Microbes Environ. Vol. 26, No. 4, 339–346, 2011 http://wwwsoc.nii.ac.jp/jsme2/ doi:10.1264/jsme2.ME11212

Seasonal Variations in the Community Structure of Actively Growing in Neritic Waters of Hiroshima Bay, Western Japan

AKITO TANIGUCHI1†, YUYA TADA1, and KOJI HAMASAKI1* 1Atmosphere and Ocean Research Institute, The University of Tokyo, 5–1–5 Kashiwanoha, Kashiwa, Chiba 277–8564, Japan

(Received May 6, 2011—Accepted June 30, 2011—Published online July 27, 2011)

Using bromodeoxyuridine (BrdU) magnetic beads immunocapture and a PCR-denaturing gradient gel electrophoresis (DGGE) technique (BUMP-DGGE), we determined seasonal variations in the community structures of actively growing bacteria in the neritic waters of Hiroshima Bay, western Japan. The community structures of actively growing bacteria were separated into two clusters, corresponding to the timing of phytoplankton blooms in the autumn–winter and spring–summer seasons. The trigger for changes in bacterial community structure was related to organic matter supply from phytoplankton blooms. We identified 23 phylotypes of actively growing bacteria, belonging to Alphaproteobacteria (Roseobacter group, 9 phylotypes), Gammaproteobacteria (2 phylotypes), (8 phylotypes), and Actinobacteria (4 phylotypes). The Roseobacter group and Bacteroidetes were dominant in actively growing bacterial communities every month, and together accounted for more than 70% of the total DGGE bands. We revealed that community structures of actively growing bacteria shifted markedly in the wake of phytoplankton blooms in the neritic waters of Hiroshima Bay. Key words: actively growing bacteria, bromodeoxyuridine, PCR-DGGE, seasonal variation

Phytoplankton blooms are significant events in coastal eco- nities are still largely unknown. Actively growing bacteria, systems, greatly impacting the structure and composition of defined in this study as rapidly dividing bacteria, should organic matter fields (23); bacterial community structures, in utilize increased levels of organic matter to maintain their particular, must respond and adapt to perturbations (37). Sev- rapid growth rates, and thus, should be lysed by viruses eral reports have demonstrated that phytoplankton blooms and/or grazed by protozoa (34, 49). Because of this, actively lead to marked changes in the abundance, composition, and growing bacteria should be regarded as key components driv- activities of bacterial communities (13, 40). Riemann et al. ing biogeochemical processes in oceanic environments. (38) showed that bacterial abundance, growth rates, and enzy- Bromodeoxyuridine (BrdU), a halogenated nucleoside matic activities markedly increased after manipulated diatom which can serve as a thymidine analog, is useful for analyzing bloom. Additionally, they showed that Alphaproteobacteria DNA synthesis in actively growing bacterial communities and Bacteroidetes were rapidly growing key bacteria during (e.g., 16, 33, 44, 46, 50). An immunocapture technique the decay phase of a bloom. using BrdU-labeled DNA has been applied to determine In natural coastal environments, many studies have phylogenetic affiliations and functions of bacterial groups (6, reported the temporal dynamics of bacterial communities 7, 28). Using BrdU magnetic beads immunocapture and PCR- (e.g., 22, 27, 29, 35, 39, 42). Alonso-Sáez et al. (2) reported DGGE (BUMP-DGGE), previous studies investigated phy- that both bacterial abundance and composition vary season- logenetic affiliations and changes in the spatial distributions ally and that Alphaproteobacteria and Bacteroidetes bacte- of actively growing bacteria in coastal and oceanic environ- ria are predominant in coastal waters throughout the year; ments (17, 48). These studies revealed that Roseobacter and their investigations were based on PCR-denaturing gradient Bacteroidetes bacteria dominate actively growing bacteria gel electrophoresis (DGGE) and catalyzed reporter depo- groups in both coastal and oceanic regions. In addition, the sition fluorescence in situ hybridization (CARD-FISH). On results suggest that different water masses are characterized the basis of microautoradiography combined with FISH by different phylogenetic affiliations of bacteria; however, (MAR- or Micro-FISH) analyses, Alonso-Sáez and Gasol these general conclusions must be carefully examined in each (3) showed that substantial changes occur in the activities region and moreover, temporal variations in the structure and and compositions of some bacterial phylogenetic groups composition of actively growing bacterial communities are (e.g., the Roseobacter group and SAR11 bacteria of the not fully understood. Tada et al. (45) investigated seasonal Alphaproteobacteria) during the course of a year. Although variations in phylotype-specific productivity by means of temporal variations in the abundance, activities and compo- BrdU immunocytochemistry-FISH (BIC-FISH). Their results sitions of total bacteria have been reported in these previous showed that the proportion of actively growing bacteria to studies, the dynamics of actively growing bacterial commu- total bacteria varies seasonally, from 15% to 30%, with an annual mean rate of 22%. Additionally, they found that * Corresponding author. E-mail: [email protected]; Roseobacter/Rhodobacter bacteria constituted a constant Tel: +81–4–7136–6171; Fax: +81–4–7136–6171. population of proliferating microbes and that Bacteroidetes † Present address: Graduate School of Agriculture, Kinki University, populations increased markedly after phytoplankton blooms. 3327–204 Nakamachi, Nara 631–8505, Japan The objectives of this study were to determine the detailed 340 TANIGUCHI et al. phylogenetic affiliations of actively growing bacteria and to and compared bacterial community structures between these two reveal their temporal variations throughout the year in the periods. The threshold temperatures were 12°C (January–March vs. neritic waters of Hiroshima Bay, western Japan. Specifically, April–December), 15°C (January–April vs. May–December), 18°C we inquired into the effects of phytoplankton blooms on the (December–May vs. June–November), 20°C (December–June vs. July–November), and 22°C (November–July vs. August–October). community structures of actively growing bacteria, and on the concordance of results obtained by BUMP-DGGE and Sequencing and phylogenetic analyses BIC-FISH for phylotype proportions of actively growing Excised DGGE bands were sequenced directly from PCR bacteria. products that had been reamplified with the primer set described above. Prior to sequencing, the PCR products were analyzed by Materials and Methods DGGE to confirm band positions relative to the original sample. After purification of the PCR products with a Qiaquick PCR Sampling and BrdU labeling purification kit (Qiagen, Germantown, MD, USA), bidirectional Seawater samples were collected from a depth of 5 m at the pier sequencing using the 341F/907R primer set was performed by of Kure Port (34°14'30''N, 132°33'07''E), Hiroshima Bay, western SolGent (http://www.solgent.com/). Sequences were aligned to Japan, once a month in the morning, from July 2005 to June 2006. known sequences in the GenBank database using BLAST (4). Kure Port has two neighboring rivers. An approximately 5-L sample Phylogenetic trees were constructed with the neighbor-joining was pre-filtered through a 200-µm nylon mesh to remove mesozoo- method using MEGA 4 (47). All sequences were checked by the plankton. A 1-L portion of the sample was stored at 4°C for the program Bellerophon (18). analysis of environmental factors (processing within 5 h). A 2-L Nucleotide sequence accession numbers portion of the sample was immediately filtered through a 0.22-µm Sterivex filter (Millipore, Billerica, MA, USA) with a peristaltic The nucleotide sequences were deposited in the nucleotide pump to collect bacterial cells for extraction of DNA and subsequent sequence database, DNA Data Bank of Japan (DDBJ), under analysis of bacterial community structures. The final 2-L portion of accession numbers AB367452 to AB367491. the sample was incubated with BrdU (final concentration, 1 µM; Sigma-Aldrich, St. Louis, MO, USA) in a dark bottle at ambient Results temperatures for 3 h. After incubation, bacterial cells were collected as described above. Immediately after filtration, the Sterivex filters Environmental factors were stored at −20°C until further analysis. Environmental factors, including water temperature, salinity, particulate organic carbon Environmental factors show seasonal variations, as indi- (POC) and nitrogen (PON), chlorophyll a (chl a) concentration, and cated by the data in Table 1. During the study period, water bacterial abundance, were measured as described previously (45). temperatures ranged from 11.0 to 24.8°C and salinity ranged from 31.5 to 34.5, respectively. Conspicuous phytoplankton BUMP-DGGE analysis blooms occurred in September 2005 and February 2006, BUMP-DGGE analysis was performed according to procedures described previously (17), with slightly modifications. Briefly, the generating massive amounts of organic matter. Chl a µ −1 Sterivex filter was subjected to xanthogenate–SDS DNA extraction, concentrations ranged from 1.73 to 18.0 g L , with and 1 µg of the extracted DNA was used for BrdU immunocapture. maximum concentrations occurring in February 2006. POC The total and immunocaptured BrdU-labeled DNA was used as a and PON concentrations varied from 181.3 to 980.1 µg L−1 template for PCR amplification of 16S rRNA genes using the and 30.3 to 108.7 µg L−1, respectively, with trends similar to eubacterial-specific primer 341F-GC with a 40-bp GC clamp (57- those of chl a concentrations. The C/N molar ratio ranged mer; 5'-CGCCCGCCGCGCGCGGCGGGCGGGGCGGGGGCAC from 6.2 to 10.5. Bacterial abundance (mean±standard GGGGGGCCTACGGGAGGCAGCAG-3') and universal primer deviation) ranged from (5.4±2.0)×105 to (22.1±3.9)×105 cells 907R (20-mer; 5'-CCGTCAATTCMTTTGAGTTT-3') (40). For −1 DGGE, the PCR products (approximately 60 ng) were loaded onto mL , with maximum values in April 2006. Details of a 6% polyacrylamide gel in 0.5×TAE buffer, with a denaturing environmental characteristics are presented in Tada et al. (45). gradient of 25%–70%, from top to bottom. Electrophoresis was performed at 85 V for 16 h at 60°C in an INGENY PhorU system Community structures of total and BrdU-incorporated (Ingeny, Goes, Netherlands). bacteria Statistical analyses We analyzed the specificity of BrdU immunocapture prior DGGE bands with relative intensities greater than 1% were used to DGGE analysis. When seawater samples without BrdU to perform further statistical analyses, using the ‘Vegan’ routine labeling were subjected to BrdU immunocapture (in the same (32) in the R software package (Vienna, Austria). The Jaccard manner as BrdU-labeled samples), the optimized procedure coefficient was calculated from the presence/absence of DGGE yielded no PCR products (data not shown). There was no bands and was used for cluster analysis by applying a between- lane for a BrdU-immunocaptured sample on the DGGE image group average linkage method and also a non-metric multidimen- for February 2006 (Fig. 1) because the immunocapture sional scaling (nMDS) technique. We also performed permutational multivariate analysis of variance (PERMANOVA) to test the procedure was unsuccessful, probably because of the low significance (P values) of the temporal shifts in bacterial community abundance of BrdU-incorporated bacteria and high concen- structures (5). We analyzed bacterial community data, using the tration of chl a, and therefore relatively low concentration of Jaccard coefficient, with 999 permutations to test for differences in BrdU-labeled DNA (Table 1). bacterial community structures before and after phytoplankton The DGGE profiles of PCR amplified 16S rRNA genes blooms. Variation in chl a was used to divide the year into 2 periods, from total DNA (Fig. 1) clearly represent the variations in corresponding to the timing of peaks of the blooms: March–August and September–February. Variations in POC, PON and C/N ratio total bacterial community structures; cluster analysis (Fig. 2) were regarded as the corresponding change of the blooms. distinguished community structures in autumn–winter (Sep- Additionally, based on the monthly water temperature, we divided tember 2005 to January 2006) from those in spring–summer the year into 2 periods (i.e., periods of high and low temperatures) (July and August 2005, and February to June 2006) (Fig. 2). Seasonal Variations in Actively Growing Bacteria 341

Table 1. Environmental factors at the pier of Kure Port in Hiroshima Bay from July 2005 to June 2006

POC PON Chl a BA *Proportion of Year Month WT (°C) Sal µ −1 µ −1 C/N ratio µ −1 5 −1 BrdU-positive ( g L ) ( g L ) ( g L ) (×10 cells mL ) cells (%) 2005 Jul 21.6 32.0 361.2 68.3 6.2 2.24 17.3 ± 4.6 nd Aug 24.6 32.0 613.1 95.9 7.5 8.36 20.9 ± 5.7 30.2 ± 2.6 Sep 24.8 30.8 571.7 100.8 6.6 13.9 16.2 ± 3.3 20.4 ± 4.8 Oct 24.5 33.7 383.6 57.6 7.8 7.42 14.3 ± 2.2 19.5 ± 1.9 Nov 21.5 31.5 181.3 31.5 6.7 2.26 15.1 ± 3.2 14.8 ± 4.5 Dec 15.7 33.0 188.8 30.3 7.3 1.73 12.8 ± 1.8 21.1 ± 3.6 2006 Jan 11.2 33.8 226.6 33.4 7.9 2.72 9.4 ± 2.0 23.0 ± 3.5 Feb 11.1 34.2 980.1 108.7 10.5 18.0 5.4 ± 2.0 23.9 ± 2.0 Mar 11.0 34.5 456.7 64.4 8.3 nd 17.1 ± 3.0 18.9 ± 2.1 Apr 13.0 33.0 384.8 50.5 8.9 5.14 22.1 ± 3.9 18.1 ± 3.0 May 15.6 31.7 477.6 66.4 8.4 5.05 15.1 ± 4.1 25.4 ± 3.7 Jun 17.3 32.5 621.0 87.5 8.3 7.78 18.9 ± 2.4 28.0 ± 3.0 WT: water temperature, Sal: salinity, POC: particulate organic carbon, PON: particulate oerganic nitrogen, BA: bacterial abundance, nd: no data, * measured by Tada et al. (45).

Fig. 1. DGGE images of 16S rRNA genes amplified from total and BrdU-incorporated communities from July 2005 to June 2006 at the pier of Kure Port. The image of the total community is from total DNA before incubation; the image of the BrdU-incorporated community is from immunocaptured DNA after incubation for 3 h. Lanes Rb and Fig. 3. nMDS analysis of total and BrdU-incorporated communities Rt are used for markers. Each excised and sequenced band is marked from July 2005 to June 2006 in Kure Port. Open and closed circles refer and numbered. The lane for February 2004 in the BrdU-incorporated to community structures of total and BrdU-incorporated bacteria. Con- community indicates no data. secutive time points are shown by arrows. PERMANOVA showed that the change of community structures before and after phytoplankton blooms (March–August vs. September–February) was significant (P<0.01). Stress value is an index for fitness of plotting, and a lower value is better. The stress value is 0.17, indicating a usable description which is unlikely to mislead.

Similar variation was also observed in BrdU-incorporated community data. The nMDS analysis (Fig. 3) showed that the structures of both total and BrdU-incorporated commu- nities changed markedly at the timing of phytoplankton blooms (March–August vs. September–February: PER- MANOVA, P<0.01). Water temperatures affected changes in the community structure of total bacteria only (PER- MANOVA, P<0.05 for both periods divided by 15°C, 18°C, 20°C, or 22°C) and not on BrdU-incorporated bacteria (PERMANOVA, P>0.05 for all). Sequencing and phylogenetic analyses Fig. 2. Relationships of community structures between total and Partial 16S rRNA genes from 39 bands excised from BrdU-incorporated bacteria from July 2005 to June 2006. The dendrogram was constructed by using the between-group average both total and BrdU-incorporated communities were se- linkage method for clustering. quenced (Fig. 1). BrdU-incorporated phylotypes constituted 342 TANIGUCHI et al.

Table 2. The closest relative of 16S rRNA gene in Kure Port from July 2005 to June 2006

Phylogenetic Total community BrdU-incorprated community KR band no. Closest relative group JASONDJ FMAMJ JASONDJMAMJ Alphaproteobacteria 39*, 168(*) Roseobacter sp. LE17 Rhodobacterales ++++++++++++ +++++++++++ 41* Rhodobacteraceae bacterium Rhodobacterales ++++++++++++ +++++++++++ DG1250 15*, 100(*) Roseobacter sp. CSQ2 Rhodobacterales ++++++−+−−−+ +++++++++++ 159(*) Sulfitobacter sp. 7PSW7 Rhodobacterales −++−−−−+++++ +++−−−−++++ 2*, 152(*) Roseobacter sp. SYOP1 Rhodobacterales −++−−−++−+++ +++−−−−++−+ 37*, 62* marine alpha proteobacterium Rhodobacterales −−−−−−++++++ +−−−−−+++++ AS21 4*, 14* Loktanella sp. J57 Rhodobacterales +−+−++−−−−−− +−+−++−−−−− 24* Roseobacter sp. CSQ2 Rhodobacterales −−−+++−−−−−− −−−+++−−−−− 31* Roseobacter sp. TM1038 Rhodobacterales −−+−−−−+−−−− −−+−++−−−−− Gammaproteobacteria 23* Alteromonas sp. S11B8 Alteromonadales +−−−+++−−+++ +++++++−−++ 101(*), 124(*) Marine gamma proteobacterium OM60 clade ++++++++++++ −−−−+++−+−+ HTCC2149 Bacteroidetes 179*, 184(*) bacterium ++++++++++++ +++++++++++ G512M1 72(*) Coccinimonas marina strain Flavobacteriales ++++++++++−+ +++++++−−−+ IMCC1846 12*, 105(*) Lewinella nigricans Sphingobacteriales −−+++++−−−−− −−+++++−−−− 58*, 85(*) Gelidibacter algens Flavobacteriales ++−−−−−−−−−− ++−−−−−−−++ 28* Brumimicrobium mesophilum Flavobacteriales +++++++−++++ −+−−−+−−−−− strain YH207 177* Gilvibacter sediminis Flavobacteriales −+−−−−−−−−−− ++−−−−−−−−− 90(*) Balneola alkaliphila strain Sphingobacteriales −+−−−−−+−−−− −+−−−−−−−−− CM41_14b 48* Fluviicola taffensis strain Flavobacteriales −−−−−−−−−−+− −−−−−−−−−+− RW262 Actinobacteria 16*, 27*, 103(*), 128(*) Acidimicrobidae bacterium Acidimicrobiales +−+++++−−−−− −−−++++−−−− Ellin7143 109(*) Bacterium Ellin5273 Acidimicrobiales −+++++−+−−−− −+++−−−−−−− 6* Salinibacterium aquaticus Acidimicrobiales +−−−−−−−−−−+ +−−−−−−−−−+ strain CW1 46*, 150(*) ‘Candidatus Microthrix ‘Candidatus −−−−−−−+++++ −−−−−−−−+−− calida’ strain TNO24 Microthrix’ Cyanobacteria 81, 127 Synechococcus sp. WH 8016 Chroococcales +++++++−−+++ −−−−−−−−−−− Plastid 122 Dinophysis caudata strain Dinophysiales, −−−−−++−−−−− −−−−−−−−−−− DCLOHABE01 plastid *: excised band from BrdU fraction. (*): excised band from total fraction, but presence in BrdU fraction.

23 of 39 phylotypes, including 9 alphaproteobacteria, 2 the year in BrdU-incorporated communities. Ten phylotypes gammaproteobacteria, 8 Bacteroidetes bacteria, and 4 (KRs-4/14, KR-24 and KR-31 of Alphaproteobacteria, KR- actinobacteria (Fig. 4). All alphaproteobacteria belonged 28, KR-48, KRs-58/85, KR-90, and KR-177 of Bacteroidetes to Rhodobacterales, Roseobacter group; 1 of 2 gamma- bacteria, and KR-6 and KR-109 of Actinobacteria) appeared proteobacteria belonged to Alteromonadales, and the other in the months when the water temperature was greater than belonged to the OM60 clade; 3 of 4 actinobacteria belonged 15°C. to Acidimicrobiales. The migration position of the DGGE band was compared Discussion among all samples, and the presence or absence of the 39 phylotypes was determined for each month (Table 2). The BrdU approach allows us to determine not only Although phylotypes of the Roseobacter group were most phylogenetic affiliations but also the contributions of partic- predominant in BrdU-incorporated communities, the distri- ular taxa to bacterial productivity at a given time and location. bution patterns, such as presence/absence and intensity, of In our previous studies, BUMP-DGGE analyses revealed each DGGE band were different. Four phylotypes (KRs-15/ phylogenetic affiliations of actively growing bacteria in 100, KRs-39/168, and KR-41 of Alphaproteobacteria, and coastal (17) and oceanic (48) environments; however, these KRs-179/184 of Bacteroidetes bacteria) appeared throughout previous studies showed spatial and temporal snapshots of Seasonal Variations in Actively Growing Bacteria 343

Fig. 4. Neighbor-joining tree of 16S rRNA gene partial sequences of excised bands and their relatives; data from July 2005 to June 2006 at the pier of Kure Port. Sequences determined in this study are shown in bold. The bands found in lanes of BrdU-incorporated communities are indicated by asterisks; asterisks in parentheses show bands that were excised from lanes of total communities but were also present in those of BrdU-incorporated communities. Bootstrap values of >50% are indicated by branches. Scale bar represents 5% estimated sequence divergence. Aquifex aeolicus is used as the outgroup. the situation and did not indicate temporal variations in the showed a clear seasonal pattern; however, water temperatures affiliations of actively growing bacteria. Environmental do not directly affect the community structure. Phytoplankton conditions, especially in temperate coastal environments, blooms are an opportunity for marked changes in the show large seasonal fluctuations, mainly due to changes in community structures of actively growing bacteria (PER- water temperature and organic matter supply from phy- MANOVA, P<0.01). Many studies have also reported toplankton; therefore, determining the seasonal variability of seasonal variation in bacterial community structures, and that actively growing bacteria is important for understanding community structures would be affected by environmental biogeochemical processes in these areas. factors, such as water temperature, nutrients, and chl a (e.g., In this study, we revealed the seasonal variations of actively 2, 22, 27, 29, 35). Schauer et al. (42) showed that a growing bacteria in neritic coastal waters. To simulate their phytoplankton bloom could rapidly change the bacterial growth in situ as closely as possible, we added concentrations community structure. Additionally, in freshwater environ- of BrdU (1 µM) for labeling, which were much lower than ments, it was suggested that the biochemical composition of the ambient concentration of dissolved organic matter DOM is one of the factors that shifts in bacterial community (DOM), i.e., typically higher than 40 µM (30); we also structures (11). Thus, the qualitative and quantitative char- shortened the time (3 h) for BrdU incubation relative to our acteristics of the organic matter supply (e.g., from an intensive previous studies (approximately 5–12 h). phytoplankton bloom) would be a more important trigger for The community structures of actively growing bacteria changes in community structures of actively growing bacteria 344 TANIGUCHI et al. than factors such as water temperature; however, the water detect any phylotypes belonging to the SAR11 group in our mass exchanged may also affect changes in the community study. In previous studies in the western North Pacific (48) structures because environmental factors such as salinity and and the Inland Sea of Japan (17), which is adjacent to our bacterial abundance also changed from August to September study area, we also did not find actively growing SAR11 2005 and January to February 2006 (Table 1). group bacteria. Tada et al. (45) have also indicated that the The dominant phylogenetic groups in actively growing contribution of SAR11 group bacteria to total bacterial bacterial communities throughout the year were the productivity is low throughout the year at the same study Roseobacter group and Bacteroidetes (Table 2); the propor- site. On the basis of genomic information and cultivation tions of each group relative to the total number of phylo- studies, SAR11 group bacteria have been hypothesized to be types were 38%–86% (mean, 53.8%) and 11.1%–46.2% slow growing and not very active, regardless of their (mean, 26.6%), respectively. These phylotypes were also abundance (15, 37). In addition, Alonso and Pernthaler (1) determined as the dominant groups in our previous studies showed, using MAR-FISH, that SAR11 group bacteria are a (17, 48), suggesting that they should be one of the key groups minority in glucose-incorporating populations. On the other to mediate the flux of organic matter in marine environments. hand, other studies have found higher activities of SAR11 Although major phylogenetic groupings did not change group bacteria using Micro-FISH (20, 21) and rRNA/rRNA seasonally, the community structures revealed by levels of gene ratio (9) approaches. Further studies are required to DGGE band resolution changed markedly (Fig. 3). Marry et better understand the metabolic activities of SAR11 group al. (22) reported a similar phenomenon: the bacterial bacteria. community structure at lower, but not higher, phylogenetic The phylotypes of Actinobacteria were determined as groupings changes markedly with the season. Our results actively growing bacteria in this study, although observed in the community structure change of actively Actinobacteria are generally considered as freshwater taxa growing bacteria were consistent with their findings. The (24). The presence of Actinobacteria in coastal marine function of each actively growing bacterial species may vary, environments has been attributed not to their habit prefer- and different species may utilize different organic substances; ences, but to their transport from continental areas (36); therefore, quantitative analysis of bacterial productivity at however, our results suggest that indigenous species of species level resolution may be required to fully understand Actinobacteria are well adapted to the environmental con- the flux patterns of organic matter. ditions of a coastal marine ecosystem (43). A previous The community structures of both total and actively study also reported clones belonging to clusters of marine growing bacteria changed throughout the year; however, Actinobacteria the North Atlantic Ocean, and a significant structures in July 2005 closely resembled those in June 2006 seasonal increase in the relative rRNA (27). Additionally, we (Fig. 3), which was also supported by the higher values of have reported that phylotypes of Actinobacteria are actively the Jaccard coefficient at these two times. Previous studies growing in the western North Pacific (48). Thus, the dynamics have also indicated seasonal patterns of variation in bacterial and ecological roles of Actinobacteria in marine environ- community structures (12, 51). Additionally, Fuhrman et al. ments would appear to warrant further investigation. (14) have statistically shown the seasonal repeatability of In conclusion, our results show that community structures bacterial community structures. They suggest that patterns of actively growing bacteria shift markedly in the wake of of variability in several bacterial species are highly predict- phytoplankton blooms in the neritic waters of Hiroshima Bay. able on the basis of various environmental factors, such as Our results also support the idea that Roseobacter and water temperature, oxygen, and dissolved nitrite. Community Bacteroidetes groups are the primary actively growing structures of actively growing bacteria found in this study bacterial taxa responsible for organic matter fluxes attributed might show seasonal repeatability because bacterial growth to bacteria, as previously reported. Future studies should might be characterized by seasonally recurring patterns of focus on individual phylotypes, and the specific functions environmental parameters. possessed by these actively growing bacteria should be Three phylotypes of Alphaproteobacteria (KR-39, KR-41 investigated with in situ methods (31). For example, the and KR-15) and 1 phylotype of Bacteroidetes bacteria (KR- combination of the BrdU technique and metagenomic analysis 179) were identified as active bacteria year-round (Table 2). can be very insightful, as in the study by Mou et al. (28). The constant expression of their activities suggests that these Revealing the specific functions and phylogenetic affiliations phylotypes are versatile in adapting to changing conditions of actively growing bacteria, both qualitatively and quanti- in coastal environments, and utilizing a variety of resources tatively, is fundamentally important to understanding bio- to support growth. Many members of the Roseobacter group geochemical processes in the marine ecosystem. are able to utilize lignin-related aromatic compounds, which are significant components of the carbon pool in the coastal Acknowledgements marine environment (8, 25). Additionally, Bacteroidetes bacteria are reported to be especially proficient in degrading We would like to thank Shin-ichi Uye of Hiroshima University, high-molecular weight organic materials, such as cellulose, and the officers and crew of the R/V Toyoshio-maru for their support protein and chitin (10, 19). Since these organic materials are in collecting samples. We are grateful to Takeshi Miki, Institute of Oceanography, National Taiwan University, for valuable sugges- abundant in eutrophic areas, Bacteroidetes bacteria should tions and help with statistical analyses. We are also grateful to the be one of the key groups in the biogeochemical process. Associate Editor and two anonymous reviewers for their construc- The SAR11 group of bacteria is one of the most abundant tive criticism. This research was supported by grants-in-aid (Nos. groups in marine environments (26); however, we did not 18310011, 18201003, 19030006, 21200029 and 21014005) from Seasonal Variations in Actively Growing Bacteria 345 the Japan Society for the Promotion of Science (JSPS), and JSPS 22. Mary, I., D.G. Cummings, I.C. Biegala, P.H. Burkill, S.D. Archer, Research Fellowships for Young Scientists (No. 05J05361). and M.V. Zubkov. 2006. Seasonal dynamics of bacterioplankton community structure at a coastal station in the western English Channel. Aquat. Microb. Ecol. 42:119–126. References 23. Meon, B., and D.L. Kirchman. 2001. Dynamics and molecular composition of dissolved organic material during experimental 1. Alonso, C., and J. Pernthaler. 2006. Roseobacter and SAR11 domi- phytoplankton blooms. Mar. Chem. 75:185–199. nate microbial glucose uptake in coastal North Sea waters. Environ. 24. Methe, B.A., W.D. Hiorns, and J.P. Zehr. 1998. Contrasts between Microbiol. 8:2022–2030. marine and freshwater bacterial community composition: analyses of 2. Alonso-Sáez, L., V. Balagué, and E.L. Sà, et al. 2007. Seasonality communities in Lake George and six other Adirondack lakes. Limnol. bacterial diversity in north-west Mediterranean coastal waters: assess- Oceanogr. 43:368–374. ment through clone libraries, fingerprinting and FISH. FEMS Micro- 25. Moran, M.A., and R.E. Hodson. 1994. Dissolved humic substances biol. Ecol. 60:98–112. of vascular plant origin in a coastal marine environment. Limnol. 3. Alonso-Sáez, L., and J.M. Gasol. 2007. Seasonal variations in the Oceanogr. 39:762–771. contributions of different bacterial groups to the uptake of low- 26. Morris, R.M., M.S. Rappé, S.A. Connon, K.L. Vergin, W.A. Siebold, molecular-weight compounds in northwestern Mediterranean coastal C.A. Carlson, and S.J. Giovannoni. 2002. SAR11 clade dominates waters. Appl. Environ. Microbiol. 73:3528–3535. ocean surface bacterioplankton communities. Nature 420:806–810. 4. Altschul, S.F., T.L. Madden, A.A. Schäffer, J. Zhang, Z. Zhang, W. 27. Morris, R.M., K.L. Vergin, J.C. Cho, M.S. Rappé, C.A Carlson, Miller, and D.J. Lipman. 1997. Gapped BLAST and PSI-BLAST: a and S.J. Giovannoni. 2005. Temporal and spatial response of bac- new generation of protein database search programs. Nucleic Acids terioplankton lineages to annual convective overturn at the Bermuda Res. 25:3389–3402. Atlantic time-series study site. Limnol. Oceanogr. 50:1687–1696. 5. Anderson, M.J. 2001. A new method for non-parametric multivariate 28. Mou, X., S. Sun, R.A. Edwards, R.E. Hodson, and M.A. Moran. analysis of variance. Austral Ecol. 26:32–46. 2008. Bacterial carbon processing by generalist species in the coastal 6. Artursson, V., and J.K. Jansson. 2003. Use of bromodeoxyuridine ocean. Nature 451:708–711. immunocapture to identify active bacteria associated with arbuscular 29. Murray, A.E., C.M. Preston, R. Massana, L.T. Taylor, A. Blakis, mycorrhizal hyphage. Appl. Environ. Microbiol. 69:6208–6215. K. Wu, and E.F. DeLong. 1998. Seasonal and spatial variability of 7. Borneman, J. 1999. Culture-independent identification of micro- bacterial and archaeal assemblages in the coastal waters near Anvers organisms that respond to specified stimuli. Appl. Environ. Micro- Island, Antarctica. Appl. Environ. Microbiol. 64:2585–2595. biol. 65:3398–3400. 30. Ogawa, H., and E. Tanoue. 2003. Dissolved organic matter in the 8. Buchan, A., J.M. Gonzalez, and M.A. Moran. 2005. Overview of the oceanic waters. J. Oceanogr. 59:129–147. Marine Roseobacter lineage. Appl. Environ. Microbiol. 71:5665–5677. 31. Okabe, S., M. Oshiki, Y. Kamagata, et al. 2010. A great leap forward 9. Campbell, B.J., L. Yu, T.R.A. Straza, and D.L. Kirchman. 2009. in microbial ecology. Microbes Environ. 25:230–240. Temporal changes in bacterial rRNA and rRNA genes in Delaware 32. Oksanen, J., F. Blanchest, R. Kindt, P. Legendre, R.G. O’Hara, G. (USA) coastal waters. Aquat. Microb. Ecol. 57:123–135. Simpson, P. Solymos, M.H. Stevens, and H. Wagner. 2010. Vegan: 10. Cottrell, M.T., and D.L. Kirchman. 2000. Natural assemblages of community ecology package. http://vegan.r-forge.r-project.org/. marine proteobacteria and members of the Cytophaga–Flavobacter 33. Pernthaler, A., J. Pernthaler, M. Schattenhofer, and R. Amann. 2002. cluster consuming low- and high-molecular-weight dissolved organic Identification of DNA-synthesizing bacterial cells in coastal North matter. Appl. Environ. Microbiol. 66:1692–1697. Sea plankton. Appl. Environ. Microbiol. 68:5728–5736. 11. Crump, B.C., G.W. Kling, M. Bahr, and J.E. Hobbie. 2003. Bacterial 34. Petnthalter, J. 2006. Predation on prokaryotes in the water column community shifts in an Arctic lake correlate with seasonal changes in and its ecological implications. Nat. Rev. Microbiol. 3:537–546. organic matter source. Appl. Environ. Microbiol. 69:2253–2268. 35. Pinhassi, J., L. Gómez-Consarnau, L. Alonso-Sáez, M.M. Sala, M. 12. Crump, B.C., and J.E. Hobbie. 2005. Synchrony and seasonality in Vidal, C. Pedrós-Alió, and J.M. Gasol. 2006. Seasonal changes in bacterioplankton communities of two temperate rivers. Limnol. bacterioplankton nutrient limitation and their effects on bacterial Oceanogr. 50:1718–1729. community composition in the NW Mediterranean Sea. Aquat. 13. Fandino, L.B., L. Riemann, G.F. Steward, R.A. Long, and F. Azam. Microb. Ecol. 44:241–252. 2001. Variations in bacterial community structure during a 36. Rappé, M.S., K. Vergin, and S.J. Giovannoni. 2000. Phylogenetic dinoflagellate bloom analyzed by DGGE and 16S rDNA sequencing. comparisons of a coastal bacterioplankton community with its Aquat. Microb. Ecol. 23:119–130. counterparts in open ocean and freshwater systems. FEMS Microbiol. 14. Fuhrman, J.A., I. Hewson, M.S. Schwalbach, J.A. Steele, M.V. Ecol. 33:219–232. Brown, and S. Naeem. 2006. Annually reoccurring bacterial commu- 37. Rappé, M.S., S.A. Connon, K.L. Vergin, and S.J. Giovannoni. 2002. nities are predictable from ocean conditions. Proc. Natl. Acad. Sci. Cultivation of the ubiquitous SAR11 marine bacterioplankton clade. USA 35:13104–13109. Nature 418:630–633. 15. Giovannoni, S.J., H.J. Tripp, S. Givan, et al. 2005. Genome stream- 38. Riemann, L., G.F. Steward, and F. Azam. 2000. Dynamics of lining in a cosmopolitan oceanic bacterium. Science 309:1242–1245. bacterial community composition and activity during a mesocosm 16. Hamasaki, K., R.A. Long, and F. Azam. 2004. Individual cell growth diatom bloom. Appl. Environ. Microbiol. 66:578–587. rates of marine bacteria, measured by bromodeoxyuridine incorpora- 39. Sala, M.M., M. Estrada, and J.M. Gasol. 2006. Seasonal changes in the tion. Aquat. Microb. Ecol. 35:217–227. functional diversity of bacterioplankton in contrasting coastal environ- 17. Hamasaki, K., A. Taniguchi, Y. Tada, R.A. Long, and F. Azam. 2007. ments of the NW Mediterranean. Aquat. Microb. Ecol. 44:1–9. Actively growing bacteria in the Inland Sea of Japan, identified by 40. Sapp, M., A. Wichels, K.H. Wiltshire, and G. Gerdts. 2007. Bacterial combined bromodeoxyuridine immunocapture and denaturing gradi- community dynamics during the winter–spring transition in the North ent gel electrophoresis. Appl. Environ. Microbiol. 73:2787–2798. Sea. FEMS Microbiol. Ecol. 59:622–637. 18. Huber, T., G. Faulkner, and P. Hugenholtz. 2004. Bellerophon; a 41. Schäfer, H., and G. Muyzer. 2001. Denaturing gradient gel electro- program to detect chimeric sequences in multiple sequence align- phoresis in marine microbial ecology, p. 425–468. In J.H. Paul (ed.), ments. Bioinformatics 20:2317–2319. Methods in Microbiology, vol. 30. Academic Press, New York, USA. 19. Kirchman, D.L. 2002. The ecology of Cytophaga–Flavobacteria in 42. Schauer, M., V. Balagué, C. Pedrós-Alió, and R. Massana. 2003. aquatic environments. FEMS Microbiol. Ecol. 39:91–100 Seasonal changes in the taxonomic composition of bacterioplankton 20. Malmstrom, R.R., R.P. Kiene, and D.L. Kirchman. 2004. Identifica- in a coastal oligotrophic system. Aquat. Microb. Ecol. 31:163–174. tion and enumeration of bacterial assimilating dimethylsulfoniopropi- 43. Stevens, H., T. Brinkhoff, B. Rink, J. Vollmers, and M. Simon. 2007. onate (DMSP) in the North Atlantic and Gulf of Mexico. Limnol. Diversity and abundance of Gram positive bacteria in a tidal flat Oceanogr. 49:597–606. ecosystem. Environ. Microbiol. 9:1810–1822. 21. Malmstrom, R.R., M.T. Cottrell, H. Elifantz, and D.L. Kirchman. 44. Steward, G.F., and F. Azam. 1999. Bromodeoxyuridine as an alterna- 2005. Biomass production and assimilation of dissolved organic tive to 3H-thymidine for measuring bacterial productivity in aquatic matter by SAR11 bacteria in the Northwest Atlantic Ocean. Appl. samples. Aquat. Microb. Ecol. 19:57–66. Environ. Microbiol. 71:2979–2986. 346 TANIGUCHI et al.

45. Tada, Y., A. Taniguchi, and K. Hamasaki. 2009. Phylotype-specific 48. Taniguchi, A., and K. Hamasaki. 2008. Community structures of productivity of marine bacterial population in eutrophic seawater, as actively growing bacteria shift along a north–south transect in the revealed by bromodeoxyuridine immunocytochemistry combined with western North Pacific. Environ. Microbiol. 10:1007–1017. fluorescence in situ hybridization. Microbes Environ. 24:315–321. 49. Thingstad, T.F., and R. Lingnell. 1997. Theoretical models for the 46. Tada, Y., A. Taniguchi, and K. Hamasaki. 2010. Phylotype-specific control of bacterial growth rate, abundance, diversity and carbon growth rates of marine bacteria measured by bromodeoxyuridine demand. Aquat. Microb. Ecol. 13:19–27. immunocytochemistry and fluorescence in situ hybridization. Aquat. 50. Urbach, E., K.L. Vergin, and S.J. Giovannoni. 1999. Immunological Microb. Ecol. 59:229–238. detection and isolation of DNA from metabolically active and 47. Tamura, K., J. Dudley, M. Nei, and S. Kumar. 2007. MEGA4: bacteria. Appl. Environ. Microbiol. 65:1207–1213. Molecular Evolutionary Genetics Analysis (MEGA) software version 51. Van der Gucht, K., K. Sabbe, L. De Meester, N. Vloemans, G. Zwart, 4.0. Mol. Biol. Evol. 24:1596–1599. M. Gillis, and W. Vyverman. 2001. Contrasting bacterioplankton community composition and seasonal dynamics in two neighbouring hypertrophic freshwater lakes. Environ. Microbiol. 3:680–690.