Music & Video

A novel phylogenetic clade of picocyanobacteria from the Mazurian lakes (Poland) reflects the early ontogeny of glacial lakes

Description
A novel phylogenetic clade of picocyanobacteria from the Mazurian lakes (Poland) reflects the early ontogeny of glacial lakes
Categories
Published
of 10
All materials on our website are shared by users. If you have any questions about copyright issues, please report us to resolve them. We are always happy to assist you.
Related Documents
Share
Transcript
  RESEARCH ARTICLE AnovelphylogeneticcladeofpicocyanobacteriafromtheMazurianlakes(Poland)re£ectstheearlyontogenyofglaciallakes Iwona Jasser 1 , Adriana Kr´olicka 1 & Anna Karnkowska-Ishikawa 2 1 Microbial Ecology Department, Institute of Botany, University of Warsaw, Warsaw, Poland; and  2 Department of Plant Systematics and Geography,Institute of Botany, University of Warsaw, Warsaw, Poland Correspondence:  Iwona Jasser, MicrobialEcology Department, Institute of Botany,University of Warsaw, ul. Miecznikowa 1, 02-096 Warsaw, Poland. Tel.: 1 48 22 554 1444;fax: 1 48 22 554 1413; e-mail: jasser.iwona@biol.uw.edu.plReceived 6 May 2010; revised 3 August 2010;accepted 3 October 2010.Final version published online 9 November2010.DOI:10.1111/j.1574-6941.2010.00990.xEditor: Patricia Sobecky Keywords picocyanobacteria phylogeny; lake ontogeny;16S rRNA gene;  cpcBA . Abstract The community of picocyanobacteria inhabiting the Great Mazurian Lakes system(comprising lakes ranging from mesotrophic to hypertrophic) is dominated by phycoerythrin-rich cells, which outnumber phycocyanin-rich cells, even in hyper-trophic lakes. The genetic diversity and phylogeny of 43 strains of picocyanobac-teria isolated from four Mazurian lakes were studied by analyzing the nucleotidesequences of the 16S rRNA gene and  cpcBA -IGS operon. Phylogenetic analysesassigned some of the strains to several previously described clusters (Groups A, B,C, E and I) and revealed the existence of a novel clade, Group M (Mazurian),which exhibited a low level of similarity to the other clusters. Both phycocyaninand phycoerythrin picocyanobacteria were assigned to this clade based on ananalysis of the 16S rRNA gene. The  cpcBA  sequence analysis assigned only phycocyanin strains to Group M, whereas the phycoerythrin strains from the Mribogroup were assigned to Groups B and E. We hypothesize that Group Msrcinally contained only phycocyanin picocyanobacteria. The phycoerythrinfound in strains belonging to ribogroup M seems to have been acquired throughhorizontal gene transfer as an adaptation to the changing environment early in theontogeny of these glacial lakes. Introduction Picocyanobacteria, first discovered three decades ago, inha-bit various types of aquatic environment in terms of chemistry, trophic status and underwater light climate, andplay a very important role as primary producers in bothmarine and freshwater ecosystems (Callieri & Stockner,2002). Depending on the trophic status and light dominat-ing the water column, different groups of picocyanobacteriaprevail. Oligo- and mesotrophic clearwater lakes are usually dominated by orange-fluorescing, phycoerythrin-rich pico-cyanobacteria, which efficiently absorb the blue/green lightprevailing in these types of waters. Red-fluorescing, phyco-cyanin-rich picocyanobacteria, which optimally use redlight, are dominant in more productive ecosystems (V¨or¨os et al  ., 1998). Historically, picocyanobacteria were classifiedinto two genera:  Synechococcus  and  Synechocystis  (Water-bury   et al  ., 1986; Andreoli  et al  ., 1989; Ernst  et al  ., 1995).Their discrimination was based on cell size, shape anddivision type. In the late 1990s, Kom´arek and colleagues(Kom´arek, 1996; Kom´arek   et al  ., 1999) proposed a new systematics of picocyanobacteria based on cytomorphologi-cal, biochemical and molecular features. According to thisapproach, all freshwater picocyanobacteria belonging to theso-called  Synechococcus -type were divided into three genera: Synechococcus ,  Cyanobium  and  Cyanobacterium .Molecular techniques such as denaturing gradient gelelectrophoresis analysis and the sequencing of amplifiedgenes have created novel opportunities for the taxonomicand phylogenetic analysis of picoplanktonic autotrophs(Becker  et al  ., 2002; Zeidner & B´eja`, 2004). These methodshave permitted  in situ  studies of community compositionand the seasonal dynamics of picocyanobacteria without theneed to isolate the studied organisms. The cultivation of picocyanobacterial strains has also received more attentionfollowing the introduction of new isolation methods (Cros-bie  et al  ., 2003b). Several recent studies have examined thediversity and phylogenetic relationships of picocyanobacter-ia inhabiting various environments. The results of 16S rRNAgene analyses, which depict the phylogenetic relationship FEMS Microbiol Ecol  75  (2011) 89–98  c   2010 Federation of European Microbiological SocietiesPublished by Blackwell Publishing Ltd. All rights reserved    M   I   C   R   O   B   I   O   L   O   G   Y   E   C   O   L   O   G   Y  between sequences, are considered too conservative to revealrecent differences between picocyanobacterial isolates. Thus,the analyses of less well-conserved genes, such as thephycocyanin  cpcBA  and phycoerythrin  cpeBA  operons andthe internal transcribed spacer region between the 16S and23S rRNA genes, have been carried out (Crosbie  et al  .,2003a; Ernst  et al  ., 2003; Haverkamp  et al  ., 2008). It hasbeen suggested that these analyses may be used to track possible adaptation to the environment (Ernst  et al  ., 2003;Six   et al  ., 2007).The goal of this study was to characterize the geneticdiversity and phylogeny of picocyanobacterial strains isolatedfrom selected Mazurian lakes. The Mazurian region, locatedin north-eastern Poland, comprisesabout2000lakesofglacialsrcin. Several of them, including the two largest lakes inPoland, are interconnected, forming a waterway called theGreat Mazurian Lakes system, which extends for about100km north to south. The lakes represent a gradient of trophic status from mesotrophy to hypertrophy, with numer-ical trophic state index (TSI) values ranging from 40 to  4 70(Carlson, 1977), and a mean chlorophyll  a  content rangingfrom 3.3 to 87.3mgchl a L  1 (Chr´ost & Siuda, 2006). Thereare both shallow and deep lakes (mean depth between 1.1 and14m, maximum from 3 to 50m), with most developingstable thermal stratification of the water column during thesummer. In this environment, phycoerythrin picocyanobac-teria prevail, even in the most eutrophic and hypertrophiclakes, accounting, on average, for  4 90% of picocyanobac-terial abundance in less productive lakes and 55–60% inhypertrophic ones. These results were obtained in a survey of 15 lakes of the Great Mazurian Lakes system and Lake MajczWielki in 2006, and were confirmed in more extensive studiesof selected lakes conducted in 2007 (Jasser  et al  ., 2010). Theunusual predominance ofthephycoerythrinphenotypeintheproductive lakes is intriguing, and poses the question as tohow diverse the picocyanobacterial community in this systemactually is. Another issue is how closely related (phylogeneti-cally) are the phycoerythrin and phycocyanin phenotypesoccurring in these lakes to known strains and clades? Weaddressed these questions by sequencing selected amplifiedmolecular markers ofisolatesobtained fromsome ofthe lakes(Jasser  et al  ., 2010).The specific aims of our study were (1) to characterize thegenetic diversity of isolated picocyanobacterial strains by nucleotide sequence analysis of 16S rRNA and  cpcBA -IGSgenes and (2) to describe and compare the phylogeneticrelationships of these strains with already known groups. Materials and methods Study site and methods A total of 43 monoclonal picocyanobacterial isolates (13phycoerythrin and 30 phycocyanin) obtained using twoisolation methods (Jasser  et al  ., 2010) were used for thegenetic and phylogenetic analyses. The strains were isolatedfrom four Mazurian lakes in north-eastern Poland (54 1 Nand22 1 E) in 2006. Three of the lakes are interconnected andsituated within the Great Mazurian Lakes system, whereasthefourth,Lake Majcz Wielki,islocatedin thesame area, butconnected to the system by a river. The studied lakesexemplify various kinds of glacial lake. Lake Mikołajskie andLake Bełdany represent gully lakes, whereas Lake S´niardwy,the largest lake in Poland, and Lake Majcz Wielki representkettle-type glacial lakes. The lakes were formed by ice sheetsin the late Pleistocene during the Pomeranian phase of Vistulian glaciations (Bajkiewicz-Grabowska, 1989; Wacnik,2009). At present, two of the lakes are relatively low produ-cing (mesotrophic Lake Majcz Wielki and meso-eutrophicLake S´niardwy), whereas the other two are eutrophic (LakeMikołajskie and Lake Bełdany). All four lakes are character-ized by low background water turbidity and the predomi-nance of phycoerythrin-rich picocyanobacteria (Table 1).Both phycoerythrin and phycocyanin strains were isolatedfrom each of the lakes; however, most of the phycoerythrinisolates were obtained from the eutrophic lakes, and most of the phycocyanin isolates came from the mesotrophic lake.The picocyanobacterial isolates were analyzed microscopi-cally and molecularly to verify their taxonomic purity (Jasser et al  ., 2010). All strains were cultivated in BG11 medium in aplant growth chamber at a constant temperature of 20 1 C.‘Cool white’ (4000K, Osram Lumilux, Munich, Germany)fluorescent tubes provided light at 16 m molphotonsm  2 s  1 in a 14/10h day/night cycle. Table 1.  Basic physical, chemical and biological characteristics of the lakes from which picocyanobacteria were isolated in 2006Lakes Area (ha)Meandepth (m)Maximumdepth (m)Chlorophyll a  ( m gL  1 ) SD (m)Carlson’sindex (TSI) K PAR K BG (484) (m  1 ) PE (%)Majcz Wielki 160 6.0 16.5 5.1 4.0 43.3 0.61 0.69 95S´ niardwy 11340 5.8 23.4 14.0 2.2 52.5 0.72 0.53 88Bełdany 941 10.0 46.0 27.2 1.6 58.1 1.06 0.66 88Mikołajskie 498 11.2 25.9 28.1 1.6 58.3 1.04 0.59 87K PAR  – light attenuation coefficient estimated from vertical light profiles (PAR range, 400–700nm). K BG  (484) – background turbidity at 484nmmeasured according to the method of Stomp  et al  . (2007). FEMS Microbiol Ecol  75  (2011) 89–98 c   2010 Federation of European Microbiological SocietiesPublished by Blackwell Publishing Ltd. All rights reserved 90  I. Jasser  et al  .  Pigment analysis The  in vivo  absorbance spectra of the pigments of whole cellswere analyzed using a Hitachi U-1800 spectrophotometer(Tokyo, Japan). The spectra were recorded between 380 and750nm at 1nm intervals with 4nm slits. DNA extraction and PCR amplification DNAwas extracted from 2mL cultures of the isolated strainsusing a commercial kit (EURx, Gda´nsk, Poland) designedfor the extraction of genomic DNA from bacteria (Jasser et al  ., 2010). Amplification of the  cpcBA -IGS region of thephycocyanin operon (Robertson  et al  ., 2001) for directsequencing was performed using the DNA from 39 isolateswith the specific primers cpcBF(UFP) and cpcAR(URP). Afragment of 16S rRNA gene was amplified from the DNA of 43 isolates using the primers 16S5_F (Scheldman  et al  .,1999) and B23S5_R (Lepe`re  et al  ., 2000). PCR products werepurified using a kit (EURx) and then directly sequenced inboth directions with the srcinal PCR primers using an ABI(3130xl) capillary sequencer. Sequence accession numbers, alignments andphylogenetic analyses The 16S rRNA gene and  cpcBA- IGS nucleotide sequences of picocyanobacterial isolates from the Mazurian lakes weresubmitted to the DDBJ/EMBL/GenBank databases withthe following accession numbers: FJ63765–FJ763789,GQ130142–GQ130156 and FJ763795–FJ763832. The Gen-Bank accession numbers for all the DNA sequences reportedhere and used for phylogenetic analyses are shown inSupporting Information, Table S1. Sequence alignmentsproduced using  CLUSTALX   1.83 (Thompson  et al  ., 1997) withdefault options were manually edited. The  cpcBA -IGS align-ment was edited according to the primary structure of theprotein using the  GENETIC DATA ENVIRONMENT  (2.3) software(Smith  et al  ., 1994). Areas that could not be alignedunambiguously were excluded from further analysis. Forphylogenetic analyses, a 16S rRNA gene dataset of 1341characters in the alignment of 85 sequences and a  cpcBA -IGSdataset of 388 characters in the alignment of 68 sequenceswere generated. A partitioned dataset was used for the cpcBA -IGS alignment, with three partitions (three codonpositions).Neighbor joining (NJ), maximum likelihood (ML) andmaximum parsimony (MP) analyses were performed using PAUP  version 4.0b10 for Macintosh OS X (Swofford, 2002).To identify the best tree in MP analysis, the heuristic searchoption was used with MULTREES, tree-bisection-reconnec-tion (TBR) branch swapping, with ACCTRAN optimizationand random addition, for 10 replicates. The single ML treewas obtained by a heuristic search using a random stepwiseaddition for 10 replicates, as well as with TBR branchswapping and MULTREES on. Bootstrap support for spe-cific nodes (Felsenstein, 1985) was estimated with defaultoptions using 100 replications. Bayesian analyses (BA) wereperformed and the values of their model parameters wereestimated using  MRBAYES  3.1.2 software (Ronquist & Huel-senbeck, 2003). Two independent analyses were run withthree heated and one cold chain (temperature parameter0.1) for 3000000 generations, discarding the first 25% of trees.Models of sequence evolution for the ML, NJ and BAmethods, and their parameter values for the ML and NJmethods, were estimated using  MODELTEST  3.7 (Posada & Crandall, 1998). For 16S rRNA gene alignment, theTrN 1 I 1 G model and a general time-reversible(GTR  1 I 1 G) model were used. For  cpcBA -IGS alignment, aGTR  1 I 1 G model was used for the first codon, a SY-M 1 I 1 G model for the second and a TVM 1 G model forthe third. The sequences from strains PCC 6301, PCC 7002and PCC 7942 were used to root the trees (Ernst  et al  ., 2003;Haverkamp  et al  ., 2008). Results Pigment analyses The pigment patterns of the 43 studied strains representedtwo main phycobilin pigment types: I and II. Type I strains(30 isolates) carried only phycocyanin and appeared blue/green in culture, with a maximum absorbance peak ataround 630–635nm. All phycoerythrin strains exhibited atype II pigment absorption pattern, with maximum absor-bance at 570nm. No maxima between 490 and 495nm (thephycourobilin peak) were detected for any of the strains. The 16S rRNA gene region The molecular identification of the 43 strains by 16S rRNAgene nucleotide sequence analysis revealed several clusters of picocyanobacteria. The genetic distance between individualstrains was up to 3.9%, whereas the maximum geneticdistance between all analyzed strains was 7.1%. Furtheranalysis demonstrated that the phycoerythrin and phyco-cyanin cyanobacteria isolated from the studied lakes be-longed to four previously described clades and one novelclade (Fig. 1). The existence of this new clade, named ‘theMazurian clade’ (M), was confirmed in four differentphylogenetic analyses: BA=0.93, ML=71, NJ=62 andMP=61. Both phycoerythrin and phycocyanin picocyano-bacteria, isolated from the eutrophic Lake Mikołajskie andthe mesotrophic Lake Majcz Wielki, respectively, wereassigned to this clade. The similarity between the 16S rRNAgene nucleotide sequences within this clade varied between99.1% and 100%. The rest of the isolated phycoerythrin FEMS Microbiol Ecol  75  (2011) 89–98  c   2010 Federation of European Microbiological SocietiesPublished by Blackwell Publishing Ltd. All rights reserved 91 A picocyanobacterial clade reflecting lake ontogeny  picocyanobacteria clustered, with a high bootstrap supportand posterior probability, with Group B – subalpine cluster Iand Group E. The strains assigned to Group B came fromthree of the studied lakes (mesotrophic, meso-eutrophic andeutrophic) and shared high sequence similarity (up to99.8%) with  Synechococcus rubescens  (SAG 3.81) and isolateMW 10#1 from Lake Mondsee in Austria. One phycoery-thrin isolate, BE0807D, could be assigned to Group E. Thisstrain was isolated from the second eutrophic lake, Bełdany,and the sequence similarity to the strain PS 717 from LakeBiwa was 99.5%.The majority of the remaining phycocyanin picocyano-bacteria were found to convene with Group A, the  Cyano-bium gracile  cluster. Group A was supported by only three Fig. 1.  Phylogenetic tree of the 16S rRNA genenucleotide sequences obtained by Bayesianinference (model: GTR 1 I 1 G). Terminal branchesdisplay strain names and GenBank accessionnumbers (sequences determined in this studyare shown in bold) and pigment group (  ,phycoerythrin rich;   , phycocyanin rich).Numbers at the nodes show posteriorprobabilities ( 4 0.75, first number) of the treebipartitions, as well as the bootstrap valuesobtained by ML ( 4 50%, second number) andNJ ( 4 50%, third number) analyses (model:GTR 1 I 1 G), and MP analysis ( 4 50%, fourthnumber). Clades not present in particularanalyses or without substantial support areindicated by ‘–’. FEMS Microbiol Ecol  75  (2011) 89–98 c   2010 Federation of European Microbiological SocietiesPublished by Blackwell Publishing Ltd. All rights reserved 92  I. Jasser  et al  .  analytical methods (BA=0.83, NJ=78, MP=80), and the16S rRNA gene nucleotide sequences exhibited a similarity of between 98.4% and 100%. The strains grouped in thiscluster were obtained from all four sampled lakes. A few other phycocyanin isolates, srcinating from the eutrophicLake Bełdany, exhibited high sequence similarity (up to98.6%) to strains JJ27STR (GenBank accession number:AM710383) and JJM10D4 (GenBank accession number:AM710359) obtained from lakes and reservoirs in the CzechRepublic. This group was tentatively named Group Cz. Theaffiliation of the isolates from Lake Bełdany to this clusterwas robust, as confirmed using three methods (BA=0.96,NJ=59, MP=90). The position of one of the isolatedphycocyanin strains, MA0607K, was not resolved (Fig. 1). The phycocyanin operon Comparison of the  cpcBA  operon (IGS excluded) sequencesof 39 of the Mazurian strains and those available in GenBank showed that the isolates were clustered according to pigmentphenotype. The isolated picocyanobacteria exhibited geneticdistances between strains of up to 23.7%, whereas themaximum genetic distance between all analyzed strains was29.6%. The  cpcBA  phylogenetic analysis (Fig. 2) revealedthat in most cases, the isolates were assigned to the sameclusters as those determined by the 16S rRNA gene analysis,i.e. Groups A, B, E and the newly established Group M. Afew strains, however, fell into different clades. Specifically, 10isolates, including nine from the mesotrophic Lake Majcz Fig. 2.  Phylogenetic tree of the  cpcBA -IGSsequences obtained by Bayesian inference(models for partitions are listed in Materials andmethods).Terminalbranchesdisplaystrainnamesand GenBank accession numbers (sequencesdetermined in this study are shown in bold),pigmentgroup(  ,phycoerythrin;  ,phycocyanin)and IGS length. Numbers at the nodes showposterior probabilities ( 4 0.75, first number) ofthe tree bipartitions, as well as the bootstrapvalues obtained by ML ( 4 50%, secondnumber), NJ ( 4 50%, third number) and MP( 4 50%, fourth number) analyses. Clades notpresent in particular analyses or withoutsubstantial support are indicated by ‘–’. FEMS Microbiol Ecol  75  (2011) 89–98  c   2010 Federation of European Microbiological SocietiesPublished by Blackwell Publishing Ltd. All rights reserved 93 A picocyanobacterial clade reflecting lake ontogeny

26-ICCPM2011A00048

Jan 10, 2019

SB2P

Jan 10, 2019
Search
Similar documents
View more...
Related Search
We Need Your Support
Thank you for visiting our website and your interest in our free products and services. We are nonprofit website to share and download documents. To the running of this website, we need your help to support us.

Thanks to everyone for your continued support.

No, Thanks