您所在位置： &
&nbsp&&nbsp&nbsp&&nbsp
中国深圳海域和九龙江口细菌生态特征研究.pdf 210页
本文档一共被下载：
次 ,您可全文免费在线阅读后下载本文档。
下载提示
1.本站不保证该用户上传的文档完整性，不预览、不比对内容而直接下载产生的反悔问题本站不予受理。
2.该文档所得收入(下载+内容+预览三)归上传者、原创者。
3.登录后可充值，立即自动返金币，充值渠道很便利
需要金币：300 &&
中国深圳海域和九龙江口细菌生态特征研究
你可能关注的文档：
··········
··········
博士学位论文
中国深圳海域及九龙江口细菌生态特征研究
姓名：陈明霞
申请学位级别：博士
指导教师：郑天凌；李和阳
海洋弧菌是海洋微食物环的重要组成部分，对海洋环境营养物的循环起着非
常重要的作用。此外，弧菌还是人类及水生动物(包括养殖动物)的重要致病菌。
对弧菌及其相关类群的种类组成、数量分布、与环境相关关系以及耐药性的研究
是环境细菌学研究的重要组成部分，其研究对环境健康、疾病防治及生态安全等
方面具有重要意义。本文运用传统的分离培养方法和现代的分子分类检测方法，
对深圳海域及九龙江口弧菌(包括可培养、未培养的)的生态分布及其与环境相
关关系进行研究。
研究内容包括两大部分：
1．深圳海域弧菌种类组成、数量分布及其与环境相互关系研究
2．九龙江口TCBS类群(生长于TCBS琼脂培养基上的细菌类群)的分布及其
耐药性研究
主要研究结果包括：
1．深圳海域可培养弧菌数量分布具有季节性。东部海域弧菌数量春季(2008
年4月)(范围为5．10x102-4．40x104CFU／mL，平均为1．50x104CFU／mL)高于
秋季(2008年9月)(范围为1．4l×102．2．57x103CFU／mL，平均为8．89x102
CFU／mL)；西部海域秋季(范围为0．1．56x103CFU／mL，平均为5．09x102
略高于春季(范围为0-6．62xCFU／mL，平均为2．66xCFU／mL)；弧菌数
量高发区出现在4月份大亚湾(4．40x104CFU／mL)。弧菌分布与温度、有机物
浓度及盐度等相关，各环境冈素互相交联，其相互关系复杂。
2．珠江口盐度低于11的站位未检测到可培养的弧菌，盐度高于ll的站位，
可培养弧菌数量随盐度增加而增大。九龙江口盐度较低的上游区域沉积物中未检
测到。口J‘培养弧菌，而在盐度较高的河口下游区域有弧菌分布，而且其分布比例随
盐度增加而增加。
3．深圳海域清洁水域(YMK001站位和GDN064站位)发现有大量的弧菌
存在，且数量高于污染严重的海域(包括深圳湾站位及珠江口下游站位)，是弧
菌数量的高发区。
4．深圳海域春季可培养弧菌有M
cyclitrophicus、M
harveyi的类似种，其中主要优势类群是K
splendidus的类似种；秋季分布有矿alginolyticus、Wnatriegens、Ⅳmytf，i、
两个季节的优势类群，其平均数量春季高于秋季；春季特有的优势类群有矿
似种主要发生于4月份水体温度较低的季节。
TCBS类群存在。弧菌占TCBS菌株总数的比例因不同站位、不同季节而有很大的
变化，其比例为0—100％。盐度较高的区域，弧菌数量与TCBS菌群数呈直接正相
关，盐度较低的区域弧荫数量与TCBS菌群数无关或关系不明显。
7．九龙江IS!TCBS菌株53％对氨苄青霉素耐药，59％对头孢唑啉耐药，72％
对头孢噻吩耐药，l％对庆大霉素耐药，4％对氟哌酸耐药，3％对四环素耐药，7％
对链霉素耐药。有些菌株为3联、4联或6联耐药菌株。
关键词：深圳海域，九龙江口，弧菌种类组成及数量分布，TCBS类群，耐药性
arcdominantinthemarineenvironmentand a
roleinnutrient
正在加载中，请稍后...【图文】SpringMVC_百度文库
两大类热门资源免费畅读
续费一年阅读会员，立省24元！
大小：1.36MB
登录百度文库，专享文档复制特权，财富值每天免费拿！
你可能喜欢Network of Interactions Between Ciliates and Phytoplankton During Spring (PDF Download Available)
See all >7 CitationsSee all >58 ReferencesSee all >7 Figures
34.15University of Zurich13.08+ 235.36Eawag: Das Wasserforschungs-Institut des ETH-Bereichs22.61Italian National Research CouncilShow
more authorsAbstractThe annually recurrent spring phytoplankton blooms in freshwater lakes initiate pronounced successions of planktonic ciliate species. Although there is considerable knowledge on the taxonomic diversity of these ciliates, their species-specific interactions with other microorganisms are still not well understood. Here we present the succession patterns of 20 morphotypes of ciliates during spring in Lake Zurich, Switzerland, and we relate their abundances to phytoplankton genera, flagellates, heterotrophic bacteria, and abiotic parameters. Interspecific relationships were analyzed by contemporaneous correlations and time-lagged co-occurrence and visualized as association networks. The contemporaneous network pointed to the pivotal role of distinct ciliate species (e.g., Balanion planctonicum, Rimostrombidium humile) as primary consumers of cryptomonads, revealed a clear overclustering of mixotrophic/omnivorous species, and highlighted the role of Halteria/Pelagohalteria as important bacterivores. By contrast, time-lagged statistical approaches (like local similarity analyses, LSA) proved to be inadequate for the evaluation of high-frequency sampling data. LSA led to a conspicuous inflation of significant associations, making it difficult to establish ecologically plausible interactions between ciliates and other microorganisms. Nevertheless, if adequate statistical procedures are selected, association networks can be powerful tools to formulate testable hypotheses about the autecology of only recently described ciliate species.Discover the world's research14+ million members100+ million publications700k+ research projectsFigures
ORIGINAL RESEARCHpublished: 20 November 2015doi: 10.3389/fmicb.Frontiers in Microbiology | www.frontiersin.org 1November 2015 | Volume 6 | Article 1289Edited by:T?lesphore Sime-Ngando,Centre National de la Recherche,France, FranceReviewed by:Angelicque White,Oregon State University, USARachael Marie Morgan-Kiss,Miami University, USA*Correspondence:Thomas Poschposch@limnol.uzh.chSpecialty section:This article was submitted toAquatic Microbiology,a section of the journalFrontiers in MicrobiologyReceived: 22 June 2015Accepted: 04 November 2015Published: 20 November 2015Citation:Posch T, Eugster B, Pomati F,Pernthaler J, Pitsch G and Eckert EM(2015) Network of InteractionsBetween Ciliates and PhytoplanktonDuring Spring.Front. Microbiol. 6:1289.doi: 10.3389/fmicb.Network of Interactions BetweenCiliates and Phytoplankton DuringSpringThomas Posch 1*, Bettina Eugster 1, Francesco Pomati 2, Jakob Pernthaler 1,Gianna Pitsch 1and Ester M. Eckert 1, 31Limnological Station, Institute of Plant Biology and Microbiology, University of Zurich, Kilchberg, Switzerland, 2DepartmentAquatic Ecology, Swiss Federal Institute of Aquatic Science and Technology, D?bendorf, Switzerland, 3Microbial EcologyGroup, Consiglio Nazionale Delle Ricerche- Istituto per lo studio degli ecosistemi, Verbania Pallanza, ItalyThe annually recurrent spring phytoplankton blooms in freshwater lakes initiatepronounced successions of planktonic ciliate species. Although there is considerableknowledge on the taxonomic diversity of these ciliates, their species-speci?c interactionswith other microorganisms are still not well understood. Here we present the successionpatterns of 20 morphotypes of ciliates during spring in Lake Zurich, Switzerland, andwe relate their abundances to phytoplankton genera, ?agellates, heterotrophic bacteria,and abiotic parameters. Interspeci?c relationships were analyzed by contemporaneouscorrelations and time-lagged co-occurrence and visualized as association networks.The contemporaneous network pointed to the pivotal role of distinct ciliate species(e.g., Balanion planctonicum,Rimostrombidium humile) as primary consumers ofcryptomonads, revealed a clear overclustering of mixotrophic/omnivorous species, andhighlighted the role of Halteria/Pelagohalteria as important bacterivores. By contrast,time-lagged statistical approaches (like local similarity analyses, LSA) proved to beinadequate for the evaluation of high-frequency sampling data. LSA led to a conspicuousin?ation of signi?cant associations, making it dif?cult to establish ecologically plausibleinteractions between ciliates and other microorganisms. Nevertheless, if adequatestatistical procedures are selected, association networks can be powerful tools toformulate testable hypotheses about the autecology of only recently described ciliatespecies.Keywords: ciliate morphotypes, ciliophora, local similarity analysis, phytoplankton spring bloom, network analysisINTRODUCTIONIn the original description of the PEG-model (Plankton Ecology Group, Sommer et al., 1986)explaining the mechanisms of plankton successions in lakes, the authors state about phytoplanktonspring bloom dynamics: ?It is clearly to be seen in most of the lakes that the ?rst herbivores to buildup abundant populations in the spring are small protozoans and rotifers, which have short generationtimes and exponential increase within a few days.? Although the role of microzooplankton groupsas the ?rst relevant grazers of algal spring blooms was highlighted by the PEG-model authors, thistrophic link has been almost forgotten -or overlooked- for decades. Conversely, several studieshave focused on a direct trophic shortcut from phytoplankton to metazooplankton (e.g., daphnids,copepods), attributing to the latter the sole control of algal development during spring. Already in
Posch et al. Ciliates Interactions During Springthe early 1990s, Helga M?ller and co-workers published apioneering work on the importance of ciliates (Ciliophora) as the?rst and most e?ective grazers of phytoplankton spring bloomsin Lake Constance (M?ller, 1989; M?ller et al., 1991a; M?llerand Weisse, 1994). These observations have also been con?rmedfor other temperate lakes (Amblard et al., 1993; Sommarugaand Psenner, 1993; Mathes and Arndt, 1995; Carrias et al.,1998), and the role of protists as consumers was highlightedin a recent description of the PEG-model (Sommer et al.,2012). Furthermore, it was recognized that the spring peak ofalgivorous ciliates is followed by a conspicuous succession ofvarious mixotrophic (Amblard et al., 1993), omnivorous andpredatory ciliate species (M?ller et al., 1991b). Due to their fastgeneration times (hours to days), the succession of ciliate speciesis characterized by several short-lived peaks of a few dominantgenera and a high sampling frequency is thus required to followtheir dynamics in ?real-time? (?imek et al., 2014).Nevertheless, there is an obvious discrepancy between theconsiderable knowledge on the diversity of freshwater ciliatemorphotypes (summarized in Foissner et al., 1999), theirsuccession during spring (Weisse and M?ller, 1998; Sonntaget al., 2006; Zingel and N?ges, 2010), and the scarce informationon species speci?c interactions between ciliates and othermicroorganisms (protists and bacteria). In order to determinethese factors, there is a need for broader studies on microbial foodwebs that examine multiple abiotic parameters in parallel withmicro-organisms at high taxonomic resolution. Simultaneousinformation of the diversity of organisms and a detailed patternof their co-occurrences can be obtained via next generationsequencing of phylogenetic marker genes and software-basednetwork analysis (Steele et al., 2011; Chow et al., 2014). However,in the case of freshwater ciliate species, this approach hasseveral limitations. (i) Sequence information is still missingfor many well-known and precisely described freshwater ciliatemorphotypes (Stoeck et al., 2014). It is therefore di?cult to relateoperational taxonomic units (OTU) with the existing knowledgeabout the autecology of morphospecies (see literature reviews inFoissner et al., 1999; Lynn, 2008). (ii) The co-occurrence patternof a ciliate OTU with, e.g., algal OTUs, does not inform about thetype of interaction between them at all, when no autecologicalbackground information is consulted. (iii) Due to the high copynumber of 18S rRNA (ribosomal ribonucleic acid) genes in singleciliate cells (Gong et al., 2013), a reliable quanti?cation of ciliateabundance is not yet possible solely on molecular techniques.As a consequence, the identi?cation and quanti?cation ofciliates based on their morphology might currently be a moredirect means to investigate interspeci?c interactions. Here wepresent data from a high frequency sampling campaign (2?4 daysampling intervals during 7 weeks) aimed at characterizing thedynamics of ciliate morphotypes during a phytoplankton springbloom in a large freshwater lake (Lake Zurich, Switzerland).A contemporaneous statistical analysis was conducted usingthe abundances data of protistan morphotypes. We searchedfor interspeci?c associations, keeping in mind autecologicalbackground information for the detected morphospecies. Wealso tested the explanatory power of an ecological network basedon local similarity analysis (LSA), a method often used for theevaluation of environmental sequences data allowing for thedetection of time-shifted co-occurrences between parameters.METHODSStudy Area and Sampling SiteLake Zurich is an oligo-mesotrophic pre-alpine lake with amaximum depth of 136 m and a surface area of 66.8 km2. Theentire water volume of 3.34 km3is theoretically renewed in 1.2years. The lake is located in a densely populated area and servesas a source of drinking water for more than 1 million people.Lake Zurich is monomictic with infrequent holomixis (Poschet al., 2012). Since the beginning of the 20th century the trophicstatus has increased and eutrophication reached its maximumin the 1960?s. Due to consequent waste water treatment totalphosphorus-concentrations have decreased from &120 ?g L?1topresently about 15 ?g L?1.Sampling StrategySamples were taken in 2?4 day intervals at one sampling site(47?19.3?N 8?33.9?E, zm=100 m) from 23 March to 12 May2009 at around 10 a.m. In spring, the weakly strati?ed waterbody of Lake Zurich is very susceptible to changes in weatherconditions and especially to storm events (Bleiker and Schanz,1989). Due to the geographic position and topographic situationof the lake, strong winds cause internal waves (seiches) withamplitudes between 2 and 6 m (Horn et al., 1986). Seiches caninduce massive displacements of strati?ed populations (Garneauet al., 2013), thus changing the depth of the spring phytoplanktonmaximum already on a daily base. Consequently, before samplingwe used a ?uoroprobe (TS-16-12, bbe Moldaenke GmbH) todetermine chlorophyll a(Chl a)in situ pro?les between 0and 30 m. This probe originally distinguishes between fourphytoplankton classes (cryptophytes, diatoms, chlorophytes,phycocyanin containing ?blue? cyanobacteria) and gives theirrelative Chl acontribution on total Chl aconcentration (Beutleret al., 2002). We calibrated the probe (optical ?ngerprint) for thequanti?cation of an additional class, namely the phycoerythrincontaining ?red? cyanobacterium Planktothrix rubescens, whichis a dominant primary producer in Lake Zurich (Poschet al., 2012). Based on in situ Chl apro?les (Figures 1B?D)we determined the current depth with the maximal Chl aconcentration (see crosses in Figure 1B) and samples (5 L) weretaken with a Ruttner water sampler (Uwitec) from this depthlayer. These samples were used for the enumeration of bacteriaand heterotrophic nano?agellates (see below), the quanti?cationof algal and ciliate morphotypes (see below), and for chemicalanalyses. All samples were transported in an insulated box tothe laboratory within 30 min. The following chemical parameterswere measured: Dissolved phosphorus (DP) with the molybdatemethod after digestion with H2SO4and H2O2, nitrate (NO3)via spectrophotometrical determination after reduction withsodium salicilate Seignette salt, Chl avia spectrophotometricmeasurement after acetone extraction, and dissolved (DOC)and total (TOC) organic carbon via high-temperature catalyticoxidation with a Shimadzu TOC analyzer. We checked thereliability of in situ Chl avalues obtained with the ?uoroprobeFrontiers in Microbiology | www.frontiersin.org 2November 2015 | Volume 6 | Article 1289
Posch et al. Ciliates Interactions During SpringFIGURE 1 | (A) Water temperature, (B) cryptophytes-related Chlorophyll a(Chl ain ?g L?1), (C) diatom-related Chl a(?g L?1), (D) Planktothrix rubescens related Chla(?g L?1) of the surface water body (0?30 m) in Lake Zurich during a phytoplankton spring bloom (March?May 2009). Crosses in (B) indicate the sampling depth,and dashes in (B,F) indicate the four phases of the investigation period (see text). (E) Maximal wind speed (ms?1) at the surface of the lake and photosynthetic activeradiation (?mol quanta m?2s?1) measured in the sampling depth. (F) Total Chl a(?g L?1) and ciliate abundance (103L?1). On three selected sampling dates wemade triplicate QPS preparations to evaluate standard deviations of ciliate counts. (G) Concentration of dissolved phosphorus (DP, ?g L?1) and dissolved organiccarbon (DOC,mg L?1, average ?standard deviation). (H) Bacterial (109L?1) and heterotrophic nano?agellate (HNF, 106L?1) abundance in the sampling depthsduring the investigation period.by comparing values with Chl adata determined by extraction(linear regression, r2=0.82).In addition, pro?les of water temperature and oxygenwere recorded with a 6600 multi-parameter probe (YellowSprings Instruments) between 0 and 30 m depth. Pro?les ofphotosynthetically active radiation were determined with aspherical quantum sensor (LI-COR) from the surface in 1-mintervals until an irradiance of &0.05 ?mol quanta m?2s?1wasreached.Abundance of Heterotrophic Bacteria andFlagellatesSamples for bacterial abundances were ?xed with formaldehyde(2% ?nal concentration, f.c.), 1 mL of ?xed samples wasstained with 10 ?L of 4?,6-diamidino-2-phenylindole (DAPI),and total numbers were measured by ?ow-cytometry (inFluxV-GS, Becton Dickinson). Excitation was set at 355 nm andDAPI emission was measured at 460 ?50 nm. Further detailson the analysis of bacterial parameters (total and ?lamentabundances) are described in Eckert et al. (2012). For thecounting of heterotrophic nano?agellates (HNF), 40 mL of rawwater were ?xed with Lugol?s solution (0.5% f.c.), followedby formaldehyde (2% f.c.), bleached with a few drops ofsodiumthiosulfate (3% stock solution), and stored in the darkat 4?C until processing. Fixed samples (5?10 mL) were stainedwith DAPI, ?ltered on polycarbonate ?lter (1-?m pore size) andmicroscopically counted (n=50?100 ?agellates per sample) at1000 x magni?cation (Zeiss Axio Imager.M1).Abundance of Phytoplankton and of CiliateGenera and MorphospeciesFor the determination of algae, 100 mL of water sampleswere ?xed with Lugol?s solution (1% f.c.) and analyzed usinginverted microscopy (Uterm?hl, 1958). Details on phytoplanktonspecies determination and quanti?cation are given in Pomatiet al. (2013). Picocyanobacteria, i.e., Synechococcus-like coccoidcyanobacteria, were not quanti?ed in this study, as theirabundances are very low during spring but start to increase inLake Zurich at the beginning of July.For the quanti?cation of ciliate morphotypes we used theQuantitative Protargol Staining (QPS) following the protocolof Skibbe (1994) with few modi?cations according to P?steret al. (1999). QPS results in permanent slides and allows forthe taxonomic assignments of counted ciliates. Three hundredmL of raw water samples were ?xed with Bouin?s solution (5%f. c., Skibbe, 1994). Samples were stored at room temperatureuntil further processing. Protargol stained ?lters (0.8-?m poresized cellulose nitrate with counting grid, Sartorius) wereanalyzed microscopically at x magni?cation. Theinspected water volume per sample was at least 9.5 mL, i.e.,Frontiers in Microbiology | www.frontiersin.org 3November 2015 | Volume 6 | Article 1289
Posch et al. Ciliates Interactions During SpringTABLE 1 | Ciliate species/genera detected during the phytoplankton spring bloom period of Lake Zurich in 2009.Frequency: % Average (Maximum) Abundance: ciliates L?1Average (Maximum)PHYLLOPHARYNGEASuctoriaEndogenidaStaurophrya elegans 0.1 (0.9) 8 (50)PROSTOMATEAProrodontidaBalanion planctonicum 16.0 (50.6) )Urotricha spp. 27.5 (47.0) )SPIROTRICHEATintinnidaCodonella cratera 0.3 (2.8) 51 (334)Tintinnidium/Tintinnopsis 2.5 (11.3) 391 (1670)Tintinnidium pusillumTintinnopsis cylindrataMembranicola tamari 0.8 (3.8) 101 (334)ChoreotrichidaRimostrombidium humile 1.0 (3.2) 245 (1280)Rimostrombidium spp. 9.3 (25.7) )Rimostrombidium hyalinum/brachykinetumRimostrombidium lacustrisStrombidiidaLimnostrombidium spp. 2.6 (5.5) 827 (2560)Limnostrombidium pelagicumLimnostrombidium viridePelagostrombidium spp. 1.4 (4.2) 499 (2152)Pelagostrombidium fallaxPelagostrombidium mirabileSporadotrichidaHalteria/Pelagohalteria 1.3 (3.6) 431 (891)Pelagohalteria cirriferaPelagohalteria viridisHalteria sp.Undetermined Spirotrich 5.7 (23.3) )LITOSTOMATEACyclotrichiidaAskenasia chlorelligera 0.6 (2.6) 261 (1336)Askenasia spp. 0.4 (1.1) 178 (668)Askenasia acrostomiaRhabdoaskenasia minima 0.2 (1.0) 72 (500)Mesodinium sp. 3.7 (9.5) )OLIGOHYMENOPHOREAPeritrichiaSessilida 2.6 (7.3) 970 (3896)Vaginicola sp.Vorticella natansVorticella vernalisPeniculiaPeniculidaStokesia vernalis 0.2 (1.0) 68 (297)ScuticociliataPleuronematidaHistiobalantium bodamicum 19.3 (57.5) )Cyclidium spp. 4.4 (25.5) 304 (1186)(Continued)Frontiers in Microbiology | www.frontiersin.org 4November 2015 | Volume 6 | Article 1289
Posch et al. Ciliates Interactions During SpringTABLE 1 | ContinuedRare species Actinobolina smalli (Litostomatea)Coleps spetai (Prostomatea) Belonophrya pelagica (Litostomatea)Epistylis anastatica (Oligohymenophorea) Monodinium armatum (Litostomatea)Epistylis pygmaeum (Oligohymenophorea) Lagynophrya acuminata (Litostomatea)Epicarchesium pectinatum (Oligohymenophorea) Pelagodileptus trachelioides (Litostomatea)Cinetochilum margaritaceum (Oligohymenophorea) Pelagovasicola cinctum (Litostomatea)The 20 clearly de?nable ciliate morphotype units were identical to described species, or comprised two or more species within a genus. Eleven clearly recognizable but rare specieswere excluded from graphical presentation and statistical analyses, as these species were found in too low numbers or only on single sampling dates.at minimum 400 ciliates per sample were counted. For ciliatespecies determination we used the taxonomic key published byFoissner et al. (1999), and we used the higher level taxonomicclassi?cation of Lynn (2008). On three selected sampling dates wemade triplicate QPS preparations to evaluate standard deviationsof ciliate counts. On each sampling occasion, we took also nethauls (mesh size 30 ?m) from 10 to 0 m depth for qualitativemicroscopic observations of living ciliate specimens. Theseobservations gave important background information for thelater species determination of ?xed specimens on the QPSslides.Statistical AnalysesFor contemporaneous analysis, all collected data were subjectedto a Pearson correlation coe?cient analysis, performed withthe Excel (Microsoft) add-in program XLSTAT-ADA. Firstparameters were tested for normal distribution and log(x+1)transformed when needed. Pearson correlation coe?cients (r-values), their signs (positive/negative) and levels of signi?cance(p-values) were extracted and exported to the software Cytoscape3.1.1 for creation of the graphical networks. Additionally, asecond graphical network was created using LSA (Ruan et al.,2006) to discover time-shifted associations. We used the eLSAphyton package (http://meta.usc.edu/softs/ Xia et al., 2013)which performs not only a LSA, but also contemporaneous andtime-shifted Pearson and Spearman correlation analyses (Xiaet al., 2013). A maximal time lag of two steps was set for LSA.As samples were taken in 2?4 day intervals, time lags of twosteps range from minimal 4 to maximal 8 days. Finally wecompared all statistical approaches concerning the total numberand the proportion of shared signi?cant correlations. Furtherdetails on the theoretical background and the applicabilityof networks for community analyses are given by severalauthors (Ruan et al., 2006; Steele et al., 2011; Fuhrman et al.,2015).RESULTSThermal Strati?cation and Spring BloomDynamicsA ?rst weak strati?cation started at the beginning of Apriland lasted for around 10 days, before being disrupted bya strong wind event (Figures 1A,E). The erosion of thermalstrati?cation was induced by an upwelling internal wave (seiche)of colder hypolimnetic water that led to a sudden cooling ofthe upper water layer by &3?C within 2 days. During the?rst strati?cation, cryptophytes, and diatoms dominated Chl aconcentrations (Figures 1B,C), and it was the only period whencryptophytes appeared in high numbers. The upwelling seichealso caused a disruption of the ?rst diatom bloom for a few days.Subsequently a second bloom formed in a depth of 6?7 m foraround 2 weeks (Figure 1C). From 20 April on, periodic seichesbelow the surface could be recorded with an amplitude of circa4 m, as re?ected in water temperatures (Figure 1A) but also inspatial concentrations of diatoms (Figure 1C) and P. rubescens(Figure 1D).Succession Phases in Lake ZurichOur sampling campaign encompassed four succession phasesof plankton dynamics. In the pre-bloom phase (23 Marchtill 3 April) low irradiance and Chl avalues, and highdissolved phosphorus (DP) concentrations were measured(Figures 1E?G). During this period, P. rubescens accountedfor 80% of total phytoplankton and ciliates reached a meanabundance of only seven cells mL?1(Figure 1F). As soon astemperature and irradiance increased, we observed a ?rst peakin DOC (Figure 1G) and Chl aconcentration (mainly due tocryptophytes and diatoms). In parallel, an increase of ciliates(50 cells mL?1), heterotrophic bacteria and nano?agellates(Figure 1H) was recorded. This classical spring bloom situationlasted for 10 days only, resulting in a strong reduction in DP.After this period, a second diatom bloom was also accompaniedby high ciliate abundances (60 cells mL?1) as well as peaksof bacteria and HNF. Due to increased thermal strati?cationfrom 30 April on, P. rubescens established a dense metalimneticpopulation in 10?12 m depth (Figure 1D) and became thedominant primary producer. At this stage, ciliate abundancedropped to about 20 cells mL?1. In previous investigations,P. rubescens continued to grow in this distinct metalimneticlayer until autumnal mixis caused the erosion of the epi- andmetalimnion (Posch et al., 2012).Succession of Phytoplankton and CiliatesThe microscopic analysis of phytoplankton showed a clearsuccession of larger taxonomic groups (Figures 2A,B).For clarity, we present the quantitative data on the 19determined algal genera/species (Figure 3) that contributedto the composition of larger taxonomic units. The pre-bloomphytoplankton community was dominated by small singlecelled diatoms but colonial forms were also present, withFrontiers in Microbiology | www.frontiersin.org 5November 2015 | Volume 6 | Article 1289
Posch et al. Ciliates Interactions During SpringFIGURE 2 | (A) Succession of taxonomic phytoplankton groups and (C) of ciliate classes based on cell counts. (B) Contribution of phytoplankton groups and (D) ofciliate groups to total abundance of algae and ciliates, respectively. Note that color code in (A) is also valid for (B), and that in (C) is also valid for (D). Gray line in thediatoms panel (A) show the contribution of colonial diatoms. The class Oligohymenophorea (in C,D) is split in two groups: without Peritrichia and Peritrichia only.a few dino?agellates and the cyanobacterium P. rubescens.The classical eukaryotic spring bloom was mainly formedby cryptophytes (Rhodomonas spp. and Cryptomonas spp.)and single celled diatoms (Cyclotella spp.). Afterwards largersized colonial diatoms (e.g., Tabellaria fenestrata,Fragilariacrotonensis;Figure 3) followed, accompanied by dino?agellates(Gymnodinium spp.) and chrysophytes (Dinobryon spp.). P.rubescens dominated the fourth successional phase. Throughoutthe entire succession of population waves, we observedcontinuous high abundances of single celled diatoms (Cyclotellasp.) and of undetermined chrysophytes.Ciliate classes also showed a clear temporal succession(Figures 2C,D). Details on the dynamics of prominentspecies or genera are given in Figure 4 and Table 1. Thepre-bloom phase was the only time when Suctoria (classPhyllopharyngea, Staurophrya elegans) and small scuticociliates(class Oligohymenophorea, Cyclidium spp.) were found inhigher numbers. Lorica bearing ciliates of the class Spirotrichea(Codonella cratera, Membranicola tamari,Tintinnidium sp.,Tintinnopsis sp.) were also detected. As soon as cryptophytesand diatoms increased (bloom phase), Balanion planctonicum(Prostomatea) and Rimostrombidium humile (Spirotrichea,Choreotrichida) showed steep increases in numbers (Figure 4).Colonial diatoms were often colonized by peritrichous ciliates(Oligohymenophorea), and algivorous as well as mixotrophicLitostomatea (e.g., Askenasia spp.) increased. During the postbloom phase ciliates of the class Prostomatea were still abundant,however, B. planctonicum was replaced by various Urotrichaspecies (Figure 4). During this period a single scuticociliateHistiobalantium bodamicum (class Oligohymenophorea) formedup to 35 cells mL?1(i.e., 55% of total abundance), accompaniedby a highly abundant genus of the class Litostomatea, namelyMesodinium sp.. Based on abundance data, groups contributedto the total ciliate assemblage in the following order (Figure 2D):Prostomatea (43.3%), Oligohymenophorea (26.4%), Spirotrichea(25.3%), Litostomatea (4.9%), and Phyllopharyngea (0.1%).Diversity of Pelagic CiliatesIn total we could quantify the succession of 20 clearly de?nableciliate morphotype units (Figure 4). In many cases thesemorphotypes were identical to described species (Table 1), whileothers comprised two or more species from one genus (e.g.,Limnostrombidium, Urotricha) or from an even larger taxonomicgroup (Sessilida) which could not be further identi?ed to specieslevel. Speci?cally, the quanti?cation of peritrichous ciliates, oftenepibionts on colonial diatoms or crustaceans, is not possibleFrontiers in Microbiology | www.frontiersin.org 6November 2015 | Volume 6 | Article 1289
Posch et al. Ciliates Interactions During SpringFIGURE 3 | Abundance (104L?1) of phytoplankton genera/speciesduring spring bloom 2009 in Lake Zurich. Note the different scaling ofy-axes. Genera/species are listed according to their af?liation with taxonomicgroups (see color codes in Figure 2A).solely by QPS due to restrictions in the ?lterable water volume(i.e., to concentrate su?cient colonial diatoms and crustaceans).By taking additional net-hauls we could nevertheless determinesome peritrichous taxa to the species level (Table 1). Finally,11 clearly recognizable but rare species (Table 1) were excludedfrom the statistical analyses, as these species were found in toolow numbers (&0.2 cells mL?1) or only on single samplingdates. These rare species nevertheless formed 34% of total ciliaterichness (31 taxa) observed within a rather short investigationperiod of 7 weeks.Species-speci?c Associations During theSpring BloomNetworks (Figure 5) show parameters (termed as nodes)and correlations/associations between nodes as lines (termedas edges, di?erent line-styles show positive and negativeconnections, respectively). For LSA, di?erent colors of edgesindicate time-shifted associations in our networks. Ciliatespecies/genera, heterotrophic nano?agellates (HNF) and largerheterotrophic ?agellates were set as central nodes, i.e., allsigni?cant correlations/associations (p?0.003) between theseparameters are depicted. Connections of central nodes withthe remaining nodes (phytoplankton species, other biotic, andabiotic parameters) are also shown, but for clarity not theconnections between the remaining nodes.The contemporaneous Pearson correlation analysis (PCC,Figure 5A) resulted in 116 signi?cant (p?0.003) pairs, i.e.,89 positive and only 27 negative correlations out of 1225possible combinations (Table 2). Through the selection ofcentral nodes, 66 positive and 14 negative correlations areshown. They are depicted in one network comprised of onelarger cluster (10 connected central nodes) and several smallerclusters (2?3 connected central nodes). Rimostrombidiumand Rhabdoaskenasia minima formed an independent cluster.Abundances of Gymnodinium helveticum and four algalmorphotypes did not correlate with any other parameter. Thealgae Rhodomonas lens and Cryptomonas spp. were linkedby only positive correlations with the algivorous ciliates B.planctonicum and R. humile (see also Figure 6), togetherwith the dino?agellate Gymnodinium lantzschii. A group ofmixotrophic/omnivorous ciliate species in our network wasformed by Askenasia chlorelligera and Stokesia vernalis, twospecies containing endosymbiotic green algae (zoochlorellae,Stoecker et al., 2009). In addition, two other kleptoplastidicciliate species were in this group, i.e., Limnostrombidium virideand Pelagostrombidium mirabile, which only temporarily retainthe chloroplasts of ingested algae (Rogerson et al., 1989). Themorphotype unit Halteria/Pelagohalteria (Table 1) showed theonly signi?cant positive correlation with the abundances ofheterotrophic bacteria (Figure 5A). The ?lter feeders Cyclidiumspp. (C. glaucoma and unidenti?ed species) were negativelycorrelated with heterotrophic bacteria. A second cluster ofciliates related to the dynamics of ?lamentous bacteria wasformed by Mesodinium sp. and H. bodamicum (Figure 5A).Three ciliate genera (Limnostrombidium spp., Askenasia spp. andUrotricha spp.) were all positively related to each other, but alsoFrontiers in Microbiology | www.frontiersin.org 7November 2015 | Volume 6 | Article 1289
Posch et al. Ciliates Interactions During SpringFIGURE 4 | Abundance (103L?1) of ciliate genera/species during spring bloom 2009 in Lake Zurich. Note the different scaling of y-axes. Ciliategenera/species are listed according to their af?liation with taxonomic groups (see also Table 1 and color codes in Figure 2C). Drawings of ciliates do not show theright size proportions of specimens to each other. (a) Staurophrya elegans, (b) Balanion planctonicum, (c) Urotricha spp., (d) Codonella cratera, (e) Membranicolatamari, (f) Tintinnids, (g) Rimostrombidium humile, (h) Rimostrombidium spp., (i), Limnostrombidium spp., (j) Pelagostrombidium spp., (k) Pelagohalteria/Halteria, (l)Askenasia chlorelligera, (m) Askenasia spp., (n) Rhabdoaskenasia minima, (o) Mesodinium spp., (p) Cyclidium spp., (q) Stokesia vernalis, (r) Histiobalantiumbodamicum, (s) Peritrichia. Drawings are original artworks by G. Pitsch.Frontiers in Microbiology | www.frontiersin.org 8November 2015 | Volume 6 | Article 1289
Posch et al. Ciliates Interactions During SpringFIGURE 5 | Microbial community network diagram based on (A) Pearson correlation analysis and (B) time-shifted local similarity analysis. Ciliatespecies/genera, heterotrophic nano?agellates (HNF), and large heterotrophic ?agellates were set as central nodes. Connections (edges) show signi?cant connections(p?0.003). Different colors in (B) show contemporaneous and time-shifted associations. Arrows point to the parameter that was delayed. Central nodes without anysigni?cant association to any parameter are shown at the bottom right. Abbreviations: Chla, DOC, diss NO3, PO4, TOC, total organic carbon.Frontiers in Microbiology | www.frontiersin.org 9November 2015 | Volume 6 | Article 1289
Posch et al. Ciliates Interactions During SpringTABLE 2 | Comparison of signi?cant (p?0.003) correlations/associations out of 1225 possible pairs detected by different contemporaneous andtime-shifted (1, 2 steps) analyses.Pearson correlation Spearman correlation Local similarity analysis Pearson correlationtime-shiftedSpearman correlationtime-shiftedContemporaneous 116 (89/27) 148 (105/43) 90 (66/24) 90 (73/17) 109 (79/30)1 step ? ? 51 (39/12) 100 (91/9) 137 (104/33)2 steps ? ? 24 (17/7) 58 (56/2) 99 (73/26)Total 116 (89/27) 148 (105/43) 165 (122/43) 248 (220 /28) 345 (256/89)Number in brackets show positive/negative correlations.FIGURE 6 | Dynamics of the cryptophyte Rhodomonas sp. and theciliates Balanion planctonicum and Rimostrombidium humile duringspring bloom 2009 in Lake Zurich.to colonial diatoms (F. crotonensis,Hannaea arcus), a single-celled diatom (Ulnaria ulna), and the colonial mixotrophicchrysophyte Dinobryon spp. (Figure 5A).The time-shifted LSA (Figure 5B) gave 165 signi?cant(p?0.003) pairs, i.e., 122 positive and 43 negativeconnections (Table 2). The majority of signi?cant associationswas contemporaneous (90 pairs), followed by 51 cases with atime shift of one step and only 24 cases with a delay of twosteps (Table 2). For the depiction of the network we selected onlycorrelations between central nodes and those of central nodeswith other factors (algae and abiotic parameters). This reducedthe number of presented correlations to 56 (contemporaneous),37 (one step delay), and 13 (two steps delay), respectively.The network based on LSA shows one large cluster with 15connected central nodes, and one smaller group of 5 connectedciliates. The two ciliates R. humile and C. cratera, two algalspecies and heterotrophic bacteria were not associated with anyother parameter. Some connections within ciliates and also ofciliates with algae were found by both, PCC and LSA, e.g.,the association of Halteria/Pelagohalteria with Chrysophyceaeand Cyclotella. A striking di?erence was detected for ciliatesconnected with Rhodomonas. In contrast to the results fromPCC, this algal genus showed a one-step time-shifted associationwith Halteria/Pelagohalteria and an undetermined spirotrichousciliate but no connections to B. planctonicum and R. humile.However, by comparing all signi?cant associations (p=0.003) detected by both, PCC and LSA, we found 66 sharedpairs (Table 3). Time-shifted Pearson and Spearman correlationanalyses resulted in much higher numbers of total and sharedsigni?cant correlations (Tables 2, 3) but were not further usedfor the creations of networks. Figure 7 shows the e?ects of LSAand the two other time shifted analyses on the distribution ofcorrelations factors. All time shifted analyses caused a distinctiveshift of correlation factors toward &0.2 or &?0.2 (Figure 7).DISCUSSIONCryptophytes and their PredatorsSeveral ?eld surveys have shown that B. planctonicum wasthe ?rst and most e?ective grazer of cryptophytes in spring(M?ller, 1991; Sommaruga and Psenner, 1993; ?imek et al.,2014). A 12 years data analysis showed that this predator-prey relationship was a predictable phenomenon in LakeConstance throughout the whole investigation period (Tirokand Gaedke, 2007). Numerical and functional response curvesof B. planctonicum isolates demonstrated that it is a typicalr-strategist. In addition, this species reached maximal growthrates at lower temperatures than competing ciliates (M?ller andSchlegel, 1999). Furthermore, there is a niche separation to othersmall sized prostomatid ciliates, namely Urotricha spp. (Weisseet al., 2001), which can feed on similar sized but other typesof prey. This observation was also re?ected in our networks(Figure 5) by the lack of direct links between B. planctonicum andUrotricha spp.Only by PCC, we found strong positive correlations of the?lter feeding ciliate R. humile with cryptophytes (Figures 5A,6). The congener R. lacustris is known as e?cient consumer ofcryptophytes in spring (M?ller and Schlegel, 1999). However, thishas not been reported for R. humile so far, although this specieswas found in higher population densities than R. lacustris inmany lakes (P?ster et al., 2002; Sonntag et al., 2006). Our results,moreover, are in accordance with laboratory studies (M?llerand Schlegel, 1999) showing that the diatom Stephanodiscuswas not a suitable food source for either B. planctonicum orfor Rimostrombidium spp., as no signi?cant relations betweenthese organisms could be detected. The colorless dino?agellateG. lantzschii was the third protistan species apparently pro?tingfrom cryptophytes as food source. This co-occurrence (Figure 5)is described as predator prey relationship (Weisse and M?ller,1998). We regularly detected Gymnodinium cells with ingestedcryptophytes in our Protargol preparations (data not shown).In fact, these dino?agellates seem to be voracious omnivores,Frontiers in Microbiology | www.frontiersin.org 10 November 2015 | Volume 6 | Article 1289
Posch et al. Ciliates Interactions During SpringTABLE 3 | Comparison of shared signi?cant (p?0.003) correlations between different contemporaneous and time-shifted analyses.Pearson correlation Spearman correlation Local similarity analysis Pearson correlationtime-shiftedSpearman correlationtime-shiftedPearson correlation 116 ? ? ? ?Spearman correlation 73 148 ? ? ?Local similarity analysis 66 103 165 ? ?Pearson correlation time-shifted 113 86 110 248Spearman correlation time-shifted 83 148 158 161 345Bold values are total signi?cant correlations per analysis.FIGURE 7 | Distribution of correlation coef?cients (LS) calculated for1225 possible pairs by different contemporaneous and time-shiftedanalyses.ingesting even prey of their own cell size. Besides cryptophytes,we also found ingested ciliates and centric diatoms in ?xed andliving Gymnodinium cells.Association of Mixotrophic/OmnivorousCiliatesThe co-occurrence of four ciliate genera/species (A. chlorelligera,S. vernalis,Limnostrombidium spp, Pelagostrombidium spp.,Figures 4,5A) re?ects observations from oligo- and mesotrophiclakes that mixotrophic/omnivorous species followed the ?rstpeak of ciliates in spring (Amblard et al., 1993; Sonntaget al., 2006; Stoecker et al., 2009). However, this speciesassociation was only obvious from the network based onPCC but not from LSA. The triggers for the rise of ciliateswith zoochlorellae/kleptoplastids are not known. Mixotrophicciliates are competitors of strictly heterotrophic species whenalgal prey is rare (see references in Stoecker et al., 2009).Additionally, successful feeding is also linked to prey accessibility(size and shape of algae). In our study we observed a modestincrease of colonial diatoms in parallel with the appearanceof mixotrophic/omnivorous ciliates (see Figures 3,4), possiblylimiting the spectrum of available food for ciliates. An alternativehypothesis for the appearance of mixotrophs was presented bySonntag et al. (2011): some Chlorella bearing ciliates were moreresistant to solar ultraviolet radiation than heterotrophic ones,thus allowing for a niche partitioning between these two lifestyles.Bacterivorous CiliatesHalteria is known as a quantitatively important bacterivore,sometimes dominating total protistan grazing rates, and thuseven exceeding the grazing impact of HNF (?imek et al., 2000).Nevertheless, only PCC but not LSA revealed a correlationbetween Halteria and bacteria (Figures 5A,B). The strongpositive associations of this taxonomic group with the centricdiatom Cyclotella probably mirrors a predator-prey relation(Skogstad et al., 1987). Abundant small sized (4?6 ?m) centricCyclotella (cf. melosiroides) in Lake Zurich were within thepreferred prey size range of Halteria (J?rgens and ?imek, 2000),thus serving as a potential second food source.We have to interpret the negative correlation (only detectedwith PCC) of Cyclidium spp. with bacterial abundance withcare, and we suppose that it re?ects non-overlapping seasonalpeaks and not necessarily a direct causal link. Cyclidium spp.typically reach highest abundances during the cold seasons inthe upper water strata in meso- to oligotrophic lakes, and theydominate the cold hypolimnion during the rest of the year(Sonntag et al., 2006). Both habitats are characterized by lowbacterial abundances and production, which contradicts the highhalf saturation constants for bacterial prey (Posch et al., 2001).Probably, the decline of Cyclidium spp. is linked to competitionwith other bacterivores or shifts in the bacterial assemblage(Eckert et al., 2012; see also Figure 2 in Salcher, 2014).Both, PCC and LSA, might indicate that Mesodinium sp.and H. bodamicum fed on ?lamentous bacteria (Figures 5A,B).Their size range of ingestible prey is large, also feeding onalgae, ?agellates and small ciliates (M?ller and Weisse, 1994;Foissner et al., 1999). The preference for large prey particlesmight explain why we found correlations of the two speciesonly with ?lamentous bacteria, but not with total heterotrophicbacteria which are dominated by tiny coccoid and rod-shapedmorphotypes (Salcher, 2014).Unresolved Co-occurrence Patterns ofCiliate and Algal SpeciesWe found numerous contemporaneous and time-shiftedassociations of Urotricha spp., Askenasia spp. andLimnostrombidium spp. with colonial diatoms and chrysophytesFrontiers in Microbiology | www.frontiersin.org 11 November 2015 | Volume 6 | Article 1289
Posch et al. Ciliates Interactions During Spring(Figure 5). These colonial algae are too large for being ingestedby Urotricha spp. Solitary Fragilaria and Dinobryon cells wereoccasionally present but in too low numbers to sustain theseciliate populations. In addition, the single celled diatom Ulnariaulna is not an appropriate food source for these ciliates dueto its large cell size (up to 350 ?m). The morphotype unitUrotricha spp. included various species, and QPS preparationsdid not allow for detailed taxonomic determinations. Thevarious associations to other organisms probably re?ects thattoo many di?erent species were pooled in this morphotypeunit. For a proper identi?cation of the so far 13 describedeuplanktonic Urotricha species, the use of silver carbonate wasrecommended (Foissner and P?ster, 1997) in combinationwith live observations. We hope that in the future 18S rDNAsequencing might allow for a detailed species determination, asUrotricha species greatly di?er in their preferred food sourcesbut also prey size ranges.Interpretation of Contemporaneous andTime Shifted CorrelationsLSA was originally developed to detect spatial or time shiftedassociations which are not discovered by contemporaneousanalyses alone (Ruan et al., 2006). Beside the analysis ofenvironmental sequencing data, LSA can be also applied onclassical counting data as stated by the authors (Ruan et al., 2006).Since then, it is discussed how to interpret these associationpatterns, and how to detect possible causalities (Faust and Raes,2012; Fuhrman et al., 2015). It was supposed that positiveassociations indicate mutualism, commensalism, and cross-feeding activities. Negative associations may point to parasitism,predation, and competition. Nevertheless, the probability thatcorrelations indeed mirror these theoretical assumptions, arein?uenced, e.g., by the generation time of involved organisms,the turnover time of available nutrients, and also the samplingresolution at which microbial dynamics are observed. Finally,a predator-prey relation between two organisms might bein?uenced by a third factor (e.g., bottom-up control throughnutrients of the prey, top-down control of the predator).Our data set o?ers a striking example: We found a clearpositive contemporaneous correlation of Rhodomonas withthe raptorial feeder B. planctonicum (Figures 5A,6), whichis a de?nite predator-prey relationship (see details above).The growth of Rhodomonas is linked to high phosphorusconcentrations, increased insolation and stable thermalstrati?cation in spring. Ciliates can reach equivalent growthrates as algae, thus, abundances of prey and predators coincided.Finally, we even observed a synchronous decline of Rhodomonasand B. planctonicum, possibly caused by physical forces of aninternal wave or by phosphorus limitation of cryptophytes.Additionally, the abundance of metazooplankton (rotifers,daphnids) increased (data not shown), exerting a top-downcontrol on both, algae and ciliates. This example highlights thatpredator-prey interactions of protists may be indeed positivelycorrelated when the sampling e?ort is high enough to followthe dynamics at high temporal resolution. However, we haveto state that the co-occurrence between Rhodomonas and B.planctonicum was only re?ected by PCC and not by LSA.Although LSA proved to be successful in ?nding time-shifted patterns in several studies (Needham et al., 2013; Chowet al., 2014), this statistical approach was not adequate forthe evaluation of our data describing population dynamicsat a high temporal resolution. The fast succession of singleand not repetitive short-living population peaks of variousprotists, which is a main character of spring bloom dynamicsin freshwater, caused an over-proportional high number ofsigni?cant connections when analyzed with LSA (Table 2,Figure 7). Nevertheless, network based analyses may help toformulate testable hypotheses about possible interactions ofspecies that co-occur, co-vary or do not co-occur (Chow et al.,2014). However, to verify these hypotheses, it is still necessary torecognize and isolate protists for further experiments.AUTHOR CONTRIBUTIONSTP, JP, FP designed research. BE performed ciliate analyses. GPperformed ciliate drawings. EE performed bacterial analyses, FPalgal analyses. TP analyzed data and created ?gures. TP and BEwrote the publication.ACKNOWLEDGMENTSThis study was ?nanced by the Swiss National Fund (SNF3 and SNF 0603). We thank ourcaptain and technician Eugen Loher for all his help duringsampling and Regula Illi for phytoplankton taxonomy. We thankJ?rg Villiger for help with LSA analyses. We are thankful toMichaela Salcher for fruitful discussions and comments on themanuscript.REFERENCESAmblard, C., Sime-Ngando, T., Rachiq, S., and Bourdier, G. (1993).Importance ofciliated protozoa in relation to the bacterial and phytoplanktonic biomass in anoligo-mesotrophic lake, during the spring diatom bloom. Aquat. Sci. 55, 1?9.doi: 10.1007/BFBeutler, M., Wiltshire, K. H., Meyer, B., Moldaenke, C., L?ring, C., Meyerh?fer, M.,et al. (2002). A ?uorometric method for the di?erentiation of algal populationsin vivo and in situ.Photosyn. Res. 72, 39?53. doi: 10.1023/A:8Bleiker, W., and Schanz, F. (1989). In?uence of environmental factors on thephytoplankton spring bloom in Lake Z?rich. Aquat. Sci. 51, 47?58. doi:10.1007/BFCarrias, J. F., Amblard, C., and Bourdier, G. (1998). Seasonal dynamics andvertical distribution of planktonic ciliates and their relationship to microbialfood resources in the oligomesotrophic Lake Pavin. Arch. Hydrobiol. 143,227?255.Chow, C.-E. T., Kim, D. Y., Sachdeva, R., Caron, D. A., and Fuhrman, J. A. (2014).Top-down controls on bacterial community structure: microbial networkanalysis of bacteria, T4-like viruses and protists. ISME J. 8, 816?829. doi:10.1038/ismej.Eckert, E. M., Salcher, M. M., Posch, T., Eugster, B., and Pernthaler, J. (2012). Rapidsuccessions a?ect microbial N-acetyl-glucosamine uptake patterns during alacustrine spring phytoplankton bloom. Environ. Microbiol. 14, 794?806. doi:10.1111/j.11.02639.xFrontiers in Microbiology | www.frontiersin.org 12 November 2015 | Volume 6 | Article 1289
Posch et al. Ciliates Interactions During SpringFaust, K., and Raes, J. (2012). Microbial interactions: from networks to models.Nat. Rev. Microbiol. 10, 538?550. doi: 10.1038/nrmicro2832Foissner, W., Berger, H., and Schaumburg, J. (1999). Identi?cation and Ecologyof Limnetic Plankton Ciliates. Informationsberichte des Bayer. Munich:Landesamtes f?r Wissenschaft.Foissner, W., and P?ster, G. (1997). Taxonomic and ecologic revision of Urotrichs(Ciliophora, Prostomatida) with three or more caudal cilia, including a user-friendly key. Limnologica 27, 311?347.Fuhrman, J. A., Cram, J. A., and Needham, D. M. (2015). Marine microbialcommunity dynamics and their ecological interpretation. Nat. Rev. Microbiol.13, 133?146. doi: 10.1038/nrmicro3417Garneau, M.-?., Posch, T., Hitz, G., Pomerleau, F., Pradalier, C., Siegwart, R., et al.(2013). Short-term displacement of Planktothrix rubescens (cyanobacteria) ina pre-alpine lake observed using an autonomous sampling platform. Limnol.Oceanogr. 58, . doi: 10.4319/lo..1892Gong, J., Dong, J., Liu, X., and Massana, R. (2013). Extremely high copynumbers and polymorphisms of the rDNA operon estimated from singlecell analysis of oligotrich and peritrich ciliates. Protist 164, 369?379. doi:10.1016/j.protis.Horn, W., Mortimer, C. H., and Schwab, D. J. (1986). Wind-induced internalseiches in Lake Zurich observed and modeled. Limnol. Oceanogr. 31,. doi: 10.4319/lo..1232J?rgens, K., and ?imek, K. (2000). Functional response and particle size selectionof Halteria cf. grandinella, a common freshwater oligotrichous ciliate. Aquat.Microb. Ecol. 22, 57?68. doi: 10.3354/ame022057Lynn, D. H. (2008). The Ciliated Protozoa. Characterization, Classi?cation, andGuide to the Literature,3rd Edn. Springer.Mathes, J., and Arndt, H. (1995). Annual cycle of protozooplankton (ciliates,?agellates and sarcodines) in relation to phyto- and metazooplankton in LakeNeum?hler See (Mecklenburg, Germany). Arch. Hydrobiol. 134, 337?358.M?ller, H. (1989). The relative importance of di?erent ciliate taxa in thepelagic food web of Lake Constance. Microb. Ecol. 18, 261?273. doi:10.1007/BFM?ller, H. (1991). Pseudobalanion planctonicum (Ciliophora, Prostomatida):ecological signi?cance of an algivorous nanociliate in a deep meso-eutrophiclake. J. Plankt. Res. 13, 247?262. doi: 10.1093/plankt/13.1.247M?ller, H., Geller, W., and Sch?ne, A. (1991a). Pelagic ciliates in Lake Constance:comparison of epilimnion and hypolimnion. Verh. Internat. Verein Limnol. 24,846?849.M?ller, H., and Schlegel, A. (1999). Responses of three freshwater planktonicciliates with di?erent feeding modes to cryptophyte and diatom prey. Aquat.Microb. Ecol. 17, 49?60. doi: 10.3354/ame017049M?ller, H., Sch?ne, A., Pinto-Coelho, R. M., Schweizer, A., and Weisse, T. (1991b).Seasonal succession of ciliates in Lake Constance. Microb. Ecol. 21, 119?138.doi: 10.1007/BFM?ller, H., and Weisse, T. (1994). Laboratory and ?eld observations on thescuticociliate Histiobalantium from the pelagic zone of Lake Constance, FRG.J. Plankt. Res. 16, 391?401. doi: 10.1093/plankt/16.4.391Needham, D. M., Chow, C.-E. T., Cram, J. A., Sachdeva, R., Parada, A., andFuhrman, J. A. (2013). Short-term observations of marine bacterial and viralcommunities: patterns, connections and resilience. ISME J. 7, . doi:10.1038/ismej.2013.19P?ster, G., Auer, B., and Arndt, H. (2002). Pelagic ciliates (Protozoa, Ciliophora)of di?erent brackish and freshwater lakes?a community analysis at the specieslevel. Limnologica 32, 147?168. doi: 10.-05-6P?ster, G., Sonntag, B., and Posch, T. (1999). Comparison of a direct live count andan improved quantitative protargol stain (QPS) in determining abundance andcell volumes of pelagic freshwater protozoa. Aquat. Microb. Ecol. 18, 95?103.doi: 10.3354/ame018095Pomati, F., Kraft, N. J. B., Posch, T., Eugster, B., Jokela, J., and Ibelings, B. W.(2013). Individual cell based traits obtained by scanning ?ow-cytometry showselection by biotic and abiotic environmental factors during a phytoplanktonspring bloom. PLoS ONE 8:e71677. doi: 10.1371/journal.pone.0071677Posch, T., Jezbera, J., Vrba, J., Simek, K., Pernthaler, J., Andreatta, S., et al. (2001).Size selective feeding in Cyclidium glaucoma (Ciliophora, Scuticociliatida) andits e?ects on bacterial community structure: a study from a continuouscultivation system. Microb. Ecol. 42, 217?227. doi: 10.000114Posch, T., K?ster, O., Salcher, M. M., and Pernthaler, J. (2012). Harmful?lamentous cyanobacteria favoured by reduced water turnover withlake warming. Nat. Clim. Change 2, 809?813. doi: 10.1038/nclimate1581Rogerson, A., Finlay, B. J., and Berninger, U. G. (1989). Sequesteredchloroplasts in the freshwater ciliate Strombidium viride (Ciliophora:Oligotrichida). Trans. Am. Microsc. Soc. 108, 117?126. doi: 10.2307/3226368Ruan, Q., Dutta, D., Schwalbach, M. S., Steele, J. A., Fuhrman, J.A., and Sun, F. (2006). Local similarity analysis reveals uniqueassociations among marine bacterioplankton species and environmentalfactors. Bioinformatics 22, . doi: 10.1093/bioinformatics/btl417Salcher, M. M. (2014). Same same but di?erent: ecological niche partitioningof planktonic freshwater prokaryotes. J. Limnol. 73, 74?87. doi:10.4081/jlimnol.?imek, K., J?rgens, K., Nedoma, J., Comerma, M., and Armengol, J. (2000).Ecological role and bacterial grazing of Halteria grandinella: small freshwateroligotrichs as dominant pelagic ciliate bacterivores. Aquat. Microb. Ecol. 22,43?56. doi: 10.3354/ame022043?imek, K., Nedoma, J., Znachor, P., Kasalicky, V., Jezbera, J., Horn?k, K., and Seda,J. (2014). A ?nely tuned symphony of factors modulates the microbial foodweb of a freshwater reservoir in spring. Limnol. Oceanogr. 59, . doi:10.4319/lo..1477Skibbe, O. (1994). An improved quantitative protargol stain for ciliates and otherplanktonic protists. Arch. Hydrobiol. 130, 339?347.Skogstad, A., Granskog, L., and Klaveness, D. (1987). Growth of freshwaterciliates o?ered planktonic algae as food. J. Plankt. Res. 9, 503?512. doi:10.1093/plankt/9.3.503Sommaruga, R., and Psenner, R. (1993). Nanociliates of the order prostomatida:their relevance in the microbial food web of a mesotrophic lake. Aquat. Sci. 55,179?187. doi: 10.1007/BFSommer, U., Adrian, R., De Senerpont Domis, L., Elser, J. J., Gaedke,U., Ibelings, B., et al. (2012). Beyond the Plankton Ecology Group(PEG) model: mechanisms driving plankton succession. Annu. Rev.Ecol. Evol. Syst. 43, 429?448. doi: 10.1146/annurev-ecolsys-60251Sommer, U., Gliwicz, Z. M., Lampert, W., and Duncan, A. (1986). The Peg-Modelof seasonal succession of planktonic events in fresh waters. Arch. Hydrobiol.106, 433?471.Sonntag, B., Posch, T., Klammer, S., Teubner, K., and Psenner, R. (2006).Phagotrophic ciliates and ?agellates in an oligotrophic, deep, alpine lake:contrasting variability with seasons and depths. Aquat. Microb. Ecol. 43,193?207. doi: 10.3354/ame043193Sonntag, B., Summerer, M., and Sommaruga, R. (2011). Factors involved in thedistribution pattern of ciliates in the water column of a transparent alpine lake.J. Plankt. Res. 33, 541?546. doi: 10.1093/plankt/fbq117Steele, J. A., Countway, P. D., Xia, L., Vigil, P. D., Beman, J. M., Kim,D. Y., et al. (2011). Marine bacterial, archaeal and protistan associationnetworks reveal ecological linkages. ISME J. 5, . doi: 10.1038/ismej.2011.24Stoeck, T., Breiner, H.-W., Filker, S., Ostermaier, V., Kammerlander, B., andSonntag, B. (2014). A morphogenetic survey on ciliate plankton from amountain lake pinpoints the necessity of lineage-speci?c barcode markersin microbial ecology. Environ. Microbiol. 16, 430?444. doi: 10.-Stoecker, D., Johnson, M., Devargas, C., and Not, F. (2009). Acquired phototrophyin aquatic protists. Aquat. Microb. Ecol. 57, 279?310. doi: 10.3354/ame01340Tirok, K., and Gaedke, U. (2007). Regulation of planktonic ciliate dynamics andfunctional composition during spring in Lake Constance. Aquat. Microb. Ecol.49, 87?100. doi: 10.3354/ame01127Uterm?hl, H. (1958). Zur vervollkommnung der quantitativen Phytoplankton-Methodik. Mitt. Int. Verein. Theor. Angew. Limnol. 9, 1?38.Weisse, T., Karstens, N., Meyer, V. C. L., Janke, L., Lettner, S., and Teichgr?ber,K. (2001). Niche separation in common prostome freshwater ciliates: thee?ect of food and temperature. Aquat. Microb. Ecol. 26, 167?179. doi:10.3354/ame026167Frontiers in Microbiology | www.frontiersin.org 13 November 2015 | Volume 6 | Article 1289
Posch et al. Ciliates Interactions During SpringWeisse, T., and M?ller, H. (1998). Planktonic protozoa and the microbial foodweb in Lake Constance. Arch. Hydrobiol. Spec. Issues Advanc. Limnol. 53,223?254.Xia, L. C., Ai, D., Cram, J., Fuhrman, J. A., and Sun, F. (2013). E?cientstatistical signi?cance approximation for local similarity analysis ofhigh-throughput time series data. Bioinformatics 29, 230?237. doi:10.1093/bioinformatics/bts668Zingel, P., and N?ges, T. (2010). Seasonal and annual populationdynamics of ciliates in a shallow eutrophic lake. Fund.Appl. Limnol. 176, 133?143. doi: 10.35/2010/0176-0133Con?ict of Interest Statement: The authors declare that the research wasconducted in the absence of any commercial or ?nancial relationships that couldbe construed as a potential con?ict of interest.Copyright ? 2015 Posch, Eugster, Pomati, Pernthaler, Pitsch and Eckert. Thisis an open-access article distributed under the terms of the Creative CommonsAttribution License (CC BY). The use, distribution or reproduction in other forumsis permitted, provided the original author(s) or licensor are credited and that theoriginal publication in this journal is cited, in accordance with accepted academicpractice. No use, distribution or reproduction is permitted which does not complywith these terms.Frontiers in Microbiology | www.frontiersin.org 14 November 2015 | Volume 6 | Article 1289
CitationsCitations7ReferencesReferences58ArticleMay 2017ArticleFull-text availableApr 2017ArticleFull-text availableNov 2016Freshwat BiolShow moreProjectPrivate Profile[...]This is a 3 years project funded by Italian Ministry of Foreign Affair (MAE) in the context of bilateral cooperation Italy-China. The full title of the project is “Efficiency of different disinfect…& Project[...]We propose investigations of life strategies related to genomic and ecophysiological traits of representative strains of the key groups of freshwater Betaproteobacteria, i.e. the genera Limnohabita…& Data provided are for informational purposes only. Although carefully collected, accuracy cannot be guaranteed. Publisher conditions are provided by RoMEO. Differing provisions from the publisher's actual policy or licence agreement may be applicable.This publication is from a journal that may support self archiving.}