Bacterial Diversity and Geochemical Profiles in Sediments ... for NO 3 −, Nessler method (SMEWW 4500-NH 3 C) for NH 4 +, and the Phenanthroline Method (SMEWW 3500-Fe) for...

This article was downloaded by: [Chongqing University]On: 30 May 2013, At: 18:43Publisher: Taylor & FrancisInforma Ltd Registered in England and Wales Registered Number: 1072954 Registered office: Mortimer House,37-41 Mortimer Street, London W1T 3JH, UK

Geomicrobiology JournalPublication details, including instructions for authors and subscription information:http://www.tandfonline.com/loi/ugmb20

Bacterial Diversity and Geochemical Profiles inSediments from Eutrophic Azorean LakesG. Martins a , I. Henriques b , D. C. Ribeiro a , A. Correia b , P. L. E. Bodelier c , J. V. Cruz d ,A. G. Brito a & R. Nogueira ea Institute for Biotechnology and Bioengineering, Centre of Biological Engineering, Universityof Minho, Braga, Portugalb CESAM and Department of Biology, University of Aveiro, Aveiro, Portugalc Netherlands Institute of Ecology (NIOO-KNAW), Department of Microbial Ecology, ACNieuwersluis, The Netherlandsd CVARG—Geosciences Department, University of Azores, Ponta Delgada, Portugale Institute of Sanitary Engineering and Waste Management, University of Hannover,Hannover, GermanyAccepted author version posted online: 27 Mar 2012.Published online: 12 Jul 2012.

To cite this article: G. Martins , I. Henriques , D. C. Ribeiro , A. Correia , P. L. E. Bodelier , J. V. Cruz , A. G. Brito & R.Nogueira (2012): Bacterial Diversity and Geochemical Profiles in Sediments from Eutrophic Azorean Lakes, GeomicrobiologyJournal, 29:8, 704-715

To link to this article: http://dx.doi.org/10.1080/01490451.2011.619633

PLEASE SCROLL DOWN FOR ARTICLE

Full terms and conditions of use: http://www.tandfonline.com/page/terms-and-conditions

This article may be used for research, teaching, and private study purposes. Any substantial or systematicreproduction, redistribution, reselling, loan, sub-licensing, systematic supply, or distribution in any form toanyone is expressly forbidden.

The publisher does not give any warranty express or implied or make any representation that the contentswill be complete or accurate or up to date. The accuracy of any instructions, formulae, and drug doses shouldbe independently verified with primary sources. The publisher shall not be liable for any loss, actions, claims,proceedings, demand, or costs or damages whatsoever or howsoever caused arising directly or indirectly inconnection with or arising out of the use of this material.

Geomicrobiology Journal, 29:704–715, 2012Copyright © Taylor & Francis Group, LLCISSN: 0149-0451 print / 1521-0529 onlineDOI: 10.1080/01490451.2011.619633

Bacterial Diversity and Geochemical Profiles in Sedimentsfrom Eutrophic Azorean Lakes

G. Martins,1 I. Henriques,2 D. C. Ribeiro,1 A. Correia,2 P. L. E. Bodelier,3

J. V. Cruz,4 A. G. Brito,1 and R. Nogueira5

1Institute for Biotechnology and Bioengineering, Centre of Biological Engineering, University of Minho,Braga, Portugal2CESAM and Department of Biology, University of Aveiro, Aveiro, Portugal3Netherlands Institute of Ecology (NIOO-KNAW), Department of Microbial Ecology, AC Nieuwersluis,The Netherlands4CVARG—Geosciences Department, University of Azores, Ponta Delgada, Portugal5Institute of Sanitary Engineering and Waste Management, University of Hannover, Hannover, Germany

In the Azores, the advanced trophic state of the lakes requiresa fast intervention to achieve the good ecological status prescribedby the Water Framework Directive. Despite the considerable ef-fort made to describe the phytoplankton growing on the watercolumn, the lack of information regarding the microbial processesin sediments is still high. Thus, for the successful implementationof internal management actions, the present work explored the re-lationships between geochemical profiles and dominant membersof the bacterial community in sediments from eutrophic Azoreanlakes. Lake Azul geochemical profiles were quite homogeneous forall parameters, while in lake Furnas the total iron profile presenteda peak below the aerobic layer. For lake Verde, the concentrationsof all studied parameters (20 ± 2% loss-on-ignition; 2.10 ± 0.08mg g−1 total phosphorus; 1.31 ± 0.50 mg g−1 total nitrogen; 8.06 ±0.13 mg g−1 total iron) in the uppermost sediment layer were ap-proximately two times higher than the ones in sediments from otherlakes, decreasing with sediment depth. The higher amounts of phos-phorus and organic matter in lake Verde suggested a higher inter-nal contribution of phosphorus to eutrophication. The dominantmembers of the sediment bacterial community, investigated by de-naturing gradient gel electrophoresis, were mostly affiliated to Pro-teobacteria phylum (Alpha-, Delta-, and Gamma-subclasses), groupBacteroidetes/Chlorobi and phylum Chloroflexi. The Cyanobacteria

Received 29 November 2010; accepted 17 August 2011.The authors are indebted to the Regional Department of Water

Resources and Land Planning for the grant (Contrato Excepcionadon◦ 4/2008/DROTRH), and its staff (Dina Pacheco), to University ofAzores (Paulo Antunes), and to INOVA (Manuela Cabral). NIOO-KNAW (Marion Meima) is acknowledged for the collaboration in es-tablishing the molecular methods. The authors also acknowledge theGrant SFRH/BD/25639/2005 from the Foundation for Science andTechnology/M.C.T., Portugal, awarded to Gilberto Martins.

Address correspondence to Gilberto Martins, IBB – Institute forBiotechnology and Bioengineering, Centre of Biological Engineering,University of Minho, Campus de Gualtar, 4710–057 Braga, Portugal.E-mail: [email protected]

phylum was solely detected in sediments from lake Verde and lakeFurnas that presented the highest amounts of nitrogen and phos-phorus both in the water column and sediments, while the otherphyla were detected in sediments from the three studied lakes. Inconclusion, management measurers to achieve the good ecologicalstatus until 2015 should be distinct for the different lakes takinginto account the relative magnitude of the nutrient sources and thebacterial diversity in sediments.

Keywords sediments, geochemical profiles, bacterial communitycomposition, eutrophication.

INTRODUCTIONEutrophication is the most common reason for lake manage-

ment and results mainly from anthropogenic activities occurringin the watershed (Martins et al. 2008; Conley et al. 2009). Reg-ulation of point sources and the attempt to reduce nonpointsources of nutrients are shifting the driver of eutrophicationfrom external to internal sources (Carpenter 2005). Phosphorus(P) release from sediments is one example of an internal driverof eutrophication (Martins et al. 2008).

Several methods have been developed to reduce P releasefrom sediments: artificial destratification, hypolimnetic aera-tion, dredging, flocculation, and sediment capping using passiveor active capping agents (Søndergaard et al. 2003; Hickey andGibbs 2009). Nevertheless, to increase the efficiency of thesemethods and decrease the potential to cause undesirable envi-ronmental effects, both to water quality and to aquatic biota, acharacterization of lake sediments is required (reference condi-tions) to the definition of the remediation process target (Hickeyand Gibbs 2009).

Sediments receive substantial inputs of both allochthonousmaterial (both organic and inorganic) and autochthonous mate-rial (mainly organic) from primary productivity (Huettel et al.

704

Dow

nloa

ded

by [

Cho

ngqi

ng U

nive

rsity

] at

18:

43 3

0 M

ay 2

013

BACTERIAL DIVERSITY IN EUTROPHIC AZOREAN LAKES 705

2003; Chen and White 2004). The composition of sediments isinfluenced by several factors, namely the chemistry of the over-lying water, the level of primary productivity, runoff or rivers, aswell as biogeochemical transformations occurring inside sedi-ments (Huettel et al. 2003; Chen and White 2004; Reitzel et al.2007). The most common geochemical profiles obtained in thesediments solid phase are total phosphorus (TP), total nitrogen(TN), organic matter (OM), and total iron (TFe). Examples ofhomogeneous (i.e., profiles that did not change significantly indepth) and non-homogeneous types of sediment profiles havebeen reported in literature (Søndergaard et al. 1996; Keldermanet al. 2005; Zeng et al. 2009).

Denaturing gradient gel electrophoresis (DGGE) is per-haps the most commonly technique used among the culture-independent fingerprinting techniques to characterize the bac-terial diversity in lake sediments (Kojima et al. 2006; Nelsonet al. 2007; Qu et al. 2008). Literature studies suggested that Pro-teobacteria, Bacteroidetes, Chloroflexi, and Actinobacteria arethe most common phyla in sediments from different eutrophiclakes and reservoirs (Tamaki et al. 2005; Kojima et al. 2006;Briee et al. 2007; Nelson et al. 2007; Schwarz et al. 2007; Quet al. 2008; Zeng et al. 2009).

Although considerable research efforts have been directed to-wards the characterization of microbial diversity in sediments,little is known about the effect of nutrients concentration on theoccurrence/predominance of certain microbial groups. To date,literature studies reported that the bacterial community diversityincreases with the amount of labile OM in sediments (Nelsonet al. 2007) and that high concentrations of nitrogen inducea shift in the dominant bacterial phylotypes (Haukka et al.2006). The lake trophic status affects the diversity of ammonia-oxidizing prokaryotes, with the increase of detected operationaltaxonomic units from oligotrophic lakes to mesotrophic lakes(Herrmann et al. 2009). The P content in sediments is corre-lated to both the number of bacteria associated to P accumu-lation/release and phosphate dissolving bacteria (phosphatasesexcretion) (Li et al. 2005; Zeng et al. 2009).

In Azorean lakes, despite the considerable effort made to de-scribe the phytoplankton growing on the water column (Santoset al. 2005; Martins et al. 2008), the lack of information re-garding the microbial processes both in the water column andsediments is still high. Thus, for the successful implementationof internal management actions to achieve the good ecologicalstatus prescribed by Water Framework Directive, the presentwork explored the relationships between geochemical profilesand dominant members of the bacterial community in sedimentsfrom three Azorean lakes with eutrophication problems (Verde,Azul, and Furnas).

MATERIALS AND METHODS

Study SiteThe three studied lakes, lake Verde, lake Azul, and lake Fur-

nas, are on the island of S. Miguel in the archipelago of Azores

(Portugal), located in the middle of the Atlantic ocean (37.75◦

latitude and −25.50◦ longitude). Lake Azul and lake Verde,jointly known as lake Sete Cidades, are connected by a narrowpassage. Morphometric and geochemical characteristics of thelakes can be found in Ribeiro et al. (2008) and in Martins etal. (2008). Lake Furnas and lake Verde are in an advanced stateof eutrophication and were classified as eutrophic, while LakeAzul is in a meso-eutrophic state. To plan future remediationmeasures to reduce the P input from sediments to the water col-umn it is also necessary to characterize the sediments referenceconditions corresponding to a situation without eutrophication.For that, a control site was chosen in another lake in the sameisland, lake Fogo, that was classified as oligotrophic.

Sampling and Geochemical ProfilesWater was sampled monthly from July to December 2007

with a 6 L Van Dorn bottle at the deepest point of the studiedlakes (24 m for lake Verde, 26 m for lake Azul, and 11 m lakeFurnas). These samples were analyzed for chemical parameters,such as TP, PO4

3−, NO3−, and NH4

+. Sediments cores with10 cm depth were taken in June 2007 with a gravitationalUwitec-corer (6 cm diameter, 40 cm length), sealed in situ insidethe core tubes, and transported (around 1 hour) to the laboratoryat 4◦C. Subsequently, each sediment core was divided into fivevertical layers (0–1 cm, 1–2 cm, 2.5–3.5 cm, 5–6 cm and, 9–10 cm) that were immediately frozen and stored until analysis.

To determine the loss-on-ignition (LOI) and water fractionof the sediments, approximately 2 g of each sediment layerwas weighed, dried at 105◦C, re-weighed, burned at 550◦C,and weighed a final time. Residues thus obtained were digestedwith 10 mL of concentrated HCl, boiled for 1 h at 150◦C, filteredthrough 0.45 µm membrane filter, and neutralized with NaOH.The final volume was adjusted to 50 mL. TP, TN, and total iron(TFe) were analyzed in digested sediments. Chemical analyseswere performed according to Standard Methods (APHA 1998).

The following spectrophotometric methods were used:molybdenum blue/stannous chloride method (SMEWW 4500-P) for PO4

3−, ultraviolet absorbance method (SMEWW 4500-NO3) for NO3

−, Nessler method (SMEWW 4500-NH3 C) forNH4

+, and the Phenanthroline Method (SMEWW 3500-Fe) forTFe. Temperature and dissolved O2 were determined in situ witha portable multiparameter meter (Multi 350i, WTW).

Microsensor ProfilingMicroprofiles of pH and oxygen concentration were obtained

ex-situ in the sediment-water interface and in the interstitialwater of the sediment from lake Furnas. Three perspex tubes(10.0 cm depth and 4.1 cm of diameter) open on top were filledwith the sediment and immersed in a parallelepiped containerfilled with 1.7 L lake water. The water column was open toair and the ratio of volume of water and area of sediment was43 m3 m−2.

The microprofiles were measured with a vertical resolution of250 µm (80 measured points in the sediment). The microsensors,Ox50 (oxygen microsensor) and pH50 (pH microsensor), were

Dow

nloa

ded

by [

Cho

ngqi

ng U

nive

rsity

] at

18:

43 3

0 M

ay 2

013

706 G. MARTINS ET AL.

operated with a level III microprofiling setup (Unisense), whichincludes a micromanipulator (MM33-2) with a motor controller(MC232). Data was acquired by computer using an analogi-cal/digital converter (Unisense, ADC216).

DNA Extraction and PCR AmplificationDNA was extracted and purified from 0.7 g sediment (wet

weight) using the UltraClean Soil DNA kit (MoBio, SolanaBeach, CA, USA), according to the manufacturer’s instructions.PCR amplification was performed in a 25 µL reaction mixturecontaining 2.5 µL of DNA template, 10 mM Tris/HCl (pH 8.3),50 mM KCl, 0.04% (w/v) bovine serum albumin, 200 µM ofeach deoxynucleotide, 1.5 mM MgCl2, 2.5 U/mL Taq DNApolymerase, and 0.5 µM of each primer (341F-GC: 5′-CGCCCG CCG CGC CCC GCG CCC GGC CCG CCG CCC CCGCC CC CCT ACG GGA GGC AGC AG-3′; 518R-5′-ATTACC GCG GCT GCT GG-3′).

The PCR amplification program started with an initial denat-uration at 94 ◦C for 10 min, followed by 30 cycles at 94◦C for 30s, 65◦C for 1 min, and 72◦C for 30 s. The annealing temperaturewas progressively decreased by 1◦C down to 55◦C. Reactionswere finished by a final extension at 72◦C for 10 min.

DGGE Profiling and AnalysisDGGE was performed as described by Muyzer et al. (1993).

Briefly, PCR products were separated on a 1.5 mm thick ver-tical gel containing 8% (w/v) polyacrylamide (37.5:1 acry-lamide:bisacrylamide) using a linear gradient of urea and for-mamide (denaturants), increasing from 30% at the top of the gelto 60% at the bottom. Here, 100% denaturant is defined as 7 Murea plus 40% (v/v) formamide.

The gel was loaded with 15 µL of PCR product previouslymixed with loading buffer (0.25 µL loading buffer per µLof PCR product). Electrophoresis was performed in 0.5 Tris-Acetate-EDTA buffer (40 mM Tris, 40 mM acetic acid, and1 mM EDTA pH 7.6) for 18 h at 75 V, using the DCode system(Bio-Rad, Hercules, CA). The gel was stained for 1 h in wa-ter containing 0.5 mg/mL ethidium bromide and the gel imagewas recorded with a Trans-illuminator Gel Doc 2000 (Bio-Rad,CA). The image was analysed and a binary (1/0) matrix wasconstructed taking into account the presence or absence of in-dividual bands in each lane.

Subsequent analysis was performed using the PRIMER v5software (Clarke and Gorley 2001). The binary matrix was trans-formed into a similarity matrix using the Bray-Curtis measure.A dendrogram was generated using the group average method.

The structural diversity of the microbial community was cal-culated using the Shannon index H (H = Pi log Pi), where Piis the importance probability of the bands in a lane (Shannonand Weaver 1963). H was calculated based on the intensity ofthe bands that were considered for the generation of the dendro-gram. The importance probability, Pi, was calculated as Pi =ni/N, where ni is the band intensity and N is the sum of all inten-sities in a lane. Band intensities were estimated using the Quan-

tityOne software package (BioRad, Richmond CA) followingthe default parameters selected in the software. The equitabilityindex, E, (Pielou 1975) was calculated for each sample as E =H/logS, where S is the number of DGGE bands (used to indicatethe number of species).

Bands Excision and CloningAt least one band for each band position detected was excised

using a sterile scalpel and incubated overnight in 50 µL sterilemilli-Q purified water at 4◦C. An aliquot of DNA solution thusprepared was 1:10 diluted and used as template for a further25 cycles PCR. The obtained PCR products were separated in aDGGE gel to confirm the recovery of the desired bands (Bodelieret al. 2005). Afterwards, the bands were cut, amplified withthe same set of primers without the GC-clamp, and the PCRproducts were inserted into the vector pCR2.1 R© (TA Cloning R©Kit, Invitrogen, Carlsbad, CA, USA). Seven clones per bandwere selected and DNA was extracted, amplified and, separatedin a DGGE gel together with the original sample to check boththe accuracy and the position of the bands in the gel.

Sequencing and Sequence AnalysisPCR products from selected clones (1 per target band) were

purified with the Jetquick PCR Product Purification Spin Kit(Genomed, Lohne, Germany) and used as template in the se-quencing reactions done with the M13R primer (Stab-Vida,Oeiras, Portugal). Band sequences were compared with se-quences available in public databases to determine their phy-logenetic closest relatives using the software BLAST (Altschulet al. 1997) from the National Center of Biotechnology In-formation (http://www.ncbi.nlm.nih.gov/BLAST/). Nucleotidesequences reported in this work have been deposited in the Gen-Bank Database under accession numbers FJ843129-FJ843163and GQ119348.

Principal Component Analysis (PCA)The multivariate statistical method PCA was used to examine

relationships between chemical and microbiological features ina data set. The PCA method was implemented according toGabriel (1971) and Huang et al. (2010). Eigenvalues and eigen-vectors were extracted from the covariance matrix of originalvariables to illustrate associations among variables, that other-wise might be hidden. The complete dataset used in this studywas a two-way table of 15 columns with the variables total andspecific number of DGGE bands and geochemical parameters(water content, LOI, TP, TN, and TFe) and 15 rows with the ob-servations obtained at five different sediment depths. All mathe-matical and statistical computations were made using XLSTAT2011.2.05 (Addinsoft, Paris, France).

Dow

nloa

ded

by [

Cho

ngqi

ng U

nive

rsity

] at

18:

43 3

0 M

ay 2

013

BACTERIAL DIVERSITY IN EUTROPHIC AZOREAN LAKES 707

RESULTS

Lake Characterization: Water Qualityand Geochemical Profiles

Lakes Verde and Azul were thermally stratified, from July toNovember, as depicted in Figure 1. Lake Furnas was well mixedmost of the time due to its lower depth (11 m maximum depth)compared to lakes Verde (24 m maximum depth) and Azul(25 maximum depth). Thermal stratification limits moleculardiffusion between the epilimnion and the hipolimnion leadingto a rapidly consumption of available O2 in lakes Verde andAzul, while the concentration of dissolved oxygen in lake Furnaswas always above 80% of saturation. pH profiles in the watercolumn varied between 6.7 and 7.7 in the studied lakes, withthe exception of the profiles obtained in July and Septemberin lakes Verde and Azul. In these last profiles, the pH valuesnear the lake surface increases to basic values; this might bedue to the uptake of inorganic carbon by phytoplankton duringphotosynthesis (Hansen 2002).

The nitrate and ammonium profiles in the water column werehomogeneous in lake Furnas. Lake Verde and lake Azul pre-sented higher values in July, September, October and Novembernear the sediments due to thermal stratification. The prevalence

of aerobic conditions might have stimulated the activity of nitri-fying bacteria while, in the other lakes, the denitrifying bacteriawere favoured by the occurrence of anoxic conditions (Rysgaardet al. 1994).

The accumulation of P near the sediments was higher in lakeVerde than in lake Azul, under anoxic conditions, despite ofthe fact that both lakes presented similar thermal stratificationprofiles. In lake Furnas, P did not accumulate near the sedimentsprobably due to the well mixed conditions in the lake.

Vertical profiles of water content and LOI in sediments aredepicted in Figure 2. The average values of water content (91 ±1%) and LOI (18 ± 1%) were higher in the sediment from lakeVerde than in sediments from the other two lakes. TP and TNvertical profiles showed that in the case of lake Verde the amountof both nutrients were higher (about twice) in the uppermostsediment layer than deeper inside the sediment where the profilewas quite homogeneous (Figure 2).

Vertical profiles of TP and TN in sediments from lakes Azuland Furnas were relatively uniform with sediment depth. Con-centrations of TN and TP in the uppermost sediment layer were1.31 mg g−1 N and 2.10 mg g−1 P for lake Verde, 0.19 mgg−1 N and 0.95 mg g−1 P for lake Azul, and 0.20 mg g−1 N and0.85 mg g−1 P for lake Furnas, respectively. The vertical profiles

FIG. 1. Water column profiles for temperature, dissolved oxygen, pH, nitrate, ammonium and TP in lakes Verde (a), Azul (b) and Furnas (c).

Dow

nloa

ded

by [

Cho

ngqi

ng U

nive

rsity

] at

18:

43 3

0 M

ay 2

013

708 G. MARTINS ET AL.

FIG. 2. Vertical profiles of water content, LOI, nutrients (TP and TN), and TFe in sediments from Azorean lakes; error bars represent standard deviations.

of TFe in sediments from lake Verde and lake Azul had higherconcentrations in the uppermost sediment layer, and decreasingconcentrations with sediment depth (Figure 2). In lake Furnas,the amount of Fe in the sediment was maximum below the oxy-gen penetration layer, experimentally measured in the presentstudy (Figure 3).

The sediments reference conditions determined in the con-trol site, the oligotrophic lake Fogo, presented average valuesof water content (61 ± 1%) and LOI (6 ± 1%) considerablylower than that of eutrophic lakes Verde, Azul, and Furnas.The concentration of TP in the uppermost sediment layer was0.2 mg g−1 P for lake Fogo, being in average ten times lowerthan in lake Verde and 5 times lower than in lakes Azul e Furnas.

The vertical profile of TFe in sediments from lake Fogowas similar to that of lake Furnas, which might be explainedby oxygen penetration in sediments due to the presence 90%oxygen saturation in the overlying water. The baseline condition

FIG. 3. Dissolved oxygen and pH profiles in sediments from lake Furnas.

of 0.2 mg g−1 P seems to be the target of future remediationmeasures to be implemented in the eutrophic lakes Verde, Azul,and Furnas.

Microprofiles of dissolved oxygen and pH determined in sed-iments from lake Furnas are depicted in Figure 3. The dissolvedoxygen concentration decreased from 7.17 mg L−1, in the water-air interface, to 0.0 mg L−1, 0.25 cm below the sediment-waterinterface. Similar values for oxygen penetration were alreadypresented in literature (Altmann et al. 2004; Himmelheber et al.2009). Oxygen microprofiles are not expected to occur in thesediments from lakes Verde and Azul because they presenteda dissolved oxygen concentration in the sediment-water inter-face close to zero. The pH profile in the sediment from lakeFurnas, obtained in the same depth as the oxygen profile, washomogeneous with an average value of 6.42 (Figure 3) .

DGGE FingerprintingThe bacterial community composition inhabiting sediments

from Azorean lakes was studied by DGGE fingerprinting of a16S rRNA gene fragment. Figure 4a depicts the DGGE gel.DGGE profiles of sediment layers collected at several depthswere quite similar. The number of bands in each profile rangedfrom 6 to 13; using the same primer set, similar results have beenreported in literature for lake sediments (Nelson et al. 2007; Quet al. 2008). Most of the bands were shared by all sedimentprofiles; nevertheless, a few bands were exclusively present insome profiles.

For example, bands AV2 (sediment depth 0–1 cm and 2.5–3.5 cm), G47 (sediment depth 5–6 cm), and I8 (sediment depth1–2 cm) were merely observed in profiles from lake Verde. BandC19 was only present in lake Furnas (sediment depth 0–1 cmand 1–2 cm) and band C22 was detected solely in lake Verde andlake Furnas (sediment depth 0–1 cm and 1–2 cm). Bands AV1 orA10 present in all sediment profiles were the most intense. Bandsabove AV1 and A10 were disregarded in this study because they

Dow

nloa

ded

by [

Cho

ngqi

ng U

nive

rsity

] at

18:

43 3

0 M

ay 2

013

BACTERIAL DIVERSITY IN EUTROPHIC AZOREAN LAKES 709

FIG. 4. (a) Denaturing gradient gel electrophoresis profiles of PCR products obtained with the bacterial primer set 357f-GC/518r from sediments of lake Verde(V1-V5), lake Furnas (F1-F5), and lake Azul (A1-A5). Sediments from each lake are identified on top of the gel; numbers represent the depth of the sedimentlayers: 1) 0–1 cm; 2) 1–2 cm; 3) 2.5–3.5 cm; 4) 5–6 cm; 5) 9–10 cm. Annotated DGGE bands were further identified by cloning and sequencing. (b) Dendrogramwith the similarity analysis of sediment profiles.

appeared in different positions in the confirmation gel, probablydue to contamination with other DNA fragments.

Based on DGGE profiles, the dendrogram depicted inFigure 4b was constructed. The similarity analysis showed thatthree main clusters could be identified corresponding to thethree lakes. The sediment bacterial community from lake Azulwas more distantly related to the ones of lake Verde and lakeFurnas. Within each cluster, samples grouped mainly accordingto depth. However, communities from different depths were notgrouped similarly in the three lakes.

For example, in lake Verde, the community from the upper-most sediment layer (V1) was distantly related to the commu-nities from other depths, while in lake Furnas and lake Azul,the community from the uppermost sediment layer (F1 andA1) grouped with the ones from contiguous depths (F2 andA2). In general, the Shannon diversity index suggested that sed-iments from the different lakes presented similar diversities,which might indicate a stable maintenance of a structurally di-verse microbial community within these lakes. The equitabilityindex suggested an almost uniform distribution of taxa within

sediments (E = 0.95–1.03 for lake Verde, 0.98–1.05 for lakeFurnas, and 0.93–1.00, for lake Azul), which may indicate sim-ilar abundances of all species identified (Moura et al. 2009).

Phylogenetic AffiliationA total of 36 bands were excised, cloned and sequenced after

DGGE analysis of amplicons obtained by the application of 16SrRNA-specific primers (Table 1). All sequences retrieved in thepresent study were related to uncultured bacteria sequences inpublic databases. Two exceptions were the sequence A10 affil-iated with an unidentified eubacterium clone and C22 affiliatedwith Microcystis sp.. Of the 36 sequences obtained, 27 had asimilarity higher than 97% with previously described sequences,7 had a similarity lower than 94%, suggesting the presence ofpreviously undescribed bacterial diversity (Yang et al. 2001),and only 1 (F49) had a similarity lower than 91%.

Phylogenetic analysis revealed that most of the bacterial16S-rRNA sequences retrieved in the present study could beassigned to class Alpha-Proteobacteria (7 sequences), phylumChloroflexi (5 sequences), and group Bacteroidetes/Chlorobi

Dow

nloa

ded

by [

Cho

ngqi

ng U

nive

rsity

] at

18:

43 3

0 M

ay 2

013

710 G. MARTINS ET AL.

FIG. 5. PCA biplots of geochemical parameters and number of DGGE bands for each detected phyla; a) plot of variables; b) plot of observations.

(5 sequences). Sequences affiliated to Actinobacteria (3sequences), Delta-Proteobacteria (2 sequences), Gamma-Proteobacteria (2 sequences), Gemmatimonadetes (1 se-quence), Cyanobacteria (1 sequence), and Candidate divisionsOP1 (1 sequence) and OP11 (1 sequence) were less abundant.

Sequences affiliated to Alpha-Proteobacteria were retrievedfrom sediments of the three Azorean lakes. Three of thesesequences were closely related to Brevundimonas aurantiacastrain 210-31 (band E35, more than 98% similarity) and Sph-ingomonas sp. (bands E31 and F48, 100% similarity) that arecarbon oxidizers, while two sequences were closely related toNitrobacter sp. (band E30, 100% similarity) and Erythrobactersp. NH89-70 (F53, 100% similarity) that are involved in thecycling of nitrogen.

Sequences belonging to the phylum Chloroflexi (band A3,B14, G36, I61 and I63) were retrieved from sediments ofthe three studied lakes; their closest relatives found in pub-lic databases were all obtained from river and sea sediments(Table 1). The environmental function of the phylum Chloroflexiis not yet fully investigated; literature studies reported that theymay contribute to the degradation of recalcitrant organic matter(Miura et al. 2007; Daniel et al. 2009).

The majority of sequences affiliated to the Bacteroidetes/Chlorobi group were retrieved from sediments of all studiedlakes (bands A15, B13, C21, D25) and band I8 was only re-trieved from lake Verde. Bacteroidetes are known as hydrolyticfermentative degraders of polymers in anaerobic habitats, in-cluding freshwater sediments (Kirchman 2002; Schwarz et al.2007). Less abundant sequences, but present in sediments fromthe three studied lakes, were affiliated to Candidate divisionOP11, commonly related to marine sediments (Harris et al.2004), Candidate division OP1, whose closest relative retrievedin public data bases was found in a hydrothermal vent biofilm(Stott et al. 2008), and Gemmatimonadetes, retrieved from lake

sediments (Qu et al. 2008; Briee et al. 2007) and deep-sea sedi-ments (Li et al. 1999). The presence of candidate division OP1 insediments from Azorean lakes might be related to their volcanicorigin (Stott et al. 2008; Hugenholtz et al. 1998). The environ-mental function of these microbial groups remains relativelyunknown.

The molecular approach used in this study presented twomain drawbacks: i) seven of the retrieved sequences could notbe assigned to a specific phylum (designated as unclassified inTable 1) probably due to either the restricted number of basesamplified with the primers (341F, 518R) used (Simpson et al.1999; Crosby and Criddle, 2003) or the novelty of the bacterialcommunities; ii) bands AV1 and A10, which presented the sameGC percentage, comigrated to the same position in the DGGEgel but were affiliated to different closest relatives suggesting alimitation of the DGGE method (Jackson et al. 2000; Siqueiraet al. 2005).

Principle Component AnalysisTo interpret the relationships between geochemical parame-

ters and bacterial community composition in sediments, a PCAanalysis with extracted principal components (PCs) was con-ducted. Two PCs with eigenvalues >1 were retained and cap-tured about 79% of the input variance using 15 observationsand 15 variables. A PCA biplot for PC1 vs PC2 (Figure 5)demonstrated that sediment bacterial community and geochem-ical variables, drawn as vectors, occupy different regions of theplot and exhibit well-defined patterns.

PC1 explained 51% of total variance and had positive load-ings for all variables (Figure 5a). PC2 explained an additional28% of the original variability and had negative loadings as wellas positive loadings on TFe, TP, TN, and number of Cyanobac-teria bands and unclassified bands. In both PCs Gemmatimon-adetes had a null contribution. Biplots of each PC (Figure 5a)

Dow

nloa

ded

by [

Cho

ngqi

ng U

nive

rsity

] at

18:

43 3

0 M

ay 2

013

BACTERIAL DIVERSITY IN EUTROPHIC AZOREAN LAKES 711

TABLE 1Sequence analysis of bands excised from DGGE gel

Clone(accession Closest relative Query Max Isolationnumber) (accession number) coverage identity Taxonomy site

AV1 (FJ843129) Uncultured bacterium cloneLWS-RSG-CH4-4075 16S ribosomalRNA gene, partial sequence(EU546553.1)

100% 98% Unclassified Lake sediment

AV2 (FJ843130) Uncultured bacterium clone FFCH830816S ribosomal RNA gene, partialsequence (EU133203.1)

99% 100% Actinobacteria Soil

AV4 (FJ843131) Uncultured bacterium partial 16S rRNAgene, clone M6A-304 (AM991224.1)

100% 91% Actinobacteria Karst spring water

A3 (FJ843132) Uncultured bacterium clone LWS-T484616S ribosomal RNA gene, partialsequence (EU546326.1)

100% 100% Chloroflexi Lake sediment

A7 (FJ843133) Uncultured bacterium partial 16S rRNAgene, specimen voucher bandDNA17(AM040914.1)

100% 98% CandidateDivisionOP11

Eutrophic lakeLoclat,hypolimnion

A10 (FJ843134) Unidentified eubacterium clonevadinBA43 16S ribosomal RNA gene,partial sequence (U81652.2)

98% 92% Unclassified Fluidized bedanaerobic digestor

A15 (FJ843135) Uncultured bacterium cloneGASP-0KB-587-C02 16S ribosomalRNA gene, partial sequence(EU043640.1)

98% 97% Bacteroidetes/Chlorobigroup

Inland dune fields

B11 (FJ843136) Uncultured candidate division OP1bacterium partial 16S rRNA gene, cloneMS12-2-H02 (AM712329.1)

96% 92% CandidateDivision OP1

Hydrothermal ventbiofilm

B12 (FJ843137) Uncultured bacterium partial 16S rRNAgene, clone AV9-153 (AM181873.1)

100% 94% Gamma-proteobacteria

Lake sediment

B13 (FJ843138) Uncultured bacterium clone 101b1 16Sribosomal RNA gene, partial sequence(EF459907.1)

100% 99% Bacteroidetes/Chlorobigroup

Sea sediment

B14 (FJ843139) Uncultured bacterium clone PR OTU-0816S ribosomal RNA gene, partialsequence (EF165510.1)

100% 99% Chloroflexi River sediment

C18 (FJ843140) Uncultured bacterium clone 2G1-22 16Sribosomal RNA gene, partial sequence(EU160208.1)

96% 97% Gemmati-monadetes

Rhizosphere soil

C19 (FJ843141) Uncultured bacterium DGGE gel bandESR BR 17 16S ribosomal RNA gene,partial sequence (AF540052.1)

100% 99% Unclassified Lake water

C21 (GQ119348) Uncultured bacterium clone FL0428BPF55 16S ribosomal RNA gene, partialsequence (FJ716474.1)

100% 99% Bacteroidetes/Chlorobigroup

Frasassi cavesystem, anoxiclake water

C22 (FJ843142) Microcystis sp. AWT139 16S ribosomalRNA gene, partial sequence (U40331.2)

100% 100% Cyanobacteria Microcystis sp.AWT139

D24 (FJ843143) Uncultured alpha proteobacterium cloneSA-B16 16S ribosomal RNAgene,partial sequence (DQ295442)

100% 99% Alpha-proteobacteria

Rhizosphere -wetland

D25 (FJ843144) Uncultured bacterium clone GD71 16Sribosomal RNA gene, partial sequence(EF613895.1)

98% 93% Bacteroidetes/Chlorobigroup

Paddy soil

Dow

nloa

ded

by [

Cho

ngqi

ng U

nive

rsity

] at

18:

43 3

0 M

ay 2

013

712 G. MARTINS ET AL.

TABLE 1Sequence analysis of bands excised from DGGE gel (Continued)

Clone(accession Closest relative Query Max Isolationnumber) (accession number) coverage identity Taxonomy site

D28 (FJ843145) Uncultured bacterium clone reef124 16Sribosomal RNA, partial sequence(EU121710.1)

98% 92% Unclassified Coral reef sediment

E30 (FJ843146) Uncultured bacterium clonePro CL-09054 OTU-5 16S ribosomalRNA gene, partial sequence(EU808746.1)

100% 100% Alpha-proteobacteria

Chloraminateddistribution system -Prospect Tank

E31 (FJ843147) Uncultured bacterium clone G10-6 16Sribosomal RNA gene, partial sequence(GQ487961.1)

100% 100% Alpha-proteobacteria

Soil polluted by heavymetals

E34 (FJ843148) Uncultured bacterium cloneADK-WSe02-91 16S ribosomal RNAgene, partial sequence (EF520621.1)

100% 95% Unclassified Acid-impacted lake

E35 (FJ843149) Uncultured bacterium gene for 16S rRNA,clone:pHAuB-1, partial sequence(AB072705)

100% 99% Alpha-proteobacteria

Subsurfaceenvironment - goldmine

F48 (FJ843150) Uncultured soil bacterium clone TF5 16Sribosomal RNA gene, partial sequence(DQ248300.1)

100% 100% Alpha-proteobacteria

Carbon tetrachloridecontaminated soil

F49 (FJ843151) Uncultured bacterium clone Er-LLAYS-6216S ribosomal RNA gene, partialsequence (EU542519.1)

98% 86% Unclassified Sediment and soilslurry

F53 (FJ843152) Uncultured bacterium HF0500 24B12genomic sequence (EU795192.1)

100% 100% Alpha-proteobacteria

HOT deep-waterstation - ALOHA

F54 (FJ843153) Uncultured bacterium clone FCPT525 16Sribosomal RNA gene, complete sequence(EF516073.1)

100% 97% Gamma-proteobacteria

Grassland soil

G36 (FJ843154) Uncultured bacterium clone PR OTU-1116S ribosomal RNA gene, partialsequence (EF165507)

100% 98% Chloroflexi River sediment

G37 (FJ843155) Uncultured soil bacterium PBS-III-30partial 16S rRNA gene (AJ390458.1)

100% 95% Unclassified Soil and rice roots

G47 (FJ843156) Uncultured bacterium clone 21 1 16Sribosomal RNA gene, partial sequence(FJ390446.1)

100% 100% Actinobacteria Water from MiyunReservoir

H55 (FJ843157) Uncultured Sphingomonadales bacteriumclone SHBZ696 16S ribosomal RNAgene, partial sequence (EU639137.1)

99% 100% Alpha-proteobacteria

Thermophilicmicrobial fuel cell

H59 (FJ843158) Uncultured bacterium partial 16S rRNAgene, clone AV9-10 (AM181950.2)

100% 100% Unclassified Lake sediment

I8 (FJ843159) Uncultured bacterium clone ORS25C b0416S ribosomal RNA gene, partialsequence (EF392979.1)

100% 99% Bacteroidetes/Chlorobi group

River sediment

I61 (FJ843160) Uncultured bacterium clone SED1000 7416S ribosomal RNA gene, partialsequence (EU557829.1)

100% 100% Chloroflexi Stream sediment

I62 (FJ843161) Uncultured bacterium partial 16S rRNAgene, clone c1LKS29 (AM086080.1)

100% 98% Delta-proteobacteria

Lake sediment

I63 (FJ843162) Uncultured bacterium clone LaC15L90 16Sribosomal RNA, partial sequence(EF667608.1)

100% 100% Chloroflexi River sediment

I64 (FJ843163) Uncultured Syntrophaceae bacterium cloneLCA1-1C 16S ribosomal RNA gene,partial sequence (EU522632.1)

98% 97% Delta-proteobacteria

Oil sand tailingsenrichment culture

Dow

nloa

ded

by [

Cho

ngqi

ng U

nive

rsity

] at

18:

43 3

0 M

ay 2

013

BACTERIAL DIVERSITY IN EUTROPHIC AZOREAN LAKES 713

also illustrated the clustering of different variables. On the oneside, TP cluster very well with Cyanobacteria bands. On theother side, TFe and total number of DGGE bands (N) seemedto be isolated from the other variables. Also, Candidate divisionOP1 and OP11 clustered together as well as Gemmatimonadetesand Proteobacteria bands. This was mainly due to the similardistribution of these bands in the three lakes.

DISCUSSIONThis study assessed the dominant members of the bacterial

community present in sediments from three nutrient impactedlakes (Verde, Azul, and Furnas) as well as the geochemicalprofiles. The detected bacterial diversity was consistent withprevious published studies (Tamaki et al. 2005; Kojima et al.2006; Nelson et al. 2007, Briee et al. 2007, Schwarz et al. 2007;Qu et al. 2008; Zeng et al. 2009). It was found that sedimentbacterial diversity was quite similar in the three Azorean lakes,as was suggested by the similarity of the equitability indexvalues obtained, but a few exceptions can be described.

The Cyanobacteria phylum (band C22) was only retrievedfrom sediments of lake Furnas and lake Verde, which presentedthe highest amounts of nitrogen and phosphorus both in thewater column and sediments (Figures 1 and 2). PCA also showeda cluster between TP, TN, and Cyanobacteria bands (Figure 4a).Previous studies of the bacterioplankton community in bothlakes (Santos et al. 2005; Martins et al. 2008) reported thepresence of Microcystis and other Cyanobacteria species in thewater column.

The presence of these photosynthetic microorganisms in sed-iments was already reported in literature and might be attributedto the deposition of the dead biomass (Kalyuzhnaya et al. 2008).Regarding the sequences affiliated to Actinobacteria, they weremore abundant in sediments from lake Verde (bands AV2, AV4,and G47) than in the ones from the other two lakes (band AV4).This higher incidence of Actinobacteria in lake Verde mightbe due to the higher amount of LOI observed in the upper-most layer of the sediment (Figure 2). PCA also showed acluster between Actinobacteria bands, LOI, and water content(Figure 5a).

In addition, members of the Actinobacteria and Gemmati-monadetes phyla, both involved in the P cycle (Seviour et al.2008; Zhang et al. 2003), were retrieved from Azorean lakessediments. Similar results were reported for an eutrophic reser-voir, with a phosphorus content in the water body (41–136 µgL−1 P) and sediments (500–850 µgP g−1) similar to the onesobserved in the present study. Nevertheless, a direct relation-ship between detected microorganisms and their environmentalfunction cannot be directly established based on molecular phy-logenetic approach (Watanabe and Hamamura, 2003).

The present study suggested that sediment depth seemed notto be a significant environmental factor affecting the diversityof the bacterial community, as already reported in literaturefor shallow lakes (Zeng et al. 2009). The absence of a defined

cluster in the plot of observations of PCA (Figure 5b) and aconsistent change in bacterial diversity in the sediment acrossdepth might reflect a discontinuous change in the environmentduring sedimentation (Qu et al. 2008).

The higher amounts of LOI, TP, and TN in the upper sedi-ment layer of lake Verde might indicate a potential for significantinternal nutrient cycling of P from sediments to the water col-umn, thus supporting phytoplankton growth (Hickey and Gibbs2009). The amount of TP in the uppermost layer of sedimentsseemed to be correlated with the annual average TP concentra-tion in the water column, for the studied lakes. The same wasnot observed for TN. As an example, lake Verde presented thehighest amount of TN in the uppermost layer of sediment, butthe highest annual average amount of TN in the water columnwas found in lake Furnas.

Regarding TFe profiles, it was interesting to observe thatTFe increased below the oxygen penetration layer in sedimentsfrom lake Furnas, which might be due to the oxidation of Fe(II)to Fe(III) by iron reducing bacteria (Koretsky et al. 2006). Insediments, Fe oxides/hydroxides sorb considerable amounts of Pthat might be released by the activity of iron reducing bacteria,mainly from Geobacteraceae and Shewanelanaceae families(Azzoni et al. 2005).

Jensen and Andersen (1992) proposed a threshold value of15 for the Fe:P ratio, below which P will be released fromsediments despite the occurrence of aerobic conditions in thesediment-water column interface. The referred ratio in sedi-ments from Azorean lakes were between 3 and 6 (6 for lakeFurnas, 5 for lake Azul, and 3 for lake Verde), thus consider-ably lower than 15, suggesting that remediation methods suchas artificial destratification and hypolimnetic aeration will notsolve the problem of internal P cycling. In the control site, theoligotrophic lake Fogo, the Fe:P ratio was around 20, meaningthat the sediment did not exhaust its sorption capacity for P. As apossible remediation approach for the euthophic Azorean lakes,it is suggested the addition of an adsorbent capping agent tothe surface of the sediments, as for example, a hybrid polymercontaining aluminium (Oliveira et al. 2010). However, it mustbe stressed that the restoration of a lake through management ofsediment nutrient release is only one part of what should be anintegrated approach including catchment nutrient management(Hickey and Gibbs 2009).

The successfully implementation of the remediation goalsshould considerer the effects of remediation processes in thesediment bacterial community and consequently in the biogeo-chemical. As an example, laboratory studies have shown that,although sediment capping agents sequester P, they can also af-fect nitrification and denitrification processes in sediments dueto the increased thickness of the diffusive boundary layer (Vopelet al. 2008; Hickey and Gibbs 2009). Further research is alsoneeded to infer the environmental function of sediment bacte-rial community in Azoreans lakes; for example, the relationshipbetween some bacterial groups and P release should be clari-fied, as well as on the evaluation of nutrients (N and P) and

Dow

nloa

ded

by [

Cho

ngqi

ng U

nive

rsity

] at

18:

43 3

0 M

ay 2

013

714 G. MARTINS ET AL.

iron fluxes from sediments to the water column under differentconditions.

In conclusion, the present study suggested that geochemicalprofiles of LOI, nutrients (TP and TN), and TFe were differentfor the different lakes (Verde, Azul and Furnas). The internal Pcontribution to eutrophication was higher in lake Verde due tothe higher amounts of P and LOI in the top sediment layer. Al-though dominant members of the sediment bacterial communityin the studied three lakes were similar (mostly affiliated to phy-lum Proteobacteria, group Bacteroidetes/Chlorobi and phylumChloroflexi), the Cyanobacteria phylum was solely detected insediments from lake Verde and lake Furnas that presented thehighest amounts of nitrogen and phosphorus both in the wa-ter column and sediments. Finally, the variability of sedimentLOI, nutrients, and TFe profiles in the Azorean lakes suggestedthat individual lake sediment characterization and site-specificassessments are required for planning future remediationmeasures.

REFERENCESAltmann D, Stief P, Amann R, Beer DD. 2004. Distribution and activity of

nitrifying bacteria in natural stream sediment versus laboratory sedimentmicrocosms. Aquat Microb Ecol 36:73–81.

Altschul SF, Madden TL, Schaffer AA, Zhang J, Zhang Z, Miller W, LipmanDJ. 1997. Gapped BLAST and PSI-BLAST: a new generation of proteindatabase search programs. Nucleic Acids Res 25:3389–3402.

APHA, AWWA, WPCF. 1998. Standard methods for the examination of waterand wastewater, 20th ed. Washington, DC: American Public Health Associ-ation.

Azzoni R, Giordani G, Viaroli P. 2005. Iron–sulphur–phosphorus interactions:implications for sediment buffering capacity in a mediterranean eutrophiclagoon (Sacca di Goro, Italy). Hydrobiologia 550(1):131–148.

Bodelier PLE, Meima-Franke M, Zwart G, Laanbroek HJ. 2005. New DGGEstrategies for the analyses of methanotrophic microbial communities usingdifferent combinations of existing 16S rRNA-based primers. FEMS MicrobiolEcol 52:163–174.

Briee C, Moreira D, Lopez-Garcıa P. 2007. Archaeal and bacterial communitycomposition of sediment and plankton from a suboxic freshwater pond. ResMicrobiol 158(3):213–227.

Carpenter SR. 2005. Eutrophication of aquatic ecosystems: Bistability and soilphosphorus. Proc Natl Acad Sci 102(29):10002–10005.

Chen G, White PA. 2004. The mutagenic hazards of aquatic sediments: a review.Mutat Res 567:151–225.

Clarke KR, Gorley RN,(2001) PRIMER v5: User manual/tutorial. Plymouth,UK: PRIMER-E, p.91.

Conley DJ, Paerl HW, Howarth RW, Boesch DF, Seitzinger SP, Havens KE,Lancelot C, Likens GE. 2009. Controlling eutrophication: nitrogen and phos-phorus. Science 323:1014–1015.

Crosby LD, Criddle CS. 2003. Understanding systematic error in microbialcommunity analysis techniques as a result of ribosomal RNA (rrn) operoncopy number. BioTechniques 34:790–803.

Daniel LMC, Pozzi E, Foresti E, Chinalia FA. 2009. Removal of ammonium viasimultaneous nitrification-denitrification nitrite-shortcut in a single packed-bed batch reactor. Bioresource Technol 100(3):1100–1107.

Gabriel KR. 1971. Biplot graphic display of matrices with application to prin-cipal component analysis. Biometrika 58(3): 453–467.

Hansen PJ. 2002. Effect of high pH on the growth and survival of ma-rine phytoplankton: implications for species succession. Aquat Microb Ecol28:279–288.

Harris JK, Kelley ST, Pace NR. 2004. New perspective on uncultured bacterialphylogenetic division OP11. Appl Environ Microbiol 70:845–849.

Haukka K, Kolmonen E, Hyder R, Hietala J, Vakkilainen K, Kairesalo T, HaariH, Sivonen K. 2006. Effect of nutrient loading on bacterioplankton commu-nity composition in Lake Mesocosms. Microb Ecol, 51:137–146.

Herrmann M, Saunders AM, Schramm A. 2009. Effect of Lake trophic sta-tus and rooted macrophytes on community composition and abundance ofammonia-oxidizing prokaryotes in freshwater sediments. Appl Environ Mi-crobiol 75(10):3127–3136.

Hickey CW, Gibbs MM. 2009. Lake sediment phosphorus release management-decision support and risk assessment Framework. New Zeal J Mar FreshwatRes 43:819–856.

Himmelheber DW, Thomas SH, Loffler FE, Taillefert M, Hughes JB. 2009.Microbial colonization of an in situ sediment cap and correlation to stratifiedredox zones. Environ Sci Technol 43:66–74.

Huang Z, Gedalanga PB, Asvapathanagul P, Olson BH. 2010. Influenceof physicochemical and operational parameters on Nitrobacter and Nitro-spira communities in an aerobic activated sludge bioreactor. Water Res44:4351–4358.

Huettel M, Røy H, Precht E, Ehrenhauss S. 2003. Hydrodynamical impact onbiogeochemical processes in aquatic sediments. Hydrobiologia 494:231–236.

Hugenholtz P, Pitulle C, Hershberger KL, Pace NR. 1998. Novel division levelbacterial diversity in a Yellowstone hot spring. J Bacteriol 180(2):366–376.

Jackson CR, Roden EE, Churchill, PF. 2000. Denaturing gradient gel elec-trophoresis can fail to separate 16S rDNA fragments with multiple basedifferences. Mol Biol Today 1(2):49–51.

Jensen HS, Andersen FO. 1992. Importance of temperature, nitrate, and pH forphosphate release from aerobic sediments of four shallow, eutrophic lakes.Limnol Oceanogr 37(3):577–589.

Kalyuzhnaya MG, Lidstrom ME, Chistoserdova L. 2008. Real-time detection ofactively metabolizing microbes by redox sensing as applied to methylotrophpopulations in Lake Washington. ISME J 2:696–706.

Kelderman P, Wei Z, Maessen M. 2005. Water and mass budgets for estimatingphosphorus sediment–water exchange in Lake Taihu (China PR). Hydrobi-ologia 544:167–175.

Kirchman DL. 2002. The ecology of Cytophaga–Flavobacteria in aquatic envi-ronments. FEMS Microbiol Ecol 39:91–100.

Kojima H, Koizumi Y, Fukui M. 2006. Community structure of bacteria associ-ated with sheaths of freshwater and brackish Thioploca species. Microb Ecol52:765–773.

Koretsky CM, Haas JR, Miller D, Ndenga NT. 2006. Seasonal variations in porewater and sediment geochemistry of littoral lake sediments (Asylum Lake,MI, USA). Geochem Trans 7:11.

Li C, Yuan HL, Huang HZ. 2005. Vertical distribution of phosphorus and P-dissolving/decomposing bacteria in the sediment of Guanting reservoir. SciChina Ser D 48:285–294.

Li L, Kato C, Horikoshi K. 1999. Bacterial diversity in deep-sea sediments fromdifferent depths. Biodivers Conserv 8:659–677.

Martins G, Ribeiro DC, Pacheco D, Cruz JV, Cunha R, Goncalves V, NogueiraR, Brito AG. 2008. Prospective scenarios for water quality and ecologicalstatus in Lake Sete Cidades (Portugal): the integration of mathematical mod-elling in decision processes. Appl Geochem 23:2171–2181.

Miura Y, Watanabe Y, Okabe S. 2007. Significance of Chloroflexi in per-formance of submerged membrane bioreactors (MBR) treating municipalwastewater. Environ Sci Technol 41(22):7787–7794.

Moura A, Tacao M, Henriques I, Dias J, Ferreira P, Correia A. 2009. Character-ization of bacterial diversity in two aerated lagoons of a wastewater treatmentplant using PCR-DGGE analysis. Microbiol Res 164(5):560–569.

Muyzer G, De Waal EC, Uitterlinden AG. 1993. Profiling of complex microbialpopulations by denaturing gradient gel electrophoresis analysis of polymerasechain reaction-amplified genes coding for 16S rRNA. Appl Environ Microbiol59:695–700.

Nealson KH. 1997. Sediment bacteria: who’s there, what are they doing, andwhat’s new? Annu Rev Earth Planet Sci 25:403–434.

Dow

nloa

ded

by [

Cho

ngqi

ng U

nive

rsity

] at

18:

43 3

0 M

ay 2

013

BACTERIAL DIVERSITY IN EUTROPHIC AZOREAN LAKES 715

Nelson DM, Ohene-Adjei S, Hu FS, Cann IKO, Mackie RI. 2007. Bacterialdiversity and distribution in the Holocene sediments of a northern temperatelake. Microb Ecol 54:252–263.

Oliveira, M., Machado, A.V., Nogueira, R.,(2010). Development of Per-meable Reactive Barrier for Phosphorus Removal. Mater Sci Forum,636–637:1365–1370.

Pielou EC. 1975. Ecological Diversity. New York: Wiley.Polymenakou PN, Bertilsson S, Tselepides A, Stephanou EG. 2005. Bacterial

community composition in different sediments from the eastern Mediter-ranean Sea: a comparison of four 16S ribosomal DNA clone libraries. MicrobEcol 50:447–462.

Qu JH, Yuan HL, Wang ET, Li C, Huang HZ. 2008. Bacterial diversity insediments of the eutrophic Guanting Reservoir, China, estimated by analysesof 16S rDNA sequence. Biodivers Conserv 17:1667–1683.

Ravenschlag K, Sahm K, Pemthaler J, Amann R. 1999. High bacterial di-versity in permanently cold marine sediments. Appl Environ Microbiol65(9):3982–3989.

Redfield AC. 1934. On the proportions of organic derivations in sea water andtheir relation to the composition of plankton. In: James Johnstone MemorialVolume, Daniel, R.J. Liverpool: University Press of Liverpool, p. 177–192.

Reitzel K, Ahlgren J, DeBrabandere H, Waldeback M, Gogoll A, Tranvik L,Rydin E. 2007. Degradation rates of organic phosphorus in lake sediment.Biogeochemistry 82(1):15–28.

Ribeiro DC, Martins G, Nogueira R, Cruz JV, Brito AG. 2008. Phosphorusfractionation in volcanic lake sediments (Azores - Portugal). Chemosphere70:1256–1263.

Rysgaard S, Risgaard-Petersen N, Sloth NP, Jensen K, Nielsen LP. 1994.Oxygen regulation of nitrification and denitrification in sediments. LimnolOceanogr 39:1643–1652.

Santos M, Pacheco D, Santana F, Muelle H. 2005. Cyanobacteria blooms in Sete-Cidades lake (S. Miguel Island—Azores). Algolog Stud 117(6):393–406.

Schwarz JIK, Eckert W, Conrad R. 2007. Community structure of Archaeaand Bacteria in a profundal lake sediment Lake Kinneret (Israel). Syst ApplMicrobiol 30:239–254.

Seviour RJ, Kragelund C, Kong Y, Eales K, Nielsen JL, Nielsen PH. 2008. Eco-physiology of the Actinobacteria in activated sludge systems. Anton Leeuwen94:21–33.

Shannon CE, Weaver W. 1963. The Mathematical Theory of Communication.Urbana, IL: University of Illinois Press.

Simpson JM, McCracken VJ, White BA, Gaskins HR, Mackie RI. 1999. Ap-plication of denaturant gradient gel electrophoresis for the analysis of theporcine gastrointestinal microbiota. J Microbiol Meth 36:167–179.

Siqueira JF, Rocas IN, Rosado AS. 2005. Application of denaturing gradientgel electrophoresis (DGGE) to the analysis of endodontic infections. JOE31(11):775–782.

Søndergaard M, Jensen JP, Jeppesen E. 2003. Role of sediment and inter-nal loading of phosphorus in shallow lakes. Hydrobiologia 506–509:135–145.

Søndergaard M, Windolf J, Jeppesen E. 1996. Phosphorus fractions andprofiles in the sediment of shallow Danish lakes as related to phospho-rus load, sediment composition and lake chemistry. Water Res 30:992–1002.

Stott MB, Saito JA, Crowe MA, Dunfield PF, Hou S, Nakasone E, Daughney CJ,Smirnova AV, Mountain BW, Takai K, Alam M. 2008. Culture-independentcharacterization of a novel microbial community at a hydrothermal vent atBrothers volcano, Kermadec arc. New Zealand J Geophys Res 113:B08S06,doi:10.1029/2007JB005477.

Tamaki H, Sekiguchi Y, Hanada S, Nakamura K, Nomura N, Matsumura M,Kamagata Y. 2005. Comparative analysis of bacterial diversity in freshwatersediment of a shallow eutrophic lake by molecular and improved cultivation-based techniques. Appl Environ Microbiol 71(4):2162–2169.

Vopel K, Gibbs M, Hickey CW, Quinn J. 2008. Modification of sediment-watersolute exchange by sediment capping agents: effects on O2 and pH. MarFreshw Res 59:1101–1110.

Watanabe K, Hamamura N. 2003. Molecular and physiological approaches tounderstanding the ecology of pollutant degradation. Curr Opin Biotechnol14:289–295.

Wobus A, Bleul C, Maassen S, Scheerer C, Schuppler M, Jacobs E, RoskeI. 2003. Microbial diversity and functional characterization of sedimentsfrom reservoirs of different trophic state. FEMS Microbiol Ecol 46:331–347.

Yang CH, Crowley DE, Borneman J, Keen NT. 2001. Microbial phyllospherepopulations are more complex than previously realized. Proc Natl Acad SciUSA 98(7):3889–3894.

Zeng J, Yang L, Li J, Liang Y, Xiao L, Jiang L, Zhao D. 2009. Verticaldistribution of bacterial community structure in the sediments of two eu-trophic lakes revealed by denaturing gradient gel electrophoresis (DGGE)and multivariate analysis techniques. World J Microbiol Biotechnol 25:225–233.

Zhang H, Sekiguchi Y, Hanada S, Hugenholtz P, Kim H,Kamagata Y, Nakamura K. 2003. Gemmatimonas aurantiaca gen.nov., sp. nov., a Gram-negative, aerobic, polyphosphate accumulatingmicro-organism, the first cultured representative of the new bacterial phylumGemmatimonadetes phyl. nov. Int J Syst Evol Microbiol 53:1155–1163.

Dow

nloa

ded

by [

Cho

ngqi

ng U

nive

rsity

] at

18:

43 3

0 M

ay 2

013