Warming effects on dynamics of microbial communities in coastal waters of temperate and subtropical zones through dilution and fractionation experiments

Warming effect on bottom-up or top-down control by grazing To determine whether the positive effect of warming on the abundance and growth rates of prokaryotes in natural community coul

This document is downloaded at: 2017-08-14T04:42:23Z

Title

Warming effects on dynamics of microbial communities incoastal waters of temperate and subtropical zones throughdilution and fractionation experiments

Warming effects on dynamics of microbial communities in coastal waters of temperate and subtropical zones through

dilution and fractionation experiments

Dao Thi Anh Tuyet

Graduate School of Science and Technology, Educational Division

Department of Environment and Energy Systems

Shizuoka University

December 2014

Warming effects on dynamics of microbial communities in coastal waters of temperate and subtropical zones through

dilution and fractionation experiments

2014 年 12 月

i

CONTENTS

Chapter 1

General introduction 1

Introduction 1

Methodological approaches 3

References for chapter 1 6

Chapter 2 Effects of warming on microbial communities in coastal waters of temperate and subtropical zones in Northern Hemisphere with a focus on Gammaproteobacteria 10

2.1 Introduction 10

2.2 Materials and methods 13

2.2.1 Study sites and sample collection 13

2.2.2 Experimental processing 14

2.2.3 Total direct counts (TDCs) of prokaryotic abundance 15

2.2.4 Fluorescence in situ hybridization (FISH) 15

2.2.5 Growth rates 16

2.2.6 Enumeration of protists 17

2.2.7 Analysis of bacterial phylotypic community composition 17

2.3 Results 20

2.3.1 Environmental parameters 20

2.3.2 Warming effect on prokaryotes 21

2.3.3 Warming effect on bottom-up or top-down control by grazing 22

2.3.4 Warming effects on Gammaproteobacteria 24

2.4 Discussions 26

Acknowledgements 30

References for chapter 2 31

ii

Figures chapter 2 37

Table chapter 2 43

Supplementary chapter 2 46

Chapter 3 Warming effect on the constituents of microbial communities 49

3.1 Introduction 49

3.2 Materials and methods 51

3.2.1 Sampling collection and experimental processing 51

3.2.2 Analysis of bacterial phylogenetic composition 51

3.2.3 Nucleotide sequence accession numbers 53

3.3 Results and discussion 54

3.3.1 Constituents of the bacterial communities in Suruga Bay 54

3.3.2 Constituents of the bacterial community in Ha Long Bay 56

References for chapter 3 58

Figure chapter 3 60

Conclusions and further works 69

Conclusions 69

Further works 70

Acknowledgements 71

as climatic models predict that sea temperature will continue to increase with 2- 4.8 °C during the 21st century (IPCC 2001, 2013) This warming may affect different aspects of the function and structure of marine ecosystems (Edwards and Richardson 2004, Montoya and Raffaelli 2010) particularly microbial food web

Among the marine environment, prokaryotes and viruses present as the smallest constituents of ecosystem and to be considered to respond swiftly to the environmental change such as temperature Preceding studies indicated a positive correlation between temperature and bacterial abundance (Write and

Coffin 1983, Lomas et al 2002) or growth rate (White et al.1991, Shiah &

Ducklow 1994) However, controling mechanism of bacterial community through warming as bottom-up or top-down control by grazer and viral infection in a given

community was not elucidated clearly Sjöstedt et al (2012) showed that a

bacterial community shifted its constituents responding to the changes in temperature To show some aspects of microbial food web under the effect of

warming, Sarmento et al (2010) summarized related studies and indicated that

warming increases bacterial respiration and bacterial losses due to their grazers and bacterial production if nutrients are available These aspects of responses in

microbial production may affect fisheries by changing the microbial loop and primary production which supply food resources for aquatic organisms such as fish, crustacean, mollusk and so on

Climate change provides many extreme climatic events which may influence different aspects of ecosystem and environmental health Another necessity of microbial study related to environmental change stems from the fact that microbial community includes pathogens which may bring problems to aquatic organism and environmental health in coastal waters A pioneering study by Colwell (1996) clearly showed warming in surface temperature increased the incidence of cholera in Bangladesh A recent retrospective analysis suggested that sea surface temperature warming is favoring the spread of vibrios (Vezzulli

et al 2012, 2013) which has different response to environmental factors than

other heterotrophic bacterial population (Simidu et al 1987)

Various studies have conducted until today on the warming effect on microbes, however, most study sites belonged to temperate (Shiah and Ducklow

1994, Apple et al 2006, Bouvy et al 2011, Sjöstedt et al 2012) or subarctic and Arctic regions (Lara et al 2013, Børsheim and Drinkwater 2014) Moreover, as

the strongest ocean warming is predicted for the surface layer in tropical and subtropical regions of Northern Hemisphere (IPCC 2013), which might lead different effect on microbes in subtropical water compared to that in temperate zone

Coastal area is regularly exposed to many anthropogenic influences such as fishery or aquaculture activities, transportation, tourism, harbor and nutrients supply Two studied sites were selected in this study belonging to temperate (Suruga Bay, Japan) and subtropical (Ha Long Bay, Viet Nam) coastal zones These two sites are completely different in climatic condition which may lead to

different responses of microbial communities against changing in temperature

As microbial ecological studies had been conducted in Suruga Bay (Takenaka et

al 2007; Hao et al 2010) which provide sufficient information on the

environmental control of prokaryotes, present study aimed to reveal warming effect on prokaryotes in this environment, which strengthen the understanding of findings supported with previous data, and become a control study for subtropical microbial ecology under the warming environment Bottom-up and top-down controls on prokaryotes under warming effect were also considered in complex interactions of a given community (chapter 2)

As a response of potential pathogens to warming is considered to be major

concern, the present study focused on Gammaproteobacteria which include

various pathogenic bacteria (Brown and Volker 2004) In addition, bacterial 16S rRNA gene was employed to evaluate how the constituents of microbial community change in response to sea temperature rise (chapter 3) as an important issue, which has not yet reported from previous study

Methodological approaches

Microbes in coastal ecosystem are controlled by the availability of substrate supply (bottom-up control) and the mortality caused by grazers and/ or viral infections (top-down control) In order to elucidate the effect of warming on microbial communities, dilution and size fractionation methods were very well utilized (Landry and Hassett 1982, Wright and Coffin 1984, Rassoulzadegan and Sheldon 1986, Šolić and Krstulović 1994, Yokohama et al 2005) Top-down control is regularly determined by two methods: 1) comparing the values obtained with labeled bacteria as food tracers for natural assemblages as

described in Sherr et al (1986), Wikner and Hagstrom (1988); 2) using size

fraction method to estimate the difference in bacterial number/ growth between ungrazed (<1µm) and grazed (>1µm) as reported by Wright and Coffin (1984), Coffin and Sharp (1987), Kuuppo-Leinikki (1990), Šolić and Krstulović (1994) Though the fractionation method tends to give somewhat lower grazing impact than the estimation from the experiment with labeled bacteria (Kuuppo-Leinikki

1990, Šolić and Krstulović 1994), the results of top-down control showed a similar trend regardless of the method employed (Šolić and Krstulović 1994) In present study, top-down control (by grazers) was estimated from difference in bacterial number/ growth between ungrazed (<1µm) and grazed (>1µm) samples as referred by Wright and Coffin (1984), Coffin and Sharp (1987), Šolić and Krstulović (1994)

Similarly to top-down control, bottom-up control on prokaryote is estimated from the differences in abundance and/ or growth rate of prokaryote between control dilution samples Dilution means sufficient nutrients as nitrogen and phosphate involved in marine water are given to bacteria

Small size fraction of sea water consists bacteria, their grazers and viruses Not only protozoa, but larger grazers such as copepods exist in the community However, the most important grazers of bacteria controlling bacterial abundance was defined as small heterotrophic nanoflagellates/ grazer with theirs own size less than 8 µm (Solic and Krstilovic 1994) Thus, in this study, pre-filtrate through larger pore-size filters of 1 µm has not been conducted

Observations from previous studies conducted in coastal water of Suruga Bay and Ha Long Bay indicated that annual seawater temperature varied about 14.6-16.4oC (Takenaka et al 2007, Hao et al 2010) and 12oC (Faxneld et al

2011), respectively Otherwise, climatic models predict that sea temperature will continue to increase with 2 - 4.8°C during the 21st century (IPCC 2001, 2013)

Consequently, warming by 3 or 5°C has been conducted in different seasons to estimate the impact of warming on microbial community in present study

References for chapter 1

1 Apple JK, PA del Giorgi, WM Kemp (2006) Temperature regulation of bacterial production, respiration, and growth efficiency in a temperate salt-marsh estuary Aquat Microb Ecol 43(3): 243-254

2 Børsheim KY, KF Drinkwater (2014) Different temperature adaptation in Arctic and Atlantic heterotrophic bacteria in the Barents Sea Polar Front region Journal of Marine Systems 130:160-166

3 Bouvy M, Y Bettarel, C Bouvier, I Domaizon, S Jacquet, E Le Floc'h, H Montanie, B Mostajir, T Sime-Ngando, JP Torreton, F Vidussi, T Bouvier (2011) Trophic interactions between viruses, bacteria and nanoflagellates under various nutrient conditions and simulated climate change Environ Microbiol 13(7):1842–

10 Hao DM, T Tashiro, M Kato, R Sohrin, T Ishibashi, C Katsuyama, K Nagaosa,

H Kimura, TD Thanh, K Kato (2010) Population dynamics of Crenarchaeota and Euryarchaeota in the mixing front of river and marine waters Microbes and Environments 25(2):126–132

11 IPCC (2001) Climate Change 2001: The scientific basis IPCC Third Assessment Report, 2001 Contribution of Working group I to the third assessment report of the Intergovernmental Panel on Climate Change [Houghton JT, Y Ding, DJ Griggs, M Noguer, PJ van der Linden, D Xiaosu (eds)] Cambridge University Press, Cambridge, United Kingdom

12 IPCC (2013) Summary for Policymakers In: Climate Change 2013: The Physical Science Basis Contribution of Working Group I to the Fifth Assessment Report of the Intergovernmental Panel on Climate Change [Stocker TF, D Qin, G.-K Plattner, M Tignor, SK Allen, J Boschung, A Nauels, Y Xia, V Bex and PM Midgley (eds)] Cambridge University Press, Cambridge, United Kingdom and New York, NY, USA

13 Landry MR, J Kirshtein, J Constantinou (1995) A refined dilution technique for measuring the community grazing impact of microzooplankton, with experimental tests in the central equatorial Pacific Mar Ecol Prog Ser 120:53-63

14 Lara E, JM Arrieta, I Garcia-Zarandona, JA Boras, CM Duarte, SAgustí, PF Wassmann, D Vaqué (2013) Experimental evaluation of the warming effect on viral, bacterial and protistan communities in two contrasting Arctic systems Aquat Microb Ecol 70:17–32

15 Lomas MW, P Glibert, F Shiah, EM Smith (2002) Microbial processes and temperature in Chesapeake Bay: Current relationships and potential impacts of regional warming Global Change Biology 8:51-70

16 Montoya JM, D Raffaelli (2010) Climate change, biotic interactions and ecosystem services Phil Trans R Soc B 365:2013-2018

17 Rassoulzadegan F, RW Sheldon (1986) Predator-prey interactions of nanozooplankton and bacteria in an oligotrophic marine environment Limnol

169-180

20 Shiah F, HW Ducklow (1994) Temperature and substrate regulation of bacterial abundance, production and specific growth rate in Chesapeake Bay, USA Mar Ecol Prog Ser 103:297-308

21 Simidu U, K Kogure, K Tsukamoto (1987) Distribution of Vibrionaceae in the

sea Nihon Biseibutsu Seitai Gakkaiho 1:65-74

22 Sjöstedt J, Å Hagström, UL Zweifel (2012) Variation in cell volume and community composition of bacteria in response to temperature Aquatic Microbial Ecology 66(3): 237–246

23 Šolić M, N Krstulović (1994) Role of predation in controllong bacterial and heterotrophic nanoflagellate standing stocks in the coastal Adriatic Sea: seasonal patterns Mar Ecol Prog Ser 114:219-235

24 Takenaka T, T Tashiro, A Ozaki, H Takakubo, Y Yamamoto, T Maruyama, K Nagaosa, H Kimura, K Kato (2007) Planktonic bacterial population dynamics

with environmental changes in coastal areas of Suruga Bay Microbes and Environments 22(3):257–267

25 Vezzulli L, I Brettar, E Pezzati, PC Reid, RR Colwell, MG Hofle, C Pruzzo (2012) Long-term effects of ocean warming on the prokaryotic community: evidence from the Vibrios ISME J 6(1):21–30

26 Vezzulli L, RR Colwell, C Pruzzo (2013) Ocean warming and pread of pathogenic Vibrios in the aquatic environment Microb Ecol 65:817–825

27 White PA, J Kalff, JB Rasmussen, JM Gasol (1991) The effect of temperature and algal biomass on bacterial production and specific growth rate in freshwater and marine habitats Microb Ecol 21:99-118

28 Wikner J, A Hagstrom (1988) Evidence of a tightly coupled nanoplanktonic predator-prey link regulating the bacterivores in the marine environment Mar Ecol Prog Ser 50: 137-145

29 Wright RT, Coffin RB (1983) Planktonic bacteria in estuaries and coastal waters of northern Massachusetts: spatial and temporal distribution Mar Ecol Prog Ser 11:205-216

30 Wright RT, RB Coffin (1984) Measuring microzooplankton grazing on planktonic marine bacteria by its impact on bacterial production Microb Ecol 10: 137-149

31 Yokokawa T, T Nagata (2005) Growth and grazing mortality rates of phylogenetic groups of bacterioplankton in coastal marine environments Appl Environ Microbiol 71(11):6799–6807

Environmental warming alters food-web structures and ecosystem

functions (Petchey et al 1999) Burrows et al (2011) suggested that during the

last 50 years of thermal variations, the rate of shift of ecosystem constituents was faster in the oceans than on land at some latitudes (in sub-Arctic zones and within 15o of the equator) and that changes in seasonal timing were generally greater in the ocean owing to smaller seasonal thermal variationscompared to these variations on land In addition to these global implications, phenomena that occur within individual ecosystems caused by climate change need to be more thoroughly studied to understand the possible impacts of global warming

Among the constituents of a coastal ecosystem, microbes are considered to respond quickly to changes in temperature because of their small cell sizes (i.e large surface/volume ratios) and potentially rapid growth rates

Daufresne et al (2009) showed that the proportion of small-sized species in a

given population increased as a result of global warming by meta-analysis of data obtained from long-term surveys and experimental analyses Other studies indicated a positive correlation between temperature and bacterial abundance

(Write and Coffin 1983, Lomas et al 2002) and growth rate (White et al 1991,

Shiah & Ducklow 1994) However, effect of temperature on the physiology of microbes does not simply explain the effect of environmental warming on the

microbial community Elucidating whether “top-down” (mortality) or “bottom-up” (increased rate of production) control strongly affects the microbial community is

an approach to understanding the mechanism of the effect of warming on the organisms Several studies have been conducted with this approach Šolić et al

(2009) found that, under in situ conditions, bottom-up control dominated during

colder period whereas top-down control dominated during warmer periods

Sjöstedt et al (2012) showed that a microbial community shifted its composition

in response to changes in temperature Although the evidence obtained from

controlled experiments and observations is not always consistent, Sarmento et

al (2010) summarized related studies and indicated that warming increases

bacterial respiration and bacterial losses due to their grazers, and increases bacterial production if nutrients are available These aspects of responses of microbial production and the resulting carbon flux may affect higher trophic levels including fishery resources Fish culture is an important component of fisheries and is conducted in coastal environments Thus we focus on coastal environments in the present study

With regard to bacteria, the effects of global warming on the distribution and growth of pathogens must be particularly emphasized A pioneering study by Colwell (1996) showed clearly that a three-degree increase in surface temperature increased the incidence of cholera in Bangladesh A recent retrospective analysis suggested that sea surface temperature warming is

favoring the spread of vibrios (Vezzulli et al 2012, 2013) Because the response

of potential pathogens to warming is a major concern, we focused on

Gammaproteobacteria which include various pathogenic bacteria (Brown and

Volker 2004) To elucidate effect of warming on the prokaryotic community, we selected two study sites, Suruga Bay in Japan and Ha Long Bay in Viet Nam,

being located in a temperate and a subtropical zone, respectively Warming effect on microbes in subtropical coastal water is not yet well elucidated And both sites are important areas of fishery and fish culture In previous work

conducted in Suruga Bay, the abundance of Bacteria correlated positively with temperature (p<0.05) (Takenaka et al 2007) However, whether warming affects the growth rate of a given prokaryotic community, Gammaproteobacteria in

particular, or the mechanism which determines the effect of warming have not been elucidated In this study, therefore, we designed an experiment to evaluate whether top-down or bottom-up control strongly regulated the effect of warming

on these microbial communities

2.2 Materials and methods

2.2.1 Study sites and sample collection

Water was sampled at the center of Shimizu Port, approximately 1.5 km from the estuary of Tomoe River in Suruga Bay, Japan (35°00′56′′N–138°30′58′′E) and near the three mouths of Troi, Man and Dien Vong Rivers in

Ha Long Bay, Viet Nam (20°55′17′′N–107°05′20′′E) where the water column depths were 26.1 m and 4 m, respectively (Fig 1) The studied sites were expected to be affected by fresh water supply and anthropogenic activities in surrounding urban area Water samples were collected at a depth of 1 m using a 10-L Niskin water sampler (5026-D, Rigosha, Tokyo, Japan) as the surface water was expected to be affected directly by climate change such as temperature, sunshine and precipitation Sampling was conducted seasonally on August 3,

2011 (summer), April 17, 2012 (spring), October 26, 2012 (autumn), and February 11, 2013 (winter) in Suruga Bay and twice on August 8, 2012 (summer, rainy season) and April 6, 2013 (spring, dry season) in Ha Long Bay

Water temperature, pH, salinity and dissolved oxygen (DO), were measured and recorded directly at the sampling sites using water quality checkers (U-52, Horiba, Kyoto, Japan; WQC-24, DKK-TOA, Tokyo, Japan; DO-HQ30d, Hach, Colorado, USA)

The concentrations of chlorophyll a and nutrients (NH4+, NO2 −

Nordersted, Germany), after Takenaka et al (2007) and Hao et al (2010) DOC

was analyzed using a total organic carbon analyzer (TOC-VCSH, Shimadzu)

Samples were immediately placed into sterilized 250-mL and 500-mL

Pyrex bottles (Barloworld Scientific Ltd., Staffordshire, UK) for environmental and microbial analysis, and for incubation experiments All the samples were then transported to the laboratory in less than 2 h on ice

2.2.2 Experimental processing

Size fractionation was performed on the basis of an assumption that prokaryotic size was usually >0.2 µm and protists were mostly trapped by filtration with a 1.0-µm filter Seawater was passed through 1.0 and 0.2-µm Nuclepore filters (Whatman, Cambridge, UK) that retained 84.5% and 1.5% of prokaryotes represented in the initial seawater (defined as 100% of prokaryotic abundance), respectively when we examined seawater collected from Suruga Bay on November 4, 2013

A set of experiments was performed in three different sterilized 500-mL Duran (Mainz, Germany) bottles, with duplicates for each experiment Each bottle was filled with 500 mL of seawater The first was a natural community (control), the second (protist-free) was a filtrate from a 1.0-µm Nuclepore filter (Whatman), and the third (dilution) was approximately equal volumes of non-filtered seawater and filtered seawater passed through a 0.2-µm Nuclepore filter (Whatman) Top-down control by grazing was estimated from the different abundance of prokaryotes in the control and protist-free samples Bottom-up control was estimated from the difference abundance of prokaryotes in control and dilution samples (Supplementary Fig S1) The dilution factor was defined as a ratio of prokaryotic abundance in the control sample to that in the dilution sample

The experiments were started after less than 30 minutes adjusting of temperature for each experimental condition Each experiment was set up in incubators (Unit water bath Thermominder SD, TAITEC, Saitama, Japan) set at

different temperatures: in situ temperature (the temperature of the sampling site)

and temperature 3 or 5 degrees higher than the in situ temperature The incubations were conducted near the sampling sites under the natural light condictions

2.2.3 Total direct counts (TDCs) of prokaryotic abundance

Prokaryotic abundance (TDC) was determined initially, then after 3 (or 4),

6, 12 (or 15), and 24 (or 20) h during incubation experiments, in accordance with the methods of Porter and Feig (1980) A sub-sample was fixed with neutralized formaldehyde (Wako, Osaka, Japan; final concentration 2%, pH 7.4), collected

on a black 0.2-µm Nuclepore filter (Whatman), and then stained with 4′, 6-diamidino-2-phenylindole (DAPI; Nacalai Tesque, Kyoto, Japan; final concentration 0.01µg mL−1) Samples were directly quantified under an epifluorescence microscope (BX51-FLA, Olympus, Tokyo, Japan; Ultraviolet excitation filter U-MWU2) by using 20 microscopic fields which contained more than 400 cells for each sample

2.2.4 Fluorescence in situ hybridization (FISH)

We used rRNA-targeted oligonucleotide probes specific for Bacteria (EUB338) (Amann et al 1990) and Gammaproteobacteria (GAM42a) (Manz et

al 1992) These probes were commercially synthesized and labeled with

fluorescein isothiocyanate (FITC) An unlabeled probe cGAM42a was used as a competitor for GAM42a because of one base pair mismatch between these

probes (Manz et al 1992) (Table 1) Planktonic abundance of Bacteria and

Gammaproteobacteria was assessed using the method described by Takenaka

et al (2007) The samples were fixed with paraformaldehyde (final concentration

3%, pH 7.2) and kept at 4°C for up to 24 h The fixed samples were then filtered through a 0.2-µm Nuclepore filter (Whatman) The microbial cells trapped on the

filter were rinsed three times with filtered phosphate-buffered saline (Dulbecco’s PBS(−); Nissui ,Tokyo, Japan; pH 7.2) and dehydrated with 1 mL of 50%, 80% and 99.5% ethanol for 3 minutes each, then dried at room temperature and stored at –20°C until hybridization Hybridization were performed at 46°C for 90 minutes on filters placed on slides coated with gelatin, hybridization buffer (0.9 M NaCl, 20 mM Tris-HCl [pH 7.4], 0.01% SDS, formamide [20% for EUB338, 35% for GAM42a]) and 5 ng µL−1 of the respective labeled probe Each filter was then washed at 48°C for 15 minutes in pre-warmed washing buffer (NaCl [0.225 M for EUB338, 0.080 M for GAM42a], 20 mM Tris-HCl [pH 7.4], 5 mM EDTA, 0.01% SDS), then rinsed with MilliQ water and air-dried The filter was stained with 0.1

µg mL−1 DAPI on glass slide, for 5 minutes The abundance of Bacteria and

Gammaproteobacteria were estimated as the ratio of hybridized cells to

DAPI-stained cells on the basic of pictures taken by use of a universal epifluorescence microscopic system (BX51-FLA, Olympus) equipped with a digital camera (DP71, Olympus; Ultraviolet excitation filter U-MWU2 and blue excitation filter U-MNIB3) Twenty microscopic fields were examined for each sample with approximately 300-1,400 DAPI-positive cells The number of FITC-positive cells varied, depending upon the relative contribution of each subgroup, and ranged from 30-700 cells for each sample

(cells mL−1) initially and after incubation for t h, respectively

2.2.6 Enumeration of protists

The abundance of protists was determined for samples taken from

coastal water by modifying the method of Lim et al (1993, 1996) Samples were

fixed with neutralized formaldehyde (Wako; final concentration 3.7%) and filtered through 1.0-µm Nuclepore filters (Whatman) within 48 h The cells trapped on the filter were dehydrated and dried at room temperature

Pre-hybridization was performed at 40°C for 45 minutes with 5 × SET buffer (0.75 M NaCl, 100 mM Tris–HCl [pH 7.8], 5 mM EDTA, 0.1%SDS) The three labeled probes EUK 309 (Sogin and Gunderson 1987), EUK 1209

(Giovannoni et al 1988), and EUK 502 (Amann et al 1990) (Table 1) were

added to each filter (final concentration 0.5 ng µL−1 respectively) Hybridization was performed at 40°C for 180 minutes; samples were then washed at 45°C for

10 minutes in pre-warmed 0.2 × SET buffer (30 mM NaCl, 4 mM Tris-HCl [pH 7.8], 0.2 mM EDTA) The filters were air-dried and then stained with 0.1 µg mL−1DAPI

The numbers of protists were quantified directly under an

epifluorescence microscope (BX51-FLA, Olympus; Ultraviolet excitation filter U-MWU2 and blue excitation filter U-MNIB3) Each sample was examined in 20 microscopic fields, for duplicate filters of each sample; Total number of cells was approximately 50-450

2.2.7 Analysis of bacterial phylotypic community composition

DNA collection and extraction

DNA samples were obtained at the sampling sites and on completion of the experiments Different volumes of seawater (0.5 to 2 L) were passed through

0.22-µm Sterivex–GV filter unit (Millipore, Massachusetts, USA), and DNA from trapped organisms was extracted in accordance with the method described by

Somerville et al (1989) Cells were lysed in the filter unit with lysozyme and proteinase K Bulk DNA was extracted with a 25:24:1 (v/v) phenol–chloroform–

isoamyl alcohol at pH 8.0, then concentrated by ethanol precipitation The nucleic acid pellet was suspended in 30 μL of TE buffer (10 mM Tris-HCl, 0.1mM EDTA, pH 8.0)

Cloning and sequencing of bacterial 16S rRNA gene fragments

Extracted DNA was amplified by polymerase chain reaction (PCR), by using KOD-Plus DNA polymerase (Toyobo, Osaka, Japan) and the bacteria-specific primer set 27F (5′-AGAGTTTGATCMTGGCTCAG-3′) and Uni1492R (5′-GGY TACCTTGTTACGACTT-3′) (Lane 1991) PCR products were cloned by use of the Zero Blunt TOPO PCR Cloning Kit (Invitrogen, Paisley, UK) Clone libraries of bacterial 16S rRNA gene fragments of each sample were constructed separately Specific patterns of DNA fragments of different sizes

were digested with restriction endonuclease MspI (Takara Bio, Shiga, Japan)

The sequences of the inserted PCR amplicons from selected recombinant colonies were analyzed by use of a capillary DNA sequencer (CEQ2000XL DNA Analysis System; Beckman Coulter, California, USA) with a bacterial specific primer (27F) Approximately 550 nucleotide lengths were analyzed

Bacterial phylogenetic analysis

Sequences were checked for chimeric artifacts by use of Check-Chimera

sequences were classified into high-order taxonomic units by Classifier, Ribosomal Database Project II, RDP (http://rdp.cme.msu.edu/classifier/

classifier.jsp) on the basis of 16S rRNA sequences (Wang et al 2007)

Sequences were aligned by use of the CLUSTAL W package in the DNA Data

Bank of Japan (DDBJ) (Thompson et al 1994) Clones with homology values

>80% and >97% were classified into phylogenetic groups and operational taxonomic units (OTUs), respectively (Schloss and Handelsman 2005) The closest related sequences to 16S rRNA gene were searched for by using the Basic Local Alignment Search Tool (BLAST, http://blast.ddbj.nig.ac.jp/) (Altschul

et al 1997)

2.3 Results

2.3.1 Environmental parameters

The measured physicochemical parameters at depths of 1 m are summarized in Table 2 The temperature of the water examined in Suruga Bay ranged from 12.1°C to 25.6°C, which was within the temperature ranges

observed in previous studies (Takenaka et al 2007, Hao et al 2010)

In contrast with the temperate coastal environment, the temperature of the examined water of Ha Long Bay, located in a subtropical zone, was 29.9°C in August 2012 and 22.7°C in April 2013

Nitrate and phosphate concentrations in the water from Suruga Bay ranged from 0.88 to 14.85 µmol L−1 and from 0.07 to 0.88 µmol L−1, respectively

These values did not differ from previous observations (Takenaka et al 2007; Hao et al 2010)

Nitrate and phosphate concentrations in the water obtained from Ha Long bay in August 2012 were 2.52 µmol L−1 and 0.20 µmol L−1, respectively

DOC concentration was less than 100 µmol L−1 for the two water samples examined in Suruga Bay, whereas it was five times higher in the water samples examined in Ha Long Bay in August 2012 This very high concentration of DOC might be caused from the particles transported by fresh water as it was rainy

season in Viet Nam

Chlorophyll a concentration in Suruga Bay ranged from 11.74 to 25.70 mg

m−3, except for winter Chlorophyll a concentration in Ha Long Bay was 2.47 and

12.87 mg m−3 Dissolved oxygen (DO) concentrations and pH in Suruga Bay ranged from 9.38 to 12.66 mg O2 L−1 and from 7.59 to 8.38, respectively These values did not differ for the two observations made in Ha Long Bay (DO: 8.11 to 8.19 mg O2 L−1 and pH: 8.01 to 8.19) Salinity in Suruga Bay ranged from 27.0‰

to 30.0‰; the salinity of the two samples obtained from Ha Long Bay was different (August 2012, 24.9‰; April 2013, 35.0‰)

2.3.2 Warming effect on prokaryotes

In situ abundance of prokaryotes did not differ significantly in the water

examined (p<0.05), ranging from 0.80 ± 0.19 × 106 to 0.89 ± 0.19 × 106 cells

mL−1 (n=20) in Suruga Bay (Supplementary Table S1)

For Ha Long Bay these results were 0.59 ± 0.05 × 106 cells mL−1 on August 8, 2012 and 0.31 ± 0.10 × 106 cells mL−1 on April 6, 2013

Figure 2 shows the changes in prokaryotic abundance for all treatments during incubation of the samples collected from Suruga Bay on October 26,

2012 (above) and for samples collected from Ha Long Bay on April 6, 2013 (below) These data include the results obtained from experiments using three treatments: control, dilution, and protist-free (a protist-free experiment was not performed for the sample collected on August 3, 2011 in Suruga Bay) Prokaryotic abundance of diluted samples was multiplied by a dilution factor, ranging from 1.79 to 2.15 (Supplementary Table S1)

For the October experiment with water collected from Suruga Bay, the total number of prokaryotes increased from 0.84 ± 0.03 × 106 to 1.09 ± 0.17 ×

106 cells mL−1 at 6 h and to 1.90 ± 0.23 × 106 cells mL−1 at 12 h of incubation at

the in situ temperature of the control sample comprising an in situ intact

community (Fig 2a and Supplementary Table S1) However, on incubation at the

in situ temperature plus 5° the total number of prokaryotes increased to 3.73 ±

0.76 × 106 cells mL−1 after 6 h and to 3.18 ± 0.93 × 106 cells mL−1 after 12 h

(Supplementary Table S1) This implied that the highest growth rate (µ) was

obtained for 0–12 h of incubation, as µ = 0.068 h−1 (doubling time, Td = 10.2 h)

at the in situ temperature (Fig 3a) When the community was incubated at the in

situ temperature plus 5°C, µ increased to 0.111 h−1 (Td = 6.2 h) Thus, increasing

the incubation temperature had a positive effect on the abundance and growth rate of the examined prokaryotes Although this trend was confirmed for all the Suruga Bay samples; the effect was not positive after incubation for 6 h in February 2013 when the warming effect that appeared at 12 and 24 h was weaker among the four experiments (Fig 3a; Fig 4d)

The highest growth rates at the in situ temperature in each experiment, determined from the numbers shown in Supplementary Table S1 were µ = 0.037

h−1 (Td = 18.7 h) after 12 h in April 2012 and µ = 0.111 h−1 (Td = 6.2 h) after 6 h in

August 2011 (Fig 3a) Warming increased the growth rate from µ = 0.037 to

0.058 h−1 (Td = 11.9 h) in April 2012 and from µ = 0.111 to 0.118 h−1 (Td = 5.9 h)

in August 2011 In February 2013, a warming effect was not apparent after

incubation for 6 h but appeared with longer incubation, from µ = 0.028 h−1 (Td =

24.8 h) to µ = 0.053 h−1 (Td = 13.1 h) at 12 h and from µ = 0.023 h−1 (Td = 30.1 h)

to µ = 0.031 h−1 (Td = 22.4 h) after 24 h of incubation

For samples from Ha Long Bay, the warming effect on the abundance and growth rate of prokaryotes was apparent after incubation for 12 h and 20 h in

April 2013, from µ = 0.047 h−1 (Td = 14.7 h) to µ = 0.080 h−1 (Td = 8.7 h) and

from µ = 0.063 h−1 (Td = 11.0 h) to µ = 0.075 h−1 (Td = 9.2 h), respectively (Fig 3b) This positive effect appeared only after 6 h in August 2012, from negative growth to positive growth After incubation for 6 h, this effect disappeared in August 2012

2.3.3 Warming effect on bottom-up or top-down control by grazing

To determine whether the positive effect of warming on the abundance and growth rates of prokaryotes in natural community could be ascribed to bottom-up or top-down control, Iattempted to evaluate these effects on the basis

of the differences between control and dilution samples and between control and protist-free samples

For Suruga Bay in October 2012, the effect of warming on bottom-up control was negative, being μ = −0.043 h−1 after 12 h (when the highest growth rate of natural community was observed) and μ = −0.032 h−1 after 24h (Fig 4c) Thus, bottom-up control did not explain the effect of warming on enhanced abundance and growth rate in this case Because the effect of warming on top-down control was negative for this case (μ = −0.044 h−1 at 12 h and μ =

−0.036 h−1 at 24h, shown in Fig 4c), this indicated that the grazing pressure became weaker on warming, suggesting prokaryote mortality was reduced Top-down control by grazing was slightly stronger than bottom-up control This suppression, by warming, of top-down control may explain the overall positive effect of warming on the growth rate in a natural community observed for the Suruga Bay sample obtained in October 2012, although the difference individual strengths of the effect was not in good agreement with the increase in the growth rate of the natural community The abundance of protists was approximately 103 cells mL−1 in both Suruga Bay and Ha Long Bay samples

Similarly, warming did not positively affect bottom-up control of prokaryotes in the water obtained from Suruga Bay in August 2011, although the effect of warming on top-down control was not examined in this experiment (Fig 4b)

The effect of warming on bottom-up control for the Ha Long Bay sample obtained in April 2013 was also negative (μ = −0.020 h−1 at 12 h and μ = −0.002

h−1 at 20 h, Fig 4e), but after 12 h the strength of top-down control by grazing was reduced by more than the decrease of bottom-up control This may explain the positive effect of warming on the whole natural community However, the

phenomenon changed in 24 h

For the Suruga Bay sample obtained in April 2012, the effect of warming

on the natural community was apparent when incubation was longer than 12 h (Fig 3a) and the effect of warming on bottom-up control explained this increase (Fig 4a) After 24 h, in contrast, the apparent effect of warming on the natural community was ascribable to both a positive effect on bottom-up and a negative effect on top-down control (Fig 4a)

For the Suruga Bay sample obtained in February 2013, accelerated growth of prokaryotes as a result of the effect of warming on bottom-up control may explain the increase in the growth rate of the prokaryotes in the natural community (Fig 4d) However, the growth rate was substantially different for prokaryotes in the natural community and in bottom-up control Although the effect of warming on top-down control was positive, its strength was less than that of bottom-up control Thus, the positive effect of warming on natural community could be ascribable to bottom-up control in this instance

Because the effect of warming on either top-down or bottom-up control, or both, could not clearly explain its positive effect on the natural community after 6

h of incubation, the data are not discussed in depth

2.3.4 Warming effects on Gammaproteobacteria

FISH-reactive bacterial populations elucidated with the primer of EUB338 (EUB) comprised 53.3, 21.8, 38.6, and 57.0% of the total prokaryote counts

(TDC) for the in situ samples obtained from Suruga Bay in April 2012, August

2011, October 2012, and February 2013, respectively Thus, the corresponding

contributions of Gammaproteobacteria to EUB were 59.7, 48.9, 49.8, and 85.4%,

respectively, and these were 31.8, 10.7, 19.2 and 48.7% of TDC Fig 5a shows the results for Suruga Bay in October 2012; the abundance of

Gammaproteobacteria after incubation for 24 h at the in situ temperature plus

five degrees for control, dilution and protist-free samples were greater than for

the corresponding samples incubated at the in situ temperature Warming had a

positive effect in Suruga Bay generally, except for several cases which appeared for 0-3 h (μ = −0.078 h−1) on October 2012 and for 0-24 h (μ = −0.006 h−1) in February 2013 (calculated from the results listed in Supplementary table S2)

In Ha Long Bay, EUB comprised 46.4% and 8.7% of TDC for the in situ

samples of April 2013 and August 2012, respectively The corresponding

contributions of Gammaproteobacteria to EUB were 41.7% and 55.1%,

respectively, and these were 19.3% and 5.0% of TDC A positive effect of

warming on Gammaproteobacteria in natural community was also found in Ha

Long Bay except for longer incubation: 0-20 h (μ = −0.003 h−1) or 0-24 h (μ =

−0.003 h−1) (calculated from the results listed in Supplementary Table S2) The

abundance of Gammaproteobacteria was increased by warming, particularly for

dilution and protist-free experiments conducted in April 2013 (Fig 5b) However,

the positive effect of warming on Gammaproteobacteria in the natural

community was not simply explained by either bottom-up nor top-down control

2.4 Discussions

We report herein the effect of warming of the coastal water from

temperate and subtropical zones where in situ temperature ranged from 12.1 to

25.6oC (Suruga Bay) or were 22.7 or 29.9oC (Ha Long Bay), focusing in

particular on Gammaproteobacteria Size fractionation and dilution experiments

were conducted to determine whether bottom-up or top-down control strongly affects the microbial community Raising the temperature by three or five degrees increased the abundance of prokaryotes of the natural communities of both temperate and subtropical zones, by enhancing their growth rates, with the

exception of a sample collected from Ha Long Bay in August 2012, when the in

situ temperature was nearly 30°C This exception could be explained by an

insignificant increase in the number of Gammaproteobacteria (Supplementary

Table S2) A positive relationship between prokaryotic abundance and

temperature has previously been shown for Suruga Bay (Takenaka et al 2007, Hao et al 2010) In addition to those findings, our results showed that the

prokaryotic community in Suruga Bay increased in abundance with increasing temperature, irrespective of season Findings were obtained from incubation experiments which covered the doubling time of prokaryotes

Warming by three degrees accelerated the bacterial growth rate in a salt

lagoon of the Mediterranean Sea, where the in situ temperature ranged from

15.7°Cto 17.8°C (Bouvy et al 2011) Another study conducted in a Fjord in

Norway observed a similar effect of warming by up to plus five degreeshigher

than the in situ temperature of approximately 6°C (Lara et al 2013) In the coastal zone of the subtropical western Pacific Ocean, Tsai et al (2008)

observed that both the abundance (0.2 × 106 to 1.0 × 106 cells mL−1) and growth rate (0.001 to 0.062 h−1) were higher in the summer season, when the

temperature exceeded 25°C than in the winter season, when the temperature ranged from 16°C to 25°C, although they did not examine the effect of warming

in this study Tsai et al (2013) recently confirmed that growth rates of bacteria were positively correlated with temperature (r 2 = 0.43, P < 0.05) Šolić and

Krstulović (1994) also showed a clear positive correlation (r = 0.93; P < 0.0001)

between temperature and growth rate on the basis of seasonal observations in the Adriatic Sea In addition to those findings, we add further understanding that warming by 3 or 5°C positively affects growth rate throughout the seasons, with temperatures ranging from 12.1°C to 25.6°C However, this effect was not observed in the summer season for subtropical water

The positive effect of warming on prokaryotes in the natural community cannot be simply explained on the basic of either bottom-up or top-down effects For two of five cases studied, enhancement of the growth rate was ascribable to bottom-up control Two other cases could be ascribed to less top-down control (mortality) as a result of warming, irrespective of whether or not warming enhance bottom-up control Compared with the finding as Šolić et al (2009), that

top-down control dominated during the warmer period in their in situ observation,

warming sometimes resulted in less top-down control, which led to an overall increase in the abundance of prokaryotic organisms, which might be a characteristic finding in this study However, there remained a case that neither bottom-up nor top-down control, nor a combination of the two, could explain the positive effect of warming on the prokaryotes in the natural community This is probably ascribable to the complexity of constituents, which is not clearly revealed by the method used, and to the difference in time needed for a response among the difference in constituents of the different trophic levels In our dilution experiments, the abundance of phytoplankton was reduced by

filtration through a 0.2-µm filter This may have led to weak or negative growth of prokaryotes which require exudates of phytoplankton (Kato and Stabel 1984) Because growth of the prokaryotes in the natural community may be enhanced

by the presence of protists and/or zooplankton in that community (Kogure et al

1980), the complexities of a food web and the effects of viral infection must also

be taken into consideration

The complexity of these effects arises, first, because warming not only affects the grazing of metazoans but also the continuous growth of prokaryotes

In contrast with top-down control, warming accelerates the activities of both prokaryotes and phytoplankton when considering bottom-up control (Takenaka

et al 2007; von Scheibner et al 2013; Jiang et al 2014) Thus, the warming

effect is more clearly apparent for bottom-up control if it operates positively However, a positive correlation between phytoplankton and prokaryotes in natural systems usually appears after a time lag (Kato and Stabel 1984, von

Scheibner et al 2013) Thus, the question of “tempo” remains Furthermore, we

need to understand whether the phenomena observed in short-time experiments can be used to deduce the long-term phenomena that occur in nature (Sarmento

et al 2010)

Among the constituents of prokaryotic communities, Yokokawa et al (2004) found that the growth rate of Gammaproteobacteria in a Delaware

estuary was high in June, when the temperature was 20.9°C In contrast, the

growth rates of Alphaproteobacteria and Cytophaga –Flavobacter exceeded

those of Gammaproteobacteria in the winter season when the temperature was 12.8°C Similarly, the growth rates of Gammaproteobacteria correlated with the temperature (r = 0.71, P < 0.05) in Otsuchi Bay and in Sagami Bay (Yokokawa

and Nagata 2005) In contrast with these seasonal observations, we directly

observed that the growth rate of Gammaproteobacteria in given community in

temperate and subtropical seawaters was accelerated by warming by 3 or 5

degrees (analyses for in situ temperatures ranging from 12.1°C to 25.6°C were

conducted by using the results listed in Supplementary Table S2)

Growth of Gammaproteobacteria, in particular most Enterobacteriaceae and Vibrios, is known to be optimum at temperatures >30°C (Madigan et al 2000; Oliver et al 2013), which is higher than the in situ temperatures observed

in the this study, except in August in the subtropical seawater of Ha Long Bay

when a positive effect of warming on Gammaproteobacteria was not observed 16S rRNA gene-sequence analysis sometimes showed that some Legionella sp and Escherichia coli appeared after warming (Table 3) The exceptional observation that Gammaproteobacteria did not increase in the natural

community (control) on warming may be because of rapid growth of other constituents of the community, and it has been suggested that responses of bacterial growth to environmental variables may differ among different groups

(Yokokawa et al 2004) Although 16S rRNA gene-sequence analysis did not detect Vibrio spp., analysis showed that other possible pathogenic bacteria

became abundant on warming (Table 3) This suggested that increasing in

abundance of Gammaproteobacteria on warming might be supported by bacteria other than vibrios, for example Escherichia coli, Legionella sp (Table 3)

In addition, the coastal waters studied in this work are well supplied with nutrients from the land, as a result of increasing human activity, particularly in subtropical zones Moreover, because the strongest ocean warming is predicted for the surface layer in tropical and subtropical regions of Northern Hemisphere (IPCC 2013), changes in microbial growth and community constituents will also become serious problems in subtropical zones

Acknowledgements

We thank two anonymous reviewers of the Journal of Oceanography for their careful reading the manuscript This study was supported by the Graduate School of Science and Technology, Shizuoka University; Special Research Fund

VAST03.06/13-14, Viet Nam We thank all the laboratory members for their assistance in sampling and conducting of the experiments

References for chapter 2

1 Altschul SF, TL Madden, AA Schäffer, J Zhang, Z Zhang, W Miller, JD Lipman (1997) Gapped BLAST and PSI-BLAST: a new generation of protein database search programs Nucleic Acids Res 25:3389–3402

2 Amann RI, Binder BJ, Olson RJ, Chisholm SW, Devereux R, Stahl DA (1990) Combination of 16S rRNA-targeted oligonucleotide probes with flow cytometry for analyzing mixed microbial populations Appl Environ Microbiol 56:1919–1925

3 Bouvy M, Bettarel Y, Bouvier C, Domaizon I, Jacquet S, Le Floc'h E, Montanie

H, Mostajir B, Sime-Ngando T, Torreton JP, Vidussi F, Bouvier T (2011) Trophic interactions between viruses, bacteria and nanoflagellates under various nutrient conditions and simulated climate change Environ Microbiol 13(7):1842–1857

4 Brown JR, Volker C (2004) Phylogeny of gamma-proteobacteria: resolution of one branch of the universal tree? Bioessays 26(5):463–468

5 Burrows MT, Schoeman DS, Buckley LB, Moore P, Poloczanska ES, Brander

KM, Brown C, Bruno JF, Duarte CM, Halpern BS, Holding J, Kappel CV, Kiessling W, O'Connor MI, Pandolfi JM, Parmesan C, Schwing FB, Sydeman

WJ, Richardson AJ (2011) The pace of shifting climate in marine and terrestrial ecosystems Science 334(6056):652–655

6 Colwell RR (1996) Global climate and infectious disease: the cholera paradigm Science 274:2025–2031

7 Daufresne M, Lengfellner K, Sommer U (2009) Global warming benefits the small in aquatic ecosystems Proc Natl Acad Sci USA 106(31):12788–12793

8 Giovannoni SJ, DeLong EF, Olsen GJ, Pace NR (1988) Phylogenetic group specific oligodeoxynucleotide probes for identification of single microbia cells J Bacteriol 170:720–726

9 Hao DM, Tashiro T, Kato M, Sohrin R, Ishibashi T, Katsuyama C, Nagaosa K, Kimura H, Thanh TD, Kato K (2010) Population dynamics of Crenarchaeota and Euryarchaeota in the mixing front of river and marine waters Microbes Environ 25(2):126–132

10 IPCC (2013) Summary for Policymakers In: Climate Change 2013: The Physical Science Basis Contribution of Working Group I to the Fifth Assessment Report of the Intergovernmental Panel on Climate Change [Stocker TF, Qin D, Plattner G.-K., Tignor M, Allen SK, Boschung J, Nauels A, Xia Y, Bex V and Midgley PM (eds)] Cambridge University Press, Cambridge, United Kingdom and New York, USA

11 Jiang Z, Liub J, Chen J, Chen Q, Yan X, Xuan J, Zeng J (2014) Responses of summer phytoplankton community to drastic environmental changes in the Changjiang (Yangtze River) estuary during the past 50 years Water Research 54:1–11

12 Kato K, Stabel HH (1984) Studies on the carbon flux from phyto- to bacterioplankton communities in Lake Constance Arch Hydrobiol 102:177–192

13 Kogure K., Simidu U, Taga N (1980) Effect of phyto- and zooplankton on the growth of marine bacteria in filtered seawater Bull Jap Soc Sci Fisher 46:323-326

14 Landry MR, Hassett RP (1982) Estimating the grazing impact of marine micro-zooplankton Marine Biology 67:283–288

15 Lane JD (1991) 16S/23S rRNA sequencing In: Stackebrandt E, Goofellow M, editors Nucleic acid techniques in bacterial systematics John Wiley & Sons, pp115–175

16 Lara E, Arrieta JM, Garcia-Zarandona I, Boras JA, Duarte CM, Agustí S, Wassmann PF, Vaqué D (2013) Experimental evaluation of the warming effect

on viral, bacterial and protistan communities in two contrasting Arctic systems Aquat Microb Ecol 70:17–32

17 Lim EL, Amaral LA, Caron DA, Delong EF (1993) Application of rRNA-based probes for observing marine nanoplanktonic protists Appl Environ Microbiol 59:1647–1655

18 Lim EL, Caron DA, Delong EF (1996) Development and field application of a quantitative method for examining natural assemblages of protists with oligonucleotide probes Appl Environ Microbiol 62:1416–1423

19 Lomas MW, Glibert P, Shiah F, Smith EM (2002) Microbial processes and temperature in Chesapeake Bay: Current relationships and potential impacts of regional warming Global Change Biology 8:51-70

20 Manz W, Amann R, Ludwig W, Wagner M, Schleifer KH (1992) Phylogenetic oligodeoxynucleotide probes for the major subclasses of Proteobacteria: problems and solutions Syst Appl Microbiol 15:593–600

21 Madigan MT, Martinko JM, Parker J (2000) Brock biology of microorganisms 9th edn Prentice-Hall Inc., New Jersey, USA, pp 455-484

22 Oliver JD, Pruzzo C, Vezzulli L, Kaper JB (2013) Vibrio species In: Doyle MP, Buchanan RL (eds) Food microbiology: fundamentals and frontiers, 4th edn ASM, Washington 401–440 doi:10.1128/9781555818463.ch16

23 Petchey OL, McPhearson PT, Casey TM, Morin PJ (1999) Environmental warming alters food-web structure and ecosystem function Nature 402: 69–72

24 Porter KG, Feig YS (1980) The use of DAPI for identifying and counting aquatic microflora Limnol Oceanogr 25(5): 943–948

25 Sarmento H, Montoya JM, Vázquez-Domínguez E, Vaqué D, Gasol JM (2010) Warming effects on marine microbial food web processes: how far can we go