Comparison of the community structure of planktonic bacteria in ballast water from entry ships and l

Available online at www.sciencedirect.com小ScienceDirectProgress inNatural ScienceEL SEVIERProgress in Natural Science 19 (2009) 947-953www.elsevier. .com/locate/pnscComparison of the community structure of planktonic bacteriain ballast water from entry ships and local sea water in Xiamen PortYing Ma", Hejian Xiong b, Senming Tang°*, Qingshuang Yang d, Minjuan Lia: Fisheries College of Jimei University, Xiamen 361021, Chinab Bioengineering College of Jinei Uninersity, Xiamen 361021, China“The Third Institute of Oceanography, SOA, Xiamen 361005, ChinadXiamen Entry- Exit Inspection and Quarantine Burean, Xiamen 361012, ChinaReceived 23 July 2008; received in revised form 11 August 2008; accepted 14 September 2008AbstractIn this study, the bacterial community structures in samples of ballast water collected from a ship from Singapore and of local seawater collected from Xiamen Port were compared using restriction fragment length polymorphism (RFLP) and 16S rDNA sequenceanalysis. Except for dominant a-Proteobacteria that are common to both systems, the bacterial community structures of the two systemswere quite diferent. Most of the clones derived from the diferent systems were grouped into diferent phylogenetic clusters, and the sys-tems share only one common RFLP pattern. The ballast water, which is likely from clean ofshore waters, contains sequences specific toa- and y-Proteobacteria. Phylogenetic analysis revealed that the ballast water contained sequences belonging to attached bacteria andbacteria commonly found in the open sea, as well as many novel sequences. In addition, no known pathogenic bacteria were detectedin the ballast water samples. Conversely, water samples from Xiamen Port were apparently afected by the near shore eavironments.Speiall, in addition to C and r-Proteobacteria, water from Xiamen Port contained B- and δ-Proteobacteria, Synechococcus, Bacter-oidetes and Actinobacteria, which are common in coastal environments. Addionally, four pathogenic bacterial sequences and one plas-mid sequence of a potential red tide forming alga were detected in the water from Xiamen Port, which suggests that the local sea water ispolluted. The results of this study can be used as background information to assess the risk associated with the introduction of non-indig-enous species to local systems and to establish ballast water management systems.◎2009 National Natural Science Foundation of China and Chinese Academy of Sciences. Published by Elsevier Limited and Science inChina Press. All rights reserved.Keywords: Xiamen; Bllst water; Bacteri; 16S rDNA; Restriction fragment length polymorphism (RFLP)1. Introductionwater contains a diverse mix of phytoplankton, includingHAB, cysts and eggs of zooplankton, benthic organisms,It is estimated that 2- 3 bilion tons of ballast water are and bacteria and viruses [2]. Although many studies thatcarried around the world each year [1] As a result, translo- have been conducted to evaluate ballast invaders havecation of organisms by ships is one of the most important focused on macro-organisms such as metazoans [3],issues threatening the naturally evolved biodiversity of micro-organisms present in ballast water have largely beenlocal aquatic systems, and the consequences of the uninten-ignored. This lack of attention is likely because microbialtional introduction of alien species associated with ships invaders cannot be detected without the aid of compoundhave gained increased attention in recent years. Ballastmicroscopes and their presence is generally only noticed whenthey have large jimnarts s11ch as dring red tides or outbreaks‘Corresponding author. Tel./fax: +86 592 2195260.of iln中国煤化_ I :ntion, microbes are theEmail dress; tsm@2lcn.com (S. Tang).mostTYHCNMHG-ironment, arrive in the1002-0071/$ - see front matter 0 2009 National Natural Science Foundation of China and Chinese Academy of Sciences. Published by Elsevier Limitedand Science in China Press. All rights reserved.do:10.1016j.psc.2008.09.007948Y. Ma et al. IProgress in Natural Science 19 (2009) 947 -953greatest numbers in ballast tanks, and probably have thetrifuge tube, after which the filter species were rinsed againgreatest chances of survival upon introduction into a newwith 0.5 ml of lysis buffer. Next, the lysates were pooledenvironment. Ruiz et al. [5] reported that the mean numberand the cells were then further lysed by the addition of I/of bacteria and virus-like particles in the ballast water of100 volume of proteinase K (0.5 mg/ml) and I/10 volumevessels arriving in Chesapeake Bay from foreign portsof sodium dodecyl sulfate (1%, SDS) followed by incuba-was8.3x 108 per liter and 7.4 x 10' per liter, respectively.tion for 30 min at 55 °C. The DNA was extracted from cellHowever, to date, most studies conducted to evaluatelysates by the phenol- chloroform extraction method. Themicro-organisms in ballast water have primarily focusedPCR was performed to amplify the bacterial 16S rDNA.on harmful alga [6,7], pathogens and viruses [5,8,9], andPrimer sequences were 27F (5-AGAGTTTGATCATthere is a lack of information regarding other non-indige-GGCTCAG-3) and 1492R (5/-GGTACCTTGTT ACG-nous micro-organisms. As a result, it is dificult to fullyACTT-3). The amplification reaction mixture consistedassess the ecological and economic impacts of invadingof 0.5 μM of each primer, 200 μuM dNTPs, 5μl of 10xspecies on local communities.PCR buffer, 1 unit of Taq DNA polymerase (TaKaRa Bio-Xiamen is an important shpping center on the south-technology Co., Dalian, China) and 2 μ of DNA template.east coast of China. It has been reported that XiamenThe amplification conditions were as follows: initial dena-Port receives more than 4 million tons of ballast waterturation at 94°C for 5 min, followed by 30 cycles ofof foreign origin annually, excluding ballast water94°Cfor1min,55°Cfor1min,and72°Cfor2min,withreceived from other domestic ports [10] Some of thisa final extension at 72。C for 10 min. The amplified prod-ballast water is taken on in areas that are known to haveucts were then gel-purified and ligated into the pMD18-Tbeen impacted by infectious diseases, which may result invector (TaKaRa Biotechnology Co., Dalian, China), afterthe dispersal of pathogens or harmful algae. In this study,which they were transformed into competent Escherichiawe collcted a ballast water sample from a ship arrivingcoli DHSa cells. The ampicilin-resistant clones were thenin Xiamen Harbor and then evaluated the bacterialrandomly picked and screened for inserts by performinggenetic diversity using a 16S rRNA gene clone librarycolony PCR using M13 primers specific for the vectornd sequences analysis. We then used this information(Invitrogen, Shanghai, China).to compare the bacterial community structure of the bal-last water to that of the local seawater. It is hoped that2.3. PCR-RFLP screening and statistical analysis of thethe information generated in this study can be used as clone librariesbackground information to enable better assessment andmanagement of the introduction of non-indigenous spe-A total of 65 clones (35 from water collected fromcies, including potentially pathogenic micro-organisms,Xiamen Port and 30 from ballast water) with inserts ofvia ballast-water transfer.the expected length were selected for subsequent RFLPanalysis. A 1:10 dilution of the colony PCR products that2. Materials and methodshad been amplified using the M13 primer was then re-amplified with the 27F and 1492R primers under the same2.1. Sample collectionPCR conditions described above. The PCR products werethen separately digested using the restriction endonucle-A ballast water sample was collected from a ship anchoredases, Hhal and Afal (TaKaRa Co., Dalian, China). Thein Xiamen Harbor, China(24929'N, 118904'E), in Septemberrestriction fragments were then visualized on a 2.0% aga-of 2007. The source region of the ballast water was Singa-rose gel, after which the clones were discriminated accord-pore. The ballast water sample comprised subsamples col-ing to their RFLP patterns. The phylotype diversity in thelected from the subsurface at five different locations on-clone libraries obtained through PCR-RFLP analysis wasboard the ship. Five liters of the water were then filteredthen subjected to statistical analysis. Speifially, the fol-through 47 mm diameter polysulfone flters with a pore sizelowing indices were calculated: (1) Taxa, the total numbersof 0.2 μum (PALL, Ann Arbor, USA) at <0.03 MP. Subsur-of RFLP patterns in each library; (2) individuals, the totalface seawater (5 l) from Xiamen Harbor was also filterednumber of clones examined; (3) coverage [11], which wasusing the same procedure. The filters were then immediatelydetermined using the following equation: Coverage =frozen and stored at -20 °C until DNA extraction.1 - (N/Individuals), where N is the number of clones thatoccurred only once; (4) diversity indices (dominance, even-2.2. DNA extraction, PCR amplification and clone libraryness, Shannon), which were calculated using the statisticalconstruction中国煤化工2/ohammer/past).The filters containing the samples were cut into small 2.4.TYHCN M H Gertic analysispieces and then incubated in 1.5 ml of lysis solution(45 mM glucose, 23 mM Tris [pH 8.0], 59 mM EDTA) con-Representative clones showing unique RFLP patternstaining 0.5 mg/ml lysozyme. After being incubated at 37 °Cwere selected for sequencing, which was conducted usingfor 2 h, the lysates were transferred to another sterile cen-an ABI model 377 automated DNA sequence analyzer.Y Ma et al IProgress in Natural Science 19 (200) 947-953949(Appied Biosystems, Perkin-Elmer) and the 27F sequenc- last water (Fig. 1). These results are in accordance with theing primer. All nucleotide sequences were checked for puta-higher dominance and lower evenness values that weretive chimeras using RDP CHIMERA _CHECK [12] andobserved for water from Xiamen Port (Table l). In addi-then compared with known 16S rDNA sequences in thetion, greater than 70% of the RFLP patterns in each clonedatabase using a BLASTN search (http://www.ncbi.nlm.library were observed only one time, which reflects the sig-nih.gov/BLAST/). Multiple alignments were then per-nificant genetic diversity of the bacterial populations informed using the neighbor-joining algorithm in the Clu-both water systems. Additionally, the Shannon index wasstalX software [13], after which a phylogenetic tree was higher for water from Xiamen Port than for the ballastconstructed using the MEGA3 software [14], Bootstrapwater (Table 1), suggesting that the water in Xiamen Portvalues were obtained with 100 replicates.has a higher overall diversity than the ballast water. Theseresults are supported by the figure describing the RFLP fre-quency (Fig. 1). Specifcally, the trend line generated for2.5. Nucleotide sequence accession numbersthe ballast water sample is smoother, having a lower start-The cloned sequences have been deposited in GenBanking value and being shorter than the trend line for waterfrom Xiamen Port. These findings suggest that the concen-under Accession Nos. EU877627 to EU877670. .tration of the most dominant bacteria in the ballast waterwas lower than that of the most dominant bacteria in the3. Resultswater from Xiamen Port. Furthermore, these results indi-cate that the bacterial community structure of the ballst3.I. PCR screening, pattern frequency and statisticalwater had a higher evenness and lower diversity than thatanalysis of the two bacterial 16S rDNA clone librariesof water from Xiamen Port. In addition, the two clonelibraries shared only one common RFLP pattern (Fig. I,After screening the colonies for inserts by colony PCR3 and 7*), and this RFLP pattern only accounted for 4%using the M13 primer, a total of 65 clones were subjectedand 5% of the total RFLP patterns observed in water fromto RFLP analysis, 35 of which were from the Xiamen PortXiamen Port and ballast water, respectively. Takenwater library and 30 of which were from the ballast watertogether, these findings indicate that the bacterial groupslibrary (Table 1). Restriction analysis indicated that thein the two clone libraries differed significantly.Xiamen clone library contained 25 RFLP patterns, andthat the ballast water library contained 19 patterns.3.2. Comparison of bacterial community structures in the twoAlthough the number of clones used for RFLP analysisclone librariesfrom the ballast water (30 clones) was smaller than thenumber of clones from Xiamen Port (35 clones), the cover-One or two clones with unique RFLP patterns wereage value for the ballast water (53%) was higher than theselected from each library and were subjected to sequenc-coverage value calculated for water collected from Xiamenng. After removing the chimeric sequences and a few unde-Port (34%) (Table 1). This result indicates that the sam-termined sequences, a total of 28 and 26 sequences werepling size of the ballast water was more adequate than thatobtained from the Xiamen Port water and ballast waterof Xiamen Port water, and further sampling of Xiamenclone libraries, respectively. Sequence analysis revealedPort would reveal more unique clones. However, the cover-that clones within the same RFLP pattern shared≥97%age values of both clone libraries were not high, indicatingsequence similarity, and that clones with diferent RFLPthat the diversity was likely underestimated in this study.patterns had <97% similarity, suggesting that the divisionDetails regarding the frequency of the RFLP patternsof the RFLP patterns was reliable. Sequences were thenfor diferent clones are shown in Fig. 1. The most abundantassigned to major groups based on BLAST similaritiesRFLP pattern in water from Xiamen Port corresponded to and phylogenetic analysis. The species-composite clonenine clones (25.1% of the total clones), whereas the mostlibraries are shown in Fig. 2.abundant RFLP pattern in the ballast water only corre-All the cloned sequences fell into seven major lineages ofsponded to six clones (20% of the total clones). Further-the bacterial domains, the ac-, B-, r-, and 8-Proteobacteria,more, there were 23 unique RFLP patterns in theBacteroidetes, Actinobacteria, Cyanobacteria (Synechococ-libraries corresponding to water from Xiamen Port (92%cus), as well as a group originating from chloroplasts of theof the total patterns and 65.7% of the total clones), whereaseukaryote Skeletonema pseudocostatum (Fig. 2). The per-there were 14 unique RFLP patterns (73.7% of the totalcentage of 16S rDNA sequences from each group in thepatterns and 46.7% of the total clones) observed in the bal- total rDNA pool indicated that the community structures中国煤化工Table 1Statistical analysis of the bacterial 16S rDNA clone libraries of ballast water and wateYHCNMHG_Clone libraryIndividualsTaxaCoverage (%)DominanceEvennessShannonBallast water3019530.084440.80762.732Xiamen Port water352:340.092240.72422.896950Y. Ma et al.lProgress in Natural Science 19 (200) 947-953108■Bllast water■Xiamen port waterL从LLLLLLLLLL30579113151719212325135791113.15”1719RFLP pttreRFLP patemFig 1. RFLP pattern frequency of the bacterial 16S rDNA clone libraries of water from Xiamen Port and bllat water. Frequency values were rankeddecreasingly. The presence of the same RFLP pattern in the two libraries is indicated by a superscript ltter a.9%口Alpha proteobacteria3%/0 Beta proteobacteriaGamuma protobacteria48%国Delta proteobacteria口Bacteroidetes目Actinobacteria/62%日Synechococcus19%國PlastidXiamen port waterBallast waterFig. 2. The percentage of microbial taxa estimated using the 16S rDNA library.of the two clone libraries were different. Specifically, clonesstructed using the neighbor-joining method. The maincorresponding to a- and y-Proteobacteria were found inbranches of the tree possessed high bootstrap values, indi-both libraries, while clones corresponding to the remaindercating that the tree topology had a high confidence levelof the bacterial groups identified were only observed in the(Fig. 3).Xiamen clone library. Furthermore, the two librariesThe a-Proteobacteria group primarily comprised speciesshared only one common RFLP pattern, which indicatesafliated with Rhodobacteraceae and a few unidentified a-that the species components within the Q- and y-Proteobac- Proteobacterial clones. In addition, with the exception ofteria differed between the two clone libraries (Fig. 1). Thiscloned sequences XM-30 and XM-40, the a-Proteobacteri-finding was supported by the results of the phylogenetical sequences in water from Xiamen Port were ≥97%analysis, which revealed that most of the clones derivedhomologous with sequences already present in Genbank.from different libraries were grouped into different clustersThese sequences were primarily related to sequences(Fig. 3). When the clone abundance of the diferent groupsretrieved from coastal environments including cloneswas evaluated, even though a-Proteobacteria were domi-PL 4a9f [15], CB01D03, SIMO-4275 and BlI from coastalnant in both libraries, it accounted for only 48% of thewater (GenBank description), clone DS158 from a man-clones in the Xiamen Port library, while it accounted forgrove ecosystem, clone TH1-60 from a freshwater lakegreater than 60% of the clones in the ballast water clone[16], and Roseovarius crassostreae CV919-312 from dis-library (Fig.2). Although y-Proteobacteria were the secondeased juvenile oysters (GenBank description). These resultsmost dominant group in both libraries, it accounted for. suggest that ax-Proteobacteria in the water in Xiamen Portonly 19% of the clones in the Xiamen Port library andwere common to coastal environments. Conversely, more38% of the clones in the ballast water library, respectively. than half of the a-Proteobacteria afiliated sequences iden-tifiehad no close matches3.3. Phylogenetic analysis of the bacterial 16S rDNA(<97中国煤化工ch sugsts that the ba-sequences in the two clone librarieslast，C N M H G novel a-Proteobacteria.Clones that had close matches were primarily related toForty-four sequences (one representative sequencebacteria that were afiliated with specific bosts, bacteriafrom each RFLP pattern) were used for phylogeneticfrom the open sea, and bacteria that had a special function.analysis. For the analysis, a phylogenetic tree was con- These closest relatives included bacterium DG1297, whichY. Ma et al IProgress in Natural Science 19 (2009) 947- 9539511100 XM-38 (877645,1)12- Clone P As9f (AY50451)W49(E0877670. 3)Clom SM0425 0042146)Bactenum DG1297 00486508)100; BW-98077665. 1)- BW11E0877565 1)Tmgws BW3766561510 RWscr7e7 AC2.2 E0921a- ProteobacteriaBW-18 E077659.1Pr XM-50E0877651. im Bactenum C49 AB30e21)XM-30 E8774.111XM-40(ELe10L Clore HIANG 521 A3383. BW-(EV77653.100 XM-12E177632, )64TClonmg 2016271 (EU237453)”XM-10(EU77631. 1)10- Clone B11 (EU010136)Cione RO18BA04 (Ar947895)B-21E7877664. 1! Clone 6C233139 (EU805148)XM2&/8E8776401118tiongPB-41 (AMB49452)β- ProteobacteriaI9-BaCorumHME624 E43313121BW4(E0877652.1)220p XM-2(EU77628.1F379065); X231E1877637.6o Clome 4C230387 (082988BW14EUE XM52776471l Isolate HAI 40b (AM117932)26100.XM-341EU877642. 1)|y- Proteobacteria: Acmefobacter punvi (FM164636)3W-46 E2877669.1)Hsomm9 aghensis (AY19406629Stren 141/010105 AY612750)30100 XM-18 EU77634.1)L Vebvro pelggius (0472)∞LM-19LEL0876517492)321C Clone Belgcex00510 14016 00351791)，δ- Proteobacteria} XM-7 EU877629 ,33100 XM-15(EU877633. 1)ActinobacteriaClome GASP (E299566)5100 XM-20 EU877636,1)]Eukaryotlc plastld35[ sercrorcus spUuone c114 EF471456.5 JCyanobacterla36XM24(EL877638. 1)| Bacteroldetes3710 Clone PiITb (AE294891)3805~Clones from this study are indicated in boldface and are designated as BW-n and XM-n, in which BW and XM indicate that the sequences were derivedfrom bllsas water and water from Xiamen Port, rspetively, and n represents the number of diferent clones. The remaining sequences were obtainedfrom GenBank. Numbers in parentheses that fllow the acession numbers indicate the occurrence frequency of the RFLP pattern in the clone library.Bootstrap values above 50 (100 iterations) are shown at each node. The scale bar represents the nucleotide substitution percentage.中国煤化工was attached to a dinflagellate Scripsiella sp., Roseobac- alsoMYHc N M H Gd 20162U71 from theteria sp. AC2-A2, which was attached to a marine sponge,Pacific and Atlantic oceans, respectively (GenBank descrip-and epibiotic bacterium 11ANG521, which was attached to tion), and bacterium C49, which is known to have the abil-a squid. In addition, isolates from the ballast water wereity to degrade hydrocarbons [17].952Y. Ma et al. 1 Progress in Natural Science 19 (2009) 947- -953The y-Proteobacteria afiliated clones were the seconddisadvantages, the clone library approach provided valu.most dominant in both libraries, accounting for 19% andable information that has allowed us to identify the domi-38% of the clones in the Xiamen Port and the ballast waternant bacterial composition in the ballast water, and tolibraries, respectively. Three of the cloned sequences (XM- compare the community structure of the ballast water to34, XM-18 and XM-19) from Xiamen Port were most clo-that of water from Xiamen Port.sely related (98 99% sequence similarity) to sequences ofThe community structure of bacteria in the water fromthe following pathogenic bacteria that are associated withXiamen Port was apparently influenced by input fromdisease in flounder (based on the description in GenBank):land-based sources of fresh water. In addition to the prev-Acinetobacter junii [18] Vibrio pelagius [19] and Vibrio oli-alence of a~ and y-Proteobacteria in seawater, β- Proteobac-vaceus. Other clones (XM-1, XM-2 and XM-23) fromteria, which are typical in freshwater, were also found inwater in Xiamen Port were primarily related to cloneswater from Xiamen Port. The inshore characteristics ofrecovered from near shore waters. The dominant y-Proteo-the microbial community in water collected from Xiamenbacteria group from the ballast water library contained sixPort was also supported by the phylogenetic analysis,clones (represented by clone BW-14). These clones werewhich revealed that the cloned sequcnces were primarilymost closely related to a halophilic marine bacterium,related to sequences in GenBank that had been isolatedHAL40b，which was isolated from the boreal sponge,from various coastal systems. Additionally, clones closelyHaliclona sp. [20]. The rest of the clones occurred only oncerelated to pathogens inchuding R. crassostreae, V. pelagius,and were closely related to the algicidal bacterium, Altero-V. olivaceus and A. juni, as well as a plasmid sequence ofmonas sp. A14, to Thalassomonas ganghwensis, which wasthe potential red tide causative alga, S. pseudocostatumisolated from tidal flat sediments [21], and to strain 14II/were also found in the water from Xiamen Port. However,A01/015, which was isolated from the sea surface of coastalthese results are not surprising because Xiamen Port showsecosystems [22]a high degree of eutrophication and coastal contaminationAll the remaining clones, which included β_ and 8-Prote-as a result of discharge from aquaculture and industrialobacteria, Bacteroidetes, Actinobacteria and Cyanobacte-and domestic sewage. Furthermore, water in Xiamen Portria (Synechococus), were only found in water fromalso contained clones related to 8-Proteobacteria, Bacteroi-Xiamen Port. The B-Proteobacterial group contained threedetes, Actinobacteria and Synechococcus, while only a- andclones that were most closely related to clone PIB-41 fromy-Proteobacteria were present in ballast water. The inshorea freshwater lake in Austria and clone TLC-PA3-8, whicharea of Xiamen Port is a much more complex habitat thanwas isolated from sea water in Victoria Harbor, Hongthe ballast tanks of a ship. In addition, oxygen is probablyKong [23]. In addition, three clones (represented by clonenever limited in water in Xiamen Port because it is con-XM-36) were closely related to Synechococcus sp.stantly moving. Taken together, these factors likely enableCB1 1C04, which was isolated from surface water in thethe water in Xiamen Port to support a wider range of bac-Chesapeake Bay (GenBank description). Additionally,teria than ballast water.two clones belonged to the Actinobacteria group, whileWhen compared with the water in Xiamen Port, the bal-one clone belonged to the 8-Proteobacteria group and last water appeared to be cleaner and contained no knownone clone belonged to the Bacteroidetes group. Thesepathogens. In addition, phylogenetic analysis revealed thatclones were all closely related to clones isolated from shal-a few sequences in the ballast water were closely related tolow submarine hydrothermal systems and marine sedi-the sequences of attached bacteria and sequences of bacte-ments[24,25]. Finally, one sequence that was similar toria from the open sea. These findings suggest that the bal-the sequence encoding the plasmid of eukaryotic S. pseudo-last water was probably collected from clean offshorecostatum (similar to the common red tide causative alga,water, and that it is capable of supporting the growth ofSkeletonema costatum) was also detected, which indicatesattached bacteria. In addition, this ballast water had beenthat the water in Xiamen Port may contain harmful algae.partially replaced with water from the South China Seawhen the ship passed through the region in accordance4. Discussionwith the local ballast water discharge regulations.However, even though the ballast water was clean andWe utilized a clone library approach to analyze the bac-contained a- and y-Proteobacteria, which already existedterial community structure in the ballast water of a vesselin the local sea water, phylogenetic analysis revealed thatarriving in Xiamen Port and then compared it with the bac-the species composition varied greatly between the twoterial composition of water collected from Xiamen Port. Itclone libraries. In addition, the ballast water contained sev-is important to note that this technique may be infuencederal_the potential ecologicalby possible biases introduced by PCR, such as the forma-effec中国煤化工ast water into the localtion of chimeras, template annealing, and differences inecosCNMHMoreover, one clonedthe number of copies of 16S rDNA. In addition, only 65sequcnvcun u Uaiiasi wauci (clone BW-23) was mostclones were subjected to RFLP analysis, so it is unlikelysimilar (98% sequence similarity) to a sequence obtainedthat the inspected clones fully reflect the bacterial commu-from the dinoflgellate, Sripsiella sP., which suggests thatnity structure of the sampled water systems. Despite thesethe ballast water could also contain harmful algae.Y. Ma et al. IProgress in Natural Science 19 (2009) 947-953In the present study, only one bllst water sample was [10] Yang QS, Xiong HC, Cheng F, ct al. Risk analysis of red tide algaeanalyzed, and the residual sediments within the tank wereimport to Xiamen port by entry ship's ballast water. Insp QuarantineSci 2004;14(Suppl,):96 9 (in Chinese).not investigated. Based on the different origins of the bal-[11] Mulins TD, Brischgi TB, Krest RL, et al. Genetic comparisonsast water, future investigations that consider additionalreveal the same unknown bacterial lineages inAtlantic and Pacifieballast water samples, as well as residual sediments in bal-bacterioplankton communities. Limnol Oceanogr 1995;40:148-58.last tanks should be conducted to ensure that non-indige-[12] Maidak BL, Cole JR, Lilburn TG, et al. The RDP-II (RibosomalDatabase Project). Nucleic Acids Res 2001;29:173-4.nous species are not transferred to local aquatic systems[13] Thompson JID, Gibson TI, Plewniak F, et al. The ClustalX windowsvia ballast water.interface: flexible strategies for multiple sequence alignment aided byquality analysis tools. Nucleic Acids Res 1997;24:4876 82.Acknowledgements[14] Kumar s, Tamura K, Nei M. MEGA 3: integrated software formolccular evolutionary genetics analysis and sequence alignment.Brief Bioinform 2004:5:150- 63.This work was supported by National Special Project[15] Acinas sG, Klepac-Ceraj V, Hunt DE, et al. Fine-scale phylogenetic908-02-03-09 and the State Key Laboratory of Marinearchitecture of a complex bacterial community. NatureEnvironmental Science (Xiamen University) Project2004:4306999):S51 4.MEL0701.16] Wu QL, Zwart G, Wu J, et al. Submerged macrophytes play a keyrole in structuring bacterioplankton community composition in thelarge, shallow, subtropical Taihu Lake, China. Environ MicrobiolReferences2007911):2765-74.[17] Harwati TU, Kasai Y, Kodama Y, et al. Characterization of diverse[1] Anil AC. A perspective of marine bioinvasion. In: multiple dimen-hydrocarbon-degrading bacteria isolated from Indonesian scawater.sions of global environmental change. India: TERI Pres; 2006.Microbes Environ 20722:42-5.[2] Aridgides LJ, Doblin MA, Berke T, et al. Multiplex PCR allows[8] Linde HI, Hahn J, Holler E, et al. Septicemia due to Acinetobactersimutaneous detection of pathogens in ships' ballast water. Marjuii. J Clin Microbiol 2002:40(7):2696- 7.Pollut Bull 2004;48:1096 -101.Villamil L, Figueras A, Toranzo AE,etal. Isolation of a highly pathogenic[3] Occhipinti-Ambrogi A, Savini D. Biological invasions as a compo-Vibrio pelagius strain asciated with mass mortalities of turbot, Scophnent of global change in stressed marine ecosystems. Mar Pollut Bullthalms maxims(L), larvae. J Fish Dis 20326/):293 303.2003;46(5):542. -51.[20] Gracber 1, Kasler 1, Borchert MS, et al. Spongibacter marinus gen4] Joachimsthal EL, Ivanov V, Tay sTL, et al. Bacteriological exam-nov, sp. nov., a halophilic marine bacteriuim isolated from the borealination of ballast water in Singapore Harbour by flow cytometry withsponge Haliclona sp. 1. Int J Syst Evol Microbiol 2008;58(3):58- 90.FISH. Mar Pollut Bull 2004;49:334 43.21] Yi H, Bac KS, Chun J. Thalassomonas ganghwensis sp. nov., isolated[5] Ruiz GM, Rawlings TK, Dobbs FC, et al. Global spread offrom tidal flat sediment. Int J Syst Evol Microbiol 204;4(2):377-80.microorganisms by ships. Nature 2000;408:49 -50.[2] Agogue H, Casamayor EO, Bourrain M, et al. A survey on bacteria[6] Hllegraeff GM, Bolch CJ. Transport of dinfaellate cysts in ship'sinhabiting the sea surface microlayer of coastal ecosystems. FEMSballast water: implications for plankton biogeography and aquacul-Microbiol Ecol 2005;54:269- 80.ture. J Plankton Res 1992;14:1067 -84.[23] Zhang R, Liu B, Lau SC, et al. Particle atached and feeliving[7] Doblin MA, Popels LC, Coyne KJ, et al. Transport of the harmfulbacterial communities in a contrasting marine environment: Victoriabloom alga Aureococcus anophagefferens by oceangoing ships andHarbor, Hong Kong. FEMS Microbiol Ecol 2007;13):496 508.coastal boats. Appl EnviroMicrobiol 2004;70:6495- -500.[24] Hrayama H, Sunamura M, Takai K, et al. Culture-dependent and[8] Drake LA, Choi KH, Ruiz GM, et al. Global redistribution of-independent characterization of microbial communities associatedbacterioplankton and virioplankton communities. Biol lnvasionswith a shallow submarine hydrothermal system ocurring within a2001;3:193-9.coral reef of Taketomi Island, Japan. Appl Environ Microbiol[9] Drake LA, Ruiz GM, Galil BS, et al. Microbial ecology of ballast2007;73(23):7642-56.water during a tans-ocanic voyag. Mar Eecol Prog Ser 22:2:33 (25] Glla DC, Pemet P. Adherent baceria in heary metal contaminated13- -20.marine sediments. Biofouling 2007;23(1- 2):1-13.中国煤化工MYHCNMHG