Lysing Matrix tubes included in the extraction kit. DNA extraction was performed according to the protocol provided by the manufacturer with the exception that cell lysis was achieved using a Mini-Beadbeater (Biospec products, Bartlesville OK) set to 5000 rpm over 30 s. DNA extracts were stored at )80C until analysis. Analysis of recovered DNA by 1% agarose gel electrophoresis, ethidium bromide staining, and UV transillumination showed that fragment size typically exceeded 20 kb in length (data not shown). Nucleic acid extracts from each sample were also analyzed spectrophotometrically at 260 and 280 nm using a Lambda-40 spectrophotometer (PerkinElmer) and Uvette microcuvettes (Eppendorff). The DNA concentration in the extracts was estimated from UV-absorbance according to Sambrook and Russel [47] and varied from 15 to 107.5 ng lL)1 in the different extracts. T-RFLP Analysis. T-RFLP was used to assess the genetic heterogeneity of PCR products generated in a mixed-template PCR amplification of 16S rRNA using bacteria-specific primers. In this form, T-RFLP is a culture independent method that provides a fingerprint of the bacterial community composition and has proven to be a reproducible and accurate tool for community analysis [16, 35, 42]. The bacteria-specific primer 27f labeled with hexachlorofluorescein [59] and the universal primer 519r [31] were used in the PCR reactions. For each sample, triplicate 30-lL reactions were set up and PCR amplified using a Stratagene Robocycler with an initial 3-min denaturation at 94C, followed by 30 cycles of 1 min at 94C, 1 min at 50C, and 2 min at 72C, followed by a single 10-min extension at 72C. Each reaction tube contained 1 ng of genomic DNA template, PCR buffer (10 mM Tris-HCl, pH 9, 50 mM KCl, 0.1% Triton X-100, and 2 mM MgCl2), 200 lM of each deoxynucleoside triphosphate, 100 nM of each primer, and 2.5 U of Taq DNA polymerase (Invitrogen, Carlsbad, CA). Replicate PCR products were pooled and purified using a Qiaquick PCR purification kit (Qiagen, Valencia,
369
CA). The concentration of PCR products from the different samples was determined by comparison to a Low DNA Mass ladder (Invitrogen) using 2% agarose gel electrophoresis, ethidium bromide staining, and UVtransillumination. Approximately 60 ng aliquots of each PCR product were separately digested with the restriction enzymes HaeIII, HhaI, and RsaI at 37C overnight as specified by the manufacturer (Invitrogen). Fragment size was determined on an ABI 3700, 96-capillary sequencer after addition of a
