Effect of ocean acidification on microbial diversity and on microbe-driven biogeochemistry and ecosystem functioning

HAL Id: hal-03502078

https://hal.science/hal-03502078

Submitted on 24 Dec 2021

HAL is a multi-disciplinary open access

archive for the deposit and dissemination of sci-

entic research documents, whether they are pub-

lished or not. The documents may come from

teaching and research institutions in France or

abroad, or from public or private research centers.

L’archive ouverte pluridisciplinaire HAL, est

destinée au dépôt et à la diusion de documents

scientiques de niveau recherche, publiés ou non,

émanant des établissements d’enseignement et de

recherche français ou étrangers, des laboratoires

publics ou privés.

Distributed under a Creative Commons Attribution 4.0 International License

Eect of ocean acidication on microbial diversity and

on microbe-driven biogeochemistry and ecosystem

functioning

Jinwen Liu, Markus G Weinbauer, Cornelia Maier, Minhan Dai, Jean-Pierre

Gattuso

To cite this version:

Jinwen Liu, Markus G Weinbauer, Cornelia Maier, Minhan Dai, Jean-Pierre Gattuso. Eect of ocean

acidication on microbial diversity and on microbe-driven biogeochemistry and ecosystem functioning.

Aquatic Microbial Ecology, 2010, 61 (3), pp.291-305. �10.3354/ame01446�. �hal-03502078�

AQUATIC MICROBIAL ECOLOGY

Aquat Microb Ecol

Vol. 61: 291–305, 2010

doi: 10.3354/ame01446

Published online October 11

INTRODUCTION

The partial pressure of carbon dioxide (pCO

)

increases in the atmosphere due to the anthropogenic

input of CO

through the burning of fossil fuel, cement

production and land-use change. It has increased by

32% between 1880 and 2000 (280 to 379 µatm;

Solomon et al. 2007), leading to changes in the Earth’s

climate and the functioning of terrestrial ecosystems.

Over the past 250 yr, the world’s oceans have absorbed

about one-third of the anthropogenic CO

, which is

now distributed from the surface to depths ranging

from a few hundred to a few thousand metres (Sabine

et al. 2004). The uptake of anthropogenic CO

pro-

foundly affects the parameters of the carbonate chem-

istry (e.g. Gattuso & Lavigne 2009), leading to an

increase of pCO

and of the concentrations of dissolved

inorganic carbon (DIC) and bicarbonate ions (HCO

–

as well as to a decrease of pH and of the concentration

of carbonate ions (CO

2–

). The term ‘ocean acidifica-

Effect of ocean acidification on microbial diversity

and on microbe-driven biogeochemistry and

ecosystem functioning

Jinwen Liu

1, 2, 3

, Markus G. Weinbauer

1, 2

, Cornelia Maier

1, 2

, Minhan Dai

Jean-Pierre Gattuso

1, 2,

INSU-CNRS, Laboratoire d’Océanographie de Villefranche, BP 28, 06234 Villefranche-sur-mer Cedex, France

Université Pierre et Marie Curie, Observatoire Océanologique de Villefranche, 06230 Villefranche-sur-mer, France

State Key Laboratory of Marine Environmental Science, Xiamen University, 361005 Xiamen, China

ABSTRACT: The ocean absorbs about 25% of anthropogenic CO

emissions, which alters its chem-

istry. Among the changes of the carbonate system are an increase in the partial pressure of CO

(pCO

) and a decline of pH; hence, the whole process is often referred to as ‘ocean acidification’.

Many microbial processes can be affected either directly or indirectly via a cascade of effects through

the response of non-microbial groups and/or through changes in seawater chemistry. We briefly

review the current understanding of the impact of ocean acidification on microbial diversity and pro-

cesses, and highlight the gaps that need to be addressed in future research. The focus is on Bacteria,

Archaea, viruses and protistan grazers but also includes total primary production of phytoplankton as

well as species composition of eukaryotic phytoplankton. Some species and communities exhibit

increased primary production at elevated pCO

. In contrast to their heterocystous counterparts, nitro-

gen fixation by non-heterocystous cyanobacteria is stimulated by elevated pCO

. The experimental

data on the response of prokaryotic production to ocean acidification are not consistent. Very few

other microbial processes have been investigated at environmentally relevant pH levels. The poten-

tial for microbes to adapt to ocean acidification, at either the species level by genetic change or at the

community level through the replacement of sensitive species or groups by non- or less sensitive

ones, is completely unknown. Consequently, the impact of ocean acidification on keystone species

and microbial diversity needs to be elucidated. Most experiments used a short-term perturbation

approach by using cultured organisms; few were conducted in mesocosms and none in situ. There is

likely a lot to be learned from observations in areas naturally enriched with CO

, such as vents,

upwelling and near-shore areas.

KEY WORDS: Ocean acidification · Microbial diversity · Microbe · Bacteria · Phytoplankton · Viruses ·

Biogeochemistry · Meta-analysis

Resale or republication not permitted without written consent of the publisher

Contribution to AME Special 4 ‘Progress and perspectives in aquatic microbial ecology’

PEN

CCESS

Aquat Microb Ecol 61: 291–305, 2010

tion’ (Caldeira & Wickett 2003) relates to the decrease

in pH but does not imply that the pH of ocean surface

waters will become acidic (i.e. below 7) any time soon.

If the current trends in CO

emissions continue to

increase, the pH of the global surface ocean could

decrease by about 0.4 units by the end of the century

compared to pre-industrial times. Changes will be more

pronounced in areas such as the Southern Ocean, which

will become undersaturated with respect to aragonite in

2050 (Orr et al. 2005), and the Arctic Ocean where arag-

onite undersaturation will occur even sooner (Steinacher

et al. 2009). This change in the chemistry of the oceans is

quantifiable and predictable for a given level of atmos-

pheric pCO

. Observations at several time-series

stations, even though all of them relatively short (<20 yr),

confirm the predicted changes in the carbonate

chemistry (e.g. Santana-Casiano et al. 2007).

The biological, ecological and biogeochemical re-

sponses of marine organisms and communities have

only been actively studied in the past few years, and

there is still a high level of uncertainty and debate on

the significance and magnitude of those responses

(Hendriks et al. 2010, Dupont et al. 2010). A meta-

analysis of the response of marine organisms to ocean

acidification recently performed by Hendriks et al.

(2010) only partly covers microbial groups and micro-

bial biogeochemistry. Marine microbes, here consid-

ered as single-celled organisms (i.e. prokaryotes and

protists) and viruses, are very diverse, as they thrive in

a large range of habitats and perform a wide range of

functions. They are therefore involved in virtually all

geochemical reactions occurring in the oceans (Kirch-

man 2008). Some of these functions are even exclu-

sively performed by prokaryotes. Some prokaryotes are

able to withstand extreme pH values with optimal pH

for growth ranging from 0.7 to >10 (Cavicchioli 2002).

The microbial group that has been investigated most

thoroughly with respect to the effect of ocean acidifica-

tion is eukaryotic phytoplankton (Riebesell 2004, Gior-

dano et al. 2005, Beardall et al. 2009). In contrast, other

groups such as viruses, Archaea, Bacteria and hetero-

trophic protists have received considerably less

attention. In a recent ‘perspective’ paper, Joint et al. (in

press) asked ‘Will ocean acidification affect marine

microbes?’ In a narrative (sensu Gates 2002) review,

Joint et al. (in press) looked at some of the relevant lit-

erature and came to the conclusion that ‘perhaps the

most appropriate null hypothesis to test is that marine

microbes possess the flexibility to accommodate pH

change and there will be no catastrophic changes in

marine biogeochemical processes that are driven by

phytoplankton, bacteria and archaea’ and recognised

that calcification and photosynthesis could be affected.

Narrative reviews have the potential for serious bias,

which could lead to misleading conclusions (Gates

2002). Meta-analysis was developed to overcome most

biases of narrative reviews. It statistically combines the

results (effect size) of several studies that address a

shared research hypothesis.

Here we used a meta-analytic approach to compre-

hensively review the current understanding of the

effect of ocean acidification on microbes and microbial

processes, and to highlight the gaps that need to be

addressed in future research.

METHODS

Data were collected from the EPOCA/EUR-

OCEANS database (Nisumaa et al. 2010), the tables or

text of papers, or interpolated from figures. Only papers

reporting the effect of elevated pCO

or decreased pH in

the range of values expected during the period spanning

the last glacial maximum and the year 2100 were se-

lected. Unless mentioned otherwise, pH values are re-

ported on the total scale (Dickson 2010). pCO

levels

were calculated from pH and other ancillary data with

the R software package seacarb (Lavigne & Gattuso

2010) as described by Nisumaa et al. (2010). Data permit-

ting, 2 effect sizes were calculated for each variable:

H:C, the value at high pCO

versus the value at control

pCO

; and C:L, the value at control pCO

versus the

value at low pCO

. The ranges of high (H), control (C)

and low (L) pCO

were 450 to 1500, 300 to 450 and 100 to

300 µatm, respectively. These levels roughly correspond

to ‘future’, ‘present’ and ‘glacial’ conditions. For meso-

cosm experiments, in which pCO

was drifting, the data

were categorised into the ‘low’, ‘control’ and ‘high’ pCO

categories according to the pCO

values at the begin-

ning of the time period considered. Only the data col-

lected during the pre-bloom phase of mesocosm exper-

iments were used because the termination of the blooms

may have distinct causes in addition to nutrient exhaus-

tion (e.g. viral lysis): Days 5 to 14 for PeECE I (Engel et al.

2005), Days 0 to 14 for PeECE II (Grossart et al. 2006a)

and Days 6 to 9 for PeECE III (Riebesell et al. 2008).

There are 2 exceptions. The community respiration data

reported by Egge et al. (2009) exhibit a large day-to-day

variability and only Day 8 was considered. Tanaka et al.

(2008) did not report alkaline phosphatase activity data

for Days 6 to 9; hence data from Days 13 to 19 were used.

The effect size is shown in Fig. 1 (see Fig. 1 in ‘Results’.

It is important to note that the effect is monotonous when

H:C and L:C are both above or below 1, whereas the ef-

fect exhibits an extreme when one of the effects is above

1 and the other below 1.

Meta-analyses were performed with the R package

meta 1.1.8 (Schwarzer 2010) to test the significance of

the effect sizes H:C and L:C. The results are shown in

Tables 1 & 2. Fixed and random effects models can be

292

Liu et al.: Ocean acidification effects on microbes

used in a meta-analysis (Borenstein et al. 2009). The

fixed effect model generally assumes that all studies

shared a single effect, whereas the effect could be dif-

ferent from study to study in the random effects model.

There is no reason to assume that ocean acidification

has the same effect on a variable measured in different

species and/or under different experimental condi-

tions. However, some have argued that the fixed effect

method is valid without assuming a common true

effect size (Borenstein et al. 2009); hence, both models

were used. Inverse variance weighting was used for

pooling and the DerSimonian-Laird estimate was used

in the random effects model. Hedges’ adjusted g was

applied to calculate the standardised mean difference

(SMD). The heterogeneity of responses with a random

effects model was assessed using the Q statistics as

well as with the H and I

indices (Borenstein et al.

2009). I

is 0 when all variability in effect size estimates

is due to sampling error within studies; I

values of 25,

50 and 75% correspond to mean low, medium and

high heterogeneity, respectively. When H < 1.2, there

is no heterogeneity within the studies, while there are

obvious differences between studies when H > 1.5.

pH REGULATION

Marine microbes have an optimum pH range that

varies between species, and many physiological pro-

cesses are pH-dependent. It is therefore critical that

293

●

012345

Tanaka et al. (2008)*

Fu et al. (2008)

Levitan et al. (2007)

Hutchins et al. (2007)

Czerny et al. (2009)

Barcelos e Ramos et al. (2007)

Larsen et al. (2008)*

PeECE I (2001) (unpubl. data)*

Larsen et al. (2008)*

PeECE I (2001) (unpubl. data)*

Allgaier et al. (2008)*

Grossart et al. (2006a)*

Rochelle−Newall et al. (2004)*

Allgaier et al. (2008)*

PeECE I (2001) (unpubl. data)*

Allgaier et al. (2008)*

PeECE I (2001) (unpubl. data)*

Grossart et al. (2006a)*

Allgaier et al. (2008)*

Grossart et al. (2006a)*

Allgaier et al. (2008)*

Grossart et al. (2006a)*

Allgaier et al. (2008)*

Grossart et al. (2006a)*

Allgaier et al. (2008)*

Grossart et al. (2006a)*

Allgaier et al. (2008)*

Grossart et al. (2006a)*

Allgaier et al. (2008)*

Rochelle−Newall et al. (2004)*

Schulz et al. (2008)*

Rochelle−Newall et al. (2004)*

Egge et al. (2009)*

Engel et al. (2004)*

Engel (2002)

●

H:C response ratio

●

C:L response ratio

TEP production

DOC concentration

CDOM concentration

BPP (total)

BPP (free)

BPP (attached)

csBPP (total)

csBPP (free)

csBPP (attached)

BA (HBA)

BA (LBA)

BA (total)

VA (LVA)

VA (HVA)

Nitrogen fixation

uptake (>10 µm)

Fig. 1. Impact of ocean acidification on microbial processes and on the parameters involved. The H:C ratios, i.e. the values at high

partial pressure of CO

(pCO

) versus the values at control pCO

, are shown as filled circles, whereas the C:L ratios, i.e. the values

at control pCO

versus the values at low pCO

, are shown as open circles. The nitrogen fixation rate reported by Hutchins et al.

(2007) at low pCO

was 0; hence the C:L ratios could not be calculated. Asterisks (*) indicate data collected in a mesocosm exper-

iment during which pCO

was drifting (see ‘Methods’). TEP: transparent exopolymer particles; DOC: dissolved organic carbon;

CDOM: chromophoric or coloured dissolved organic matter; BPP: bacterial protein production; attached and free: Bacteria

attached on particles or free; csBPP: cell-specific bacterial protein production; BA: bacterial abundance; HBA and LBA: abun-

dances of high- and low-fluorescence Bacteria; VA: viral abundance; HVA and LVA: abundances of high- and low-fluorescence

viruses; APA: alkaline phosphatase activity

Fig. 1 (continued on next page)

Aquat Microb Ecol 61: 291–305, 2010

intracellular pH (pH

) is maintained by a pH homeosta-

tic system (Booth 1985). pH

depends on the external

pH, cytoplasmic buffers, the intracellular generation of

acids and bases, and an active transport of H

(or OH

–

On short time scales, micro-organisms are often able to

buffer external changes in pH, thereby preventing

damage to internal processes and functioning. Few

studies on pH control were performed on marine

microbes, and all were carried out in experimental

conditions that are not environmentally relevant

(Takeuchi et al. 1997, Labare et al. 2010). Cultures

were done in highly enriched media and the context of

disposal, hence at CO

and pH levels that are not

relevant to ‘natural’ ocean acidification. For example,

Labare et al. (2010) reported morphological changes

and a temporary inhibition of growth in the marine

bacterium Vibrio sp. grown at pH (scale not men-

tioned) 5.2. The recovery of growth after 6 h indicates

294

●

1234567

Fu et al. (2008)

Fu et al. (2007)

Hutchins et al. (2007)

Levitan et al. (2007)

Fu et al. (2008)

Fu et al. (2007)

Kranz et al. (2009)

Czerny et al. (2009)

Hutchins et al. (2007)

Levitan et al. (2007)

Barcelos e Ramos et al. (2007)

Piontek et al. (2010)

Tanaka et al. (2008)*

Grossart et al. (2005)*

Delille et al. (2005)*

Egge et al. (2009)*

Piontek et al. (2010)

Egge et al. (2009)*

Delille et al. (2005)*

Fu et al. (2008)

Fu et al. (2007)

Czerny et al. (2009)

Barcelos e Ramos et al. (2007)

Schulz et al. (2008)*

Hutchins et al. (2007)

Engel et al. (2002)

Engel et al. (2005)*

Fu et al. (2008)

Fu et al. (2007)

Czerny et al. (2009)

Barcelos e Ramos et al. (2007)

Schulz et al. (2008)*

Engel et al. (2002)

Engel et al. (2005)*

Fu et al. (2008)

Fu et al. (2007)

Levitan et al. (2007)

Kranz et al. (2009)

Czerny et al. (2009)

Barcelos e Ramos et al. (2007)

Schulz et al. (2008)*

Hutchins et al. (2007)

Engel et al. (2008)*

Engel (2002)

Engel et al. (2005)*

●

H:C response ratio

●

C:L response ratio

C:N ratio

C:P ratio

N:P ratio

Net community respiration

Carbon loss

Net primary production

Enzyme activity (protease)

Enzyme activity (

−glucosidase)

Enzyme activity (

−glucosidase)

Enzyme activity (APA)

Enzyme activity

(extracellular glucosidase)

Cyanobacterial growth rate

Cyanobacterial CO

fixation rate

Fig. 1 (continued)

Liu et al.: Ocean acidification effects on microbes

an efficient pH regulation system. It is unknown how

long such compensation mechanisms can withstand an

environmentally relevant decline of pH, and whether

adaptation may occur over longer time scales.

TEP, DOM and CDOM

Transparent exopolymer particles (TEP) are major

components of marine aggregates that are enriched in

carbon relative to nitrogen or phosphorus, contribute

to carbon export and provide habitats and sites for

attachment of Bacteria (Passow 2002). Bacterial degra-

dation can be an important pathway of TEP loss

(Passow 2002). The effect size is above 1 (Fig. 1), sug-

gesting a monotonous effect of elevated pCO

that sti-

mulates TEP production. However, the SMD of TEP

production between high and control pCO

is signifi-

cantly different from 0 with the fixed effect model but

it is not significantly different from 0 with the random

effect model (Table 1). The SMD between the control

and low pCO

is not significant with any of the models.

Engel (2002) reported that the production of TEP by

natural phytoplankton communities increases with

increasing pCO

but that at present-day pCO

, TEP

production is already saturated. Higher pCO

levels, as

predicted for the future, would not lead to higher TEP

production even though primary production might be

stimulated (see below). In another mesocosm experi-

ment, TEP production normalised to the abundance of

Emiliania huxleyi was significantly higher in the high

treatment ‘future’ (~710 µatm) than in the ‘pre-

sent’ (~410 µatm) and ‘glacial’ (~190 µatm) treatments

(Engel et al. 2004). This indicates a possible direct

effect of pCO

on polysaccharide exudation. However,

the enhanced DIC level did not cause an increase in

the particulate organic carbon (POC) concentrations,

possibly due to an increased TEP production, which

stimulated particle aggregation and accelerated sedi-

mentation as previously observed by Logan et al.

(1995) and Engel (2000). Mari (2008) also investigated

the effect of seawater acidification on aggregation and

sedimentation of TEP. Their results were similar to

those of Engel et al. (2004) and suggested that a

decrease of seawater pH would lead to a significant

increase of the TEP pool. The most recent study avail-

able to date did not find a significant effect of elevated

pCO

on the TEP concentration (Egge et al. 2009). In

contrast to other studies, Mari (2008) found that the

buoyancy of TEP is pH-dependent, and increases in

pH cause TEP to ascend in the water column.

Chromophoric or coloured dissolved organic matter

(CDOM) is the fraction of the dissolved organic matter

(DOM) that absorbs light in both the ultraviolet

and visible ranges. CDOM diminishes light and is de-

graded by UVA and UVB light and thus likely plays an

important optical and ecological role in surface waters.

There is a consensus that heterotrophic Bacteria pro-

duce CDOM (Nelson et al. 1998, Rochelle-Newall et al.

1999). However, how ocean acidification will affect

heterotrophic Bacteria and CDOM release is still

poorly understood. Only 1 study reported the effect of

elevated pCO

on the concentration of CDOM, pre-

venting the use of meta-analysis. The effect size on the

dissolved organic carbon (DOC) concentration is

around 1 (Fig. 1) and the SMD of TEP production

between high and control pCO

is significantly differ-

ent from 0 with both the fixed and random effect model

(Table 1), suggesting an overall significant impact of

elevated pCO

on the DOC concentration. Too few

data are available to test the significance of the SMD

between the control and low pCO

This result contrasts with the conclusions drawn

from the 2 individual studies available to date. In a

mesocosm experiment carried out under different ini-

tial pCO

(190, 414 and 714 µatm), no impact of pCO

was observed on the concentrations of CDOM and

DOC (Rochelle-Newall et al. 2004). In the same exper-

iment, DOC was neither related to the abundance of

Emiliania huxleyi nor to TEP concentration (Engel et

al. 2004). No statistically significant effect of the CO

treatment on DOC concentration was found, although

during the course of the bloom, the DOC concentration

increased in 2 of the 3 ‘future’ mesocosms and in 1 of

the ‘present’ mesocosms, but in none of the ‘glacial’

mesocosms (initial pCO

of 714, 414 and 190 µatm,

respectively). Engel et al. (2004) suggested that the dif-

ferent response of TEP (discussed above) and DOC

may be due to differences in their bioavailability,

which could have generated a rapid response of the

microbial food web, possibly obscuring the effect of

pCO

on the DOC production of autotrophic cells. No

statistically significant CO

treatment effects on the

concentration of DOC were detected in other PeECE

experiments (Rochelle-Newall et al. 2004, Grossart et

al. 2006a, Schulz et al. 2008) or in a mesocosm experi-

ment with similar CO

treatments (Kim et al. 2006).

CARBON CYCLE

Primary production

A relatively large number of studies have investi-

gated the effect of elevated pCO

on photosynthesis at

the organismal level, especially to clarify the physio-

logical and molecular mechanisms governing the car-

bon-concentrating mechanisms (CCMs). DIC is mostly

fixed via the enzyme ribulose biphosphate carboxylase

(RUBISCO), which has a relatively low affinity for CO

295

Aquat Microb Ecol 61: 291–305, 2010

In most species, RUBISCO is less than half saturated at

present CO

levels (Giordano et al. 2005). Hence,

autotrophs that only rely on diffusive entry of CO

have poor photosynthetic efficiency. However, all

cyanobacteria and most algae have developed CCMs

to elevate the concentration of CO

at the site of car-

boxylation. It is beyond the scope of this paper to cover

this literature extensively, as several excellent reviews

are available (Riebesell 2004, Giordano et al. 2005,

Beardall et al. 2009). We will focus on cyanobacteria

and community primary production below and simply

wish to point out that large differences in the CO

sensitivity between the major groups of eukaryotic

phytoplankton exist (Riebesell 2004).

Cyanobacteria are the largest and the most widely

distributed group of photosynthetic prokaryotes (Burns

et al. 2005). This group has a major impact on the

global carbon cycle and contributes up to 50% of the

296

Parameters High vs. control pCO

(H:C) Low vs. control pCO

(L:C)

SMD SMD

Fixed effect Random effects Fixed effect Random effects

TEP production 0.83 ± 0.18 1.06 ± 1.29 –0.26 ± 0.18 –0.26 ± 0.18

(p < 0.001) (p = 0.41) (p = 0.15) (p = 0.15)

DOC concentration –0.69 ± 0.22 –0.73 ± 0.28 na na

(p = 0.002) (p = 0.009)

BPP (total) –0.12 ± 0.21 –0.13 ± 0.78 na na

(p = 0.58) (p = 0.86)

BPP (free) –0.70 ± 0.22 –0.73 ± 0.59 na na

(p = 0.001) (p = 0.21)

BPP (attached) 0.25 ± 0.21 0.3 ± 0.6 na na

(p = 0.24) (p = 0.62)

csBPP (total) –0.22 ± 0.21 –0.24 ± 0.55 na na

(p = 0.28) (p = 0.66)

csBPP (free) –0.82 ± 0.21 –0.84 ± 0.38 na na

(p < 0.001) (p = 0.027)

csBPP (attached) 0.10 ± 0.23 0.21 ± 1.15 na na

(p = 0.65) (p = 0.85)

BA (HBA) –0.43 ± 0.18 –1.48 ± 1.34 na na

(p = 0.013) (p = 0.27)

BA (LBA) –0.63 ± 0.17 –0.7 ± 0.26 na na

(p < 0.001) (p = 0.008)

BA (total) 0.06 ± 0.1 –0.67 ± 0.50 –0.38 ± 0.10 –0.79 ± 0.78

(p = 0.53) (p = 0.18) (p < 0.001) (p = 0.32)

VA (LVA) –0.42 ± 0.23 –1.02 ± 1.22 na na

(p = 0.06) (p = 0.406)

VA (HVA) –0.61 ± 0.22 –1.13 ± 1.09 na na

(p = 0.006) (p = 0.3)

Nitrogen fixation 1.06 ± 0.26 1.95 ± 0.68 –0.76 ± 0.27 –0.66 ± 0.46

(p < 0.001) (p = 0.004) (p = 0.005) (p = 0.15)

C:N ratio 0.05 ± 0.14 –0.01 ± 0.51 –0.84 ± 0.15 –0.68 ± 0.53

(p = 0.72) (p = 0.98) (p < 0.001) (N = 130, p =0.20)

C:P ratio 0.26 ± 0.17 0.27 ± 0.26 –1.76 ± 0.23 –1.44 ± 0.52

(p = 0.14) (p = 0.3) (p < 0.001) (p =0.005)

N:P ratio –0.48 ± 0.19 0.69 ± 0.59 0.49 ± 0.27 –0.17 ± 1.56

(p = 0.01) (p = 0.24) (p = 0.08) (p =0.92)

Community respiration –0.78 ± 0.22 –0.55 ± 0.50 na na

(p = 0.001) (p =0.28)

Net primary production –0.21 ± 0.30 0.22 ± 0.81 na na

(p = 0.48) (p = 0.78)

Cyanobacterial growth rate 0.68 ± 0.21 1.06 ± 0.36 –1.41 ± 0.33 –0.68 ± 1.11

(N = 69, p = 0.001) (N = 69, p = 0.003) (N = 33, p < 0.001) (N = 33, p = 0.54)

Cyanobacterial CO

fixation rate 1.21 ± 0.33 1.58 ± 0.49 0.44 ± 0.60 –3.30 ± 3.02

(N = 43, p < 0.001) (N = 43, p = 0.001) (N = 12, p = 0.46) (N = 12, p = 0.27)

Table 1. Summary standardised mean difference (SMD ± SEM) for each parameter calculated using fixed-effect and random-

effect models. N: sample size; na: no data or the data do not meet the requirements for meta-analysis; p: probability.

Other abbreviations as in the legend of Fig. 1

Liu et al.: Ocean acidification effects on microbes

fixed carbon in marine systems (Partensky et al. 1999).

Cyanobacteria are known to utilise CCMs, such as the

active transport of HCO

–

and CO

, to facilitate CO

fixation and maintain rapid growth at low external DIC

concentrations (Badger & Price 2003, Badger et al.

2006). It is reasonable to assume that increased CO

availability will reduce the need for CCM activity and

hence reduces the allocation of energy or nutrients for

carbon acquisition (Burkhardt et al. 2001, Beardall &

Giordano 2002). This may further affect the photosyn-

thesis, growth rate and other activities.

The effect of elevated pCO

on the growth rate and/or

photosynthesis of the widespread cyanobacterium Tri-

chodesmium spp. was investigated by Barcelos e Ramos

et al. (2007), Levitan et al. (2007), Hutchins et al. (2007)

and Kranz et al. (2009). Four other cyanobacterial species

were investigated: Synechococcus sp. and Prochlorococ-

cus sp. (Fu et al. 2007), Crocosphaera watsonii (Fu et al.

2008) and Nodularia spumigena (Czerny et al. 2009).

Overall, the effect size on the growth rate is above 1

both for the H:C and C:L ratios (Fig. 1), suggesting a

monotonous increase in growth rate as a function of

increasing pCO

. One notable exception, with effect

sizes below 1, is Nodularia spumigena, which exhibits

a monotonous decrease in growth rate as a function of

increasing pCO

. The SMD of cyanobacterial growth

rate between high and control pCO

is significantly

different from 0 both with the fixed effect and random

effects models (Table 1). The SMD between the control

and low pCO

is significantly different from 0 only with

the fixed effect model. An additional demonstration of

species specificity is the result that Synechococcus sp.

exhibits a much greater response to elevated pCO

(750 versus 380 µatm) and temperature (4°C increase)

than Prochlorococcus sp. (Fu et al. 2007). There is some

evidence that the growth rate of Crocosphaera in-

creases at elevated pCO

but only under Fe-replete

conditions (Fu et al. 2008).

The effect size on the cyanobacterial photosynthesis

(i.e. CO

fixation rate) is above 1 both for the H:C and

C:L ratios (Fig. 1), suggesting a monotonous increase

in the rate of photosynthesis with increasing pCO

There is only 1 exception in the study of Levitan et al.

(2007), where photosynthesis was higher at low and

high pCO

than in the control. The SMD of cyanobac-

terial photosynthesis between high and control pCO

was significantly different from 0 both with the fixed

effect and random effects models, but the SMD

between the control and low pCO

was not signifi-

cantly different from 0 (Table 1).

The stimulation of net photosynthesis at elevated

pCO

is predominantly attributed to changes in cell

division (Hutchins et al. 2007, Levitan et al. 2007) but

also to altered elemental ratios of carbon to nitrogen

(Levitan et al. 2007, Kranz et al. 2009) or nitrogen to

phosphorus (Barcelos e Ramos et al. 2007). Fu et

al. (2007) reported that the photosynthetic parame-

ters of Synechococcus sp. significantly changed at ele-

vated pCO

but only when combined with elevated

temperature.

297

Parameters High vs. control pCO

Low vs. control pCO

Q df p HI

(%) Q df p HI

(%)

TEP production 45.7 3 <0.001 3.9 93 1.5 2 0.46 1 0

DOC concentration 1.4 1 <0.001 1.2 30 na na na na na

BPP (total) 13.4 1 <0.001 3.7 93 na na na na na

BPP (free) 7.2 1 0.01 2.7 86 na na na na na

BPP (attached) 7.9 1 0.01 2.8 87 na na na na na

csBPP (total) 6.8 1 0.009 2.6 85 na na na na na

csBPP (free) 3.1 1 0.07 1.8 68 na na na na na

csBPP (attached) 26 1 <0.001 5.1 96 na na na na na

BA (HBA) 18.9 1 <0.001 4.4 95 na na na na na

BA (LBA) 1.5 1 0.22 1.2 34 na na na na na

BA (total) 32.8 2 <0.001 4.1 94 42.8 1 <0.001 6.5 97

VA (LVA) 22 1 <0.001 4.7 95 na na na na na

VA (HVA) 17.4 1 <0.001 4.2 94 na na na na na

Nitrogen fixation 33.72 9 <0.001 1.9 73 8 4 0.09 1.4 50

C:N ratio 119.2 12 <0.001 3.2 90 61.4 6 <0.001 3.2 90

C:P ratio 11.3 7 0.13 1.3 38 13.5 4 0.01 1.8 71

N:P ratio 89 12 <0.001 2.7 87 105.7 4 <0.001 5.1 96

Community respiration 2.1 1 0.15 1.5 53 na na na na na

Net primary production 3.7 1 0.06 1.9 73 na na na na na

Cyanobacterial growth rate 27.7 12 0.01 1.5 57 17.1 3 <0.001 2.4 82

Cyanobacterial CO

fixation rate 13.5 8 0.09 1.3 41 9.4 2 0.01 2.2 79

Table 2. Estimates of the heterogeneity of the effect size (Q, H and I

; see ‘Methods’). df: degrees of freedom; na: no data or the

data do not meet the requirements for meta-analysis; p: probability; other abbreviations as in the legends to Table 1 and Fig. 1

Aquat Microb Ecol 61: 291–305, 2010

There are several reports of increased community

primary production of phytoplanktonic assemblages

(some not microbial) at elevated pCO

. Hein & Sand-

Jensen (1997) indicated that elevated CO

will stimu-

late primary production in the North Atlantic. The

PeECE mesocosm experiments also investigated the

effect of elevated pCO

on net primary production. No

conspicuous change was observed in the PeECE I

(Delille et al. 2005) and II (J. Egge unpubl. data) exper-

iments, but a significant effect was found in the PeECE

III experiment. Riebesell et al. (2007) reported a 27 and

39% increase in net primary production at 2× and 3×

ambient pCO

, and Egge et al. (2009) found a higher

cumulative primary production based on the

C-incor-

porations at higher pCO

towards the end of the exper-

iment. However, other studies had reported that

increased pCO

resulted in no significant increase in

primary production (Tortell et al. 2002).

Bacterial abundance, production and enzyme activity

Bacteria are the main group of organisms able to use

DOC. Since they can be grazed by flagellates, some

of the DOC which would otherwise be lost from the

food web, can be cycled back via grazing (microbial

loop; Azam et al. 1983). Inagaki et al. (2006) reported

that the pH values of the deep-sea sediments overlying

a CO

lake ranged from 4.0 to 6.6 units, in contrast to a

pH (presumably on the National Bureau of Standards,

NBS, scale) of 7.3 outside of the CO

-hydrate zone. In

their study, a strong decline in cell numbers and abun-

dance of specific lipid biomarkers toward the liquid CO

interface on a scale of decimetres was observed: along

this gradient high abundances (>10

–3

) of microbial

cells found in sediment pavements above the CO

lake

decreased to strikingly low cell numbers (10

–3

) at the

liquid CO

/CO

-hydrate interface. In other studies per-

formed at pCO

levels relevant to future surface ocean

acidification (190 to 1050 µatm), the total abundance of

Bacteria varied considerably in phytoplankton blooms

triggered in pelagic mesocosms, but pCO

had little or no

effect on bacterial abundance (BA; Rochelle-Newall et

al. 2004, Grossart et al. 2006a, Allgaier et al. 2008). Note,

however, that a significant impact was observed during

the decline of the bloom in 1 of the experiments together

with a higher growth rate and abundance of attached

prokaryotes at the highest pCO

level (700 µatm;

Grossart et al. 2006a). Yamada et al. (2008) also found no

significant effect of pCO

values up to 10 000 µatm.

The H:C effect sizes of the abundance of high DNA

Bacteria (HBA) and low DNA Bacteria (LBA) are lower

than 1, suggesting an inhibition of bacterial abun-

dance under high pCO

, but the C:L ratios are higher

than 1, indicating a minimum abundance at control

pCO

and larger abundances at low and high pCO

(Fig. 1). Only 2 SMDs are significantly different from 0:

the high DNA Bacteria H:C ratio and the total BA L:C

ratio, both with a fixed model.

Diverse responses of bacterial production (BPP) to

elevated pCO

were found, partly depending on the

community considered (attached versus free Bacteria)

and on the normalisation used (total BPP or cell-

specific BPP, csBPP). The effect sizes suggest a mono-

tonous response in 4 of the data sets available at low,

control and high pCO

(Fig. 1). The H:C effect size is

most often much higher than the L:C effect size. In the

other 3 data sets, the response does not appear to be

monotonous. Only a few SMDs of bacterial production

between high and control pCO

are significantly dif-

ferent from 0 with the fixed effect model (total and

csBPP of free Bacteria) and the random effect model

(csBPP of free Bacteria; Table 1). Too few data are

available to test the significance of the SMD between

the control and low pCO

Grossart et al. (2006a) reported that the total BPP

(estimated using tritiated leucine incorporation) as

well as the csBPP of total and attached Bacteria are

higher at elevated pCO

. They also found a higher pro-

tease activity at elevated pCO

, whereas the activities

of α- and β-glucosidase remained unchanged. The

preferential stimulation of the abundance and activity

of attached Bacteria may result from an increased pro-

duction of TEP, which would provide surfaces for Bac-

teria and favour aggregation (Grossart et al. 2006b).

Allgaier et al. (2008) suggested that there was no

difference in BPP among pCO

treatments. However,

linear regressions between BPP of free-living Bacteria,

BPP of attached Bacteria or csBPP of attached Bacteria

and C:N ratio of suspended matter were significantly

different between pCO

levels. Yamada et al. (2008)

also reported an increase in bacterial production, but

they used very high pCO

levels ranging between

2000 and 10 000 µatm.

Although there is some evidence that elevated CO

affects some microbial processes such as bacterial

production and degradation, the understanding of the

mechanisms involved is still poor. Extracellular en-

zymes are vital for microbial metabolism, and their

activity could provide clues on the mechanisms

involved. Grossart et al. (2006a) found that the activity

of total protease as well as α- and β-glucosidase was

highest at elevated pCO

levels, but this effect was sta-

tistically significant only for protease activity. Similar

results were reported by Piontek et al. (2010), with

higher rates of extracellular glucosidases at lower pH.

Tanaka et al. (2008) reported that the specific glucose

affinity of Bacteria at 3 pCO

levels was similar.

Kranz et al. (2009) investigated the external carbonic

anhydrase (eCA) activities of the cyanobacterium Tri-

298

Liu et al.: Ocean acidification effects on microbes

chodesmium erythraeum in response to pCO

levels of

150, 370 and 1000 µatm. eCA is an enzyme which pro-

motes the conversion of HCO

–

ions to CO

. They

reported low activities of eCA, which did not change as

a function of pCO

, indicating a minor role of eCA in

the carbon acquisition of this species.

Among the factors which determine the conse-

quences of bacterial DOC consumption are the rate at

which biomass is produced (bacterial carbon assimila-

tion) and the rate at which DOC is converted into CO

(bacterial respiration). However, bacterial respiration

and growth efficiency have not yet been studied with

respect to ocean acidification. Also, there are no data

on bacterial production rates estimated as cell division

rates using incorporation of tritiated thymidine. Thus,

it is not known whether pCO

changes result in differ-

ences of the 2 main methods used for estimating bac-

terial production, i.e. leucine and thymidine incor-

poration. Such differences are known under stress

situations such as UVB exposure.

Organic carbon consumption and loss

Piontek et al. (2010) recently tested the effect of ocean

acidification on the degradation activity of marine

Bacteria in a pH perturbation experiment. Higher loss of

polysaccharides (up to 32%) and POC were found under

lowered pH conditions, which suggested that ocean

acidification could affect the cycling of organic carbon

in the future ocean by weakening the biological carbon

pump and by increasing the respiratory production

of CO

. Riebesell et al. (2007) reported that the commu-

nity consumed up to 39% more DIC at higher pCO

whereas nutrient uptake remained the same. This ex-

cess carbon consumption was associated with higher

loss of organic carbon from the upper layer of the

mesocosm. This has an implication on a variety of marine

biological and biogeochemical processes. In the same

mesocosm experiment, community respiration did not

reveal any clear response to pCO

, neither in terms of

the timing nor of the level of cumulative consumption

for the 24 d of the experiment (Egge et al. 2009).

NUTRIENT CYCLES

Inorganic nutrients such as nitrate and phosphate

are vital for microbial growth. Almost all published

studies have investigated the nitrogen cycle. Tanaka et

al. (2008) investigated the availability of phosphate

for phytoplankton and Bacteria at 3 pCO

levels

(350 µatm: 1×CO

; 700 µatm: 2×CO

; 1050 µatm:

3×CO

). Its response was similar to that of the total

particulate phosphorus concentration and phosphate

turnover time. The phosphate transferred to the

>10 µm fraction was greater in the 3×CO

mesocosm

during the first 6 to 10 d when the phosphate concen-

tration was still high. Also, the lower availability of

inorganic nutrients after the phytoplankton bloom

reduced the bacterial capacity to consume labile DOC.

The specific alkaline phosphatase activity (APA) of

Bacteria tended to be higher at 3×CO

than at 2× and

1×CO

during the phosphate depletion period.

Nitrogen fixation

Diazotrophic cyanobacteria affect marine ecosystems

by providing reactive nitrogen to otherwise nitrogen-

limited regions. Most effect sizes for heterocystous

cyanobacteria are above 1 (Fig. 1; Barcelos e Ramos et

al. 2007, Hutchins et al. 2007, Levitan et al. 2007, Fu et

al. 2008, Kranz et al. 2009), whereas they are below 1

for the non-heterocystous species Nodularia spumi-

gena (Czerny et al. 2009). The SMDs of nitrogen fixa-

tion between high and control pCO

are significantly

different from 0 with the fixed effect model and the

random effect model (Table 1). The SMD between con-

trol and low pCO

is significantly different from 0 only

with the fixed effect model.

The filamentous non-heterocystous cyanobacterium

Trichodesmium spp. thrives in oligotrophic areas of

tropical and subtropical seas. This group contributes

about half of all marine N

fixation (Mahaffey et al.

2005). The effect of pCO

levels ranging from 140 to

850 µatm revealed that rates of N

fixation per unit of

phosphorus utilisation more than doubled at high CO

(Barcelos e Ramos et al. 2007). In 2 other studies, very

similar results were obtained (Hutchins et al. 2007,

Levitan et al. 2007). Relative to ambient or low pCO

high pCO

levels enhanced N

fixation as well as the

filament length and biomass of Trichodesmium (Levi-

tan et al. 2007). For example, N

fixation increased by

121% between 400 and 900 µatm. Hutchins et al.

(2007) reported a significant increase of N

fixation and

growth rate between 380 and 1500 µatm, with an

increase of 63% at a pCO

of 750 µatm. Kranz et al.

(2009) also reported a stimulation in nitrogen fixation

of T. erythraeum by almost 40% at a pCO

1000 µatm relative to 370 µatm. Only 1 study is avail-

able in a cyanobacterium other than Trichodesmium.

The unicellular diazotroph Crocosphaera watsonii

revealed N

fixation rates that were enhanced by 40%

at 750 µatm relative to that at 380 µatm (Fu et al. 2008).

It seems clear that CO

is limiting nitrogen fixation

and that ocean acidification could substantially in-

crease, at least in short-term experiments, the fixation

of N

(and CO

as reviewed above) of Trichodesmium.

Hutchins et al. (2009) pointed out that the magnitude

299

Aquat Microb Ecol 61: 291–305, 2010

of the response of nitrogen and carbon fixation to ele-

vated pCO

is the largest physiological response yet

reported for marine microbes. This could fundamen-

tally alter the N and C cycles. It is important to note

that the rate of N

fixation did not continue to rise as

pCO

levels were further elevated above 750 µatm.

This suggests that the observed increase of N

fixation

by Trichodesmium might level off by the end of the

century (Hutchins et al. 2009).

In contrast to the non-heterocystous cyanobacteria

mentioned above, the heterocystous, bloom-forming

diazotroph Nodularia spumigena showed a slight de-

crease of N

fixation rate at increasing pCO

levels

(Czerny et al. 2009).

Nitrification

Nitrification, another important process in the nitro-

gen cycle, is the biological oxidation of ammonia with

oxygen into nitrite followed by the oxidation of these

nitrites into nitrates. Huesemann et al. (2002) investi-

gated the effects of CO

-induced pH changes on

marine nitrification in the context of deep-sea CO

dis-

posal. They found that the rate of nitrification drops

drastically with decreasing pH. Relative to the rates at

pH 8 (presumably on the NBS scale), nitrification

decreased by ca. 50% at pH 7 and by more than 90%

at pH 6.5, while it was completely inhibited at pH 6.0.

ELEMENTAL RATIOS

Since the growth efficiency of heterotrophic micro-

bes is controlled by the quality of food, any change in

the elemental composition of particulate or dissolved

organic matter (POM or DOM) could directly or indi-

rectly affect processes such as growth rate, respiration

and nutrient recycling (Engel et al. 2005). The effect

sizes on the C:N ratio are distributed around 1 for all

studies but 1 (Fig. 1). Nevertheless, the SMD is signifi-

cantly different from 0 with the fixed effect model

between the high and control pCO

and between the

control and low pCO

(Table 1). The effect sizes on the

C:P ratio do not seem to follow a consistent pattern

(Fig. 1). The SMD between the high and control pCO

is not significantly different from 0 with both the fixed

and random effect models but is statistically different

from 0 between the control and low pCO

(Table 1).

Finally, the effect size of the N:P ratio also does not

seem to follow a coherent pattern in the 5 studies avail-

able to date (Fig. 1), but the SMD is nevertheless statis-

tically significant with the fixed effect model between

the high and control pCO

and between the control

and low pCO

(Table 1).

Significant changes in the consumption ratio of vari-

ous inorganic nutrients in response to increasing pCO

were reported by Tortell et al. (2002). They reported

that a pelagic community of the Equatorial Pacific con-

sumed NO

–

and H

SiO

in a ratio close to 1:1 (range

from 1.0 to 1.55) in the high CO

(750 µatm) treatment,

while the consumption ratio was much higher (2.16 to

2.71) at a low pCO

of 150 µatm. In contrast, the as-

similation of nitrate and phosphate was similar in the

3CO

treatments investigated by Engel et al. (2005) in

a mesocosm experiment. The concentration of parti-

culate constituents was highly variable among the

replicate mesocosms, likely disguising direct CO

related effects.

Changes in the C:N:P ratios have been shown in cul-

tures of several eukaryotic phytoplankton groups

(Riebesell 2004). Differential effects were observed on

the 2 main cyanobacterial groups: Synechococcus and

Prochlorococcus. Fu et al. (2007) reported that Syne-

chococcus sp. had a slight decrease in C:N and an in-

crease in C:P and N:P at the higher CO

concentration,

while there was no significant difference in Prochloro-

coccus sp. As mentioned in the previous section, pCO

has an effect on the N

uptake and C fixation rates of

some nitrogen-fixing cyanobacteria. This could signifi-

cantly affect the C:N and N:P ratios, but the effect

found in the various experiments is not consistent. Ele-

vated pCO

increased the C:N ratio in 3 studies (Levi-

tan et al. 2007, Czerny et al. 2009, Kranz et al. 2009)

but had no effect in another study (Barcelos e Ramos et

al. 2007). The N:P ratio of Trichodesmium sp. increases

at elevated pCO

(Barcelos e Ramos et al. 2007),

whereas the N:P ratio of Nodularia spumigena

decreases (Czerny et al. 2009).

Changes in elemental ratios were also reported at

the community level and could have a considerable

impact on the strength of the biological pump. For

example, in a mesocosm experiment, Riebesell et al.

(2007) found that the stoichiometry of C:N drawdown

increased from 6 at low CO

to 8 at high CO

, thus

exceeding the Redfield C:N ratio of 6.6 in today’s

ocean. These authors speculated that this could trans-

late to an excess CO

sequestration potential, through

the biological carbon pump, of 116 Pg C until 2100. In-

creasing C:N ratios would also lower the nutritional

value of primary-produced organic matter, which may

affect the efficiency of bacterial degradation and

zooplankton reproduction. In the same mesocosm

experiment, Bellerby et al. (2008) also found that the

cumulative C:N and C:P ratios of organic production

until the height of the bloom decreased with increas-

ingpCO

. The C:N:P ratios were 1:6.3:121 at 350 µatm,

1:7.1:144 at 700 µatm and 1:8.25:168 at 1050 µatm.

Other studies also reported species-dependent

changes of phytoplankton C:N:P ratios at elevated

300

Liu et al.: Ocean acidification effects on microbes

pCO

(Burkhardt & Riebesell 1997, Burkhardt et al.

1999, Tortell et al. 2000).

MORTALITY

Viral lysis and grazing are the 2 main factors of

microbial mortality. Lysis transfers lysis products such

as the cell content (including viruses) and cell debris

into the DOM pool. This viral shunt (Wilhelm & Suttle

1999) increases bacterial production and respiration

and enhances nutrient recycling. As mentioned above,

grazing on Bacteria transfers organic matter back to

the food web, whereas grazing on phytoplankton is the

first step in the grazing food chain, which also plays a

key role in carbon cycling. Thus, the relative contribu-

tion of viral lysis and grazing is an important factor for

shaping the fate of primary production and bacterial

production. Not much is known about how changes in

pCO

could influence mortality.

Rochelle-Newall et al. (2004) reported that elevated

pCO

had no effect on total viral abundance (VA),

thus suggesting that viral lysis was not influenced

strongly. Viral lysis rates have not yet been measured

in combination with pCO

changes. For grazing, more

information is available. Riebesell et al. (2007) noted

that the DIC consumption increased with increasing

pCO

, whereas the nutrient uptake remained stable.

This leads to an offset in the Redfield ratios, and pos-

sibly causes a deterioration of the food quality. Veloza

et al. (2006) reported that some microzooplankton

groups, e.g. some dinoflagellates, may have the capa-

city to use low quality prey. However, it is still unclear

how and to what extent this takes place. Until now,

only 2 studies investigated the response of microzoo-

plankton grazing to increasing pCO

levels. In the

PeECE III experiment, grazing was highly dynamic

over time, and no effect of CO

on microzooplankton

grazing was found (Suffrian et al. 2008). Rose et al.

(2009) investigated the potential effects of climate

change variables (temperature and pCO

) on the

trophic dynamics using a shipboard continuous cul-

ture system. They observed increases in both the

abundance and grazing rates of microzooplankton in

the high pCO

treatments (690 µatm) relative to ambi-

ent pCO

treatments (390 µatm).

COMMUNITY COMPOSITION AND DIVERSITY

Changes in community composition in the sense of

differences in the relative abundance of large plankton

groups have been documented for pCO

manipulation

experiments and are basically the result of differences

in the time developments of different groups. Changes

in diversity are considered in the following as changes

of the species composition.

Phytoplankton

Phytoplanktonic diversity was profoundly affected by

elevated pCO

in some but not all perturbation experi-

ments performed at the community level (Allgaier et al.

2008, Paulino et al. 2008). Changes in phytoplankton

diversity also led to changes in bacterial community

structure and subsequently in bacterial activities.

Tortell et al. (2002) provided direct evidence that

concentrations can influence the species com-

position of a marine phytoplankton assemblage. The

phytoplankton assemblage exposed to pCO

levels of

150 and 750 µatm was dominated by diatoms and

Phaeocystis sp. by the end of the experiment, but the

abundance of diatoms decreased by ~50% at low pCO

relative to high pCO

levels, while the abundance of

Phaeocystis increased by about 60%. This shift associ-

ated with a higher ratio of nitrate:silicate (N:Si) and

N:P consumption at low pCO

also suggested that CO

concentrations could potentially influence competition

among marine phytoplankton taxa and affect oceanic

nutrient cycling.

Prokaryotes

Mühling et al. (2006) did not find any changes in the

diversity of Bacteria subject to different pCO

levels in

a CO

perturbation experiment carried out in meso-

cosms. Vega Thurber et al. (2009) reported changes in

the diversity of coral-associated microbiota, which

shifts from a healthy-associated community to a com-

munity often found on diseased corals (e.g. Bacteri-

oidetes, Fusobacteria and Fungi) after a strong decline

in pH (8.1 to 6.7). Additionally, decreased pH as well as

other stressors, such as elevated temperature, nutri-

ents and DOC, led to an increased abundance of genes

involved in virulence, stress resistance and production

of secondary metabolites.

The diversity of free-living Bacteria of pelagic meso-

cosms assessed by community fingerprinting changes

with pCO

, whereas that of attached Bacteria seems to

be independent of pCO

and coupled to the develop-

ment of the phytoplankton bloom (Allgaier et al. 2008).

Viruses and grazers

Larsen et al. (2008) investigated how the virioplank-

ton community responded to increased levels of CO

during the PeECE III mesocosm experiment. Some

301

Aquat Microb Ecol 61: 291–305, 2010

viral populations detected and enumerated by flow

cytometry did not respond to altered CO

levels. No

clear effect was found in the ‘low-fluorescence viruses’

(LFV), ‘medium-fluorescence viruses’ (MFV) and

‘putative large viruses’ (PLV). However, the ‘high-

fluorescence viruses’ (HFV) exhibited a higher maxi-

mum abundance in the 1×CO

than in the 2× and

3×CO

mesocosms, thus suggesting changes in viral

diversity. The abundance of Emiliania huxleyi virus

(EhV) and an unidentified double-stranded DNA

(dsDNA) virus decreased with increasing CO

levels.

Only 1 study investigated the effect of elevated pCO

on the community composition of viruses (Larsen et al.

2008). In their study, 2 specific large dsDNA viruses

(EhV and CeV, infecting the haptophytes E. huxleyi

and Crysochromulina ericina) were identified. Their

results indicate that the change in parameters of the

carbonate chemistry might affect the marine pelagic

food web at the viral diversity level. It also demon-

strated that in order to unravel ecological problems as

to how pCO

and nutrients affect the relationship

between marine algal viruses and their hosts, an effort

to develop molecular markers used to identify both

hosts and viruses is needed. Nothing is known about

the effect of pCO

on the diversity of grazers.

DISCUSSION AND CONCLUSIONS

Microbial processes play an important role in the

functioning and the biogeochemical cycles of marine

ecosystems. Although some obvious effects of ocean

acidification on microbes were detected, their re-

sponse is not always consistent and our understanding

remains poor. There are many gaps and challenges for

future research.

Most data on the effect of ocean acidification on

microbial processes and diversity were gathered in

perturbation experiments carried out in the laboratory

and in mesocosms. Few studies investigated the simul-

taneous effects of ocean acidification and other pertur-

bations. Yet, it is well established that microbial pro-

cesses are greatly affected by changes in temperature

and light, as well as changes in the supply of inorganic

nutrients and organic matter (e.g. Kirchman et al.

2009). It is therefore critical that the interactions

between the carbonate chemistry and other parame-

ters are investigated. Equally critical is the need to use

pCO

gradients rather than only 2 pCO

levels to

determine critical threshold levels for parameterising

biogeochemical models. Open water CO

fertilisation

experiments still seem unrealistic (Lance 2009), but it

is potentially rewarding to take advantage of systems

naturally enriched with CO

such as shallow-water

vents (Hall-Spencer et al. 2008), deep-sea vents

(Inagaki et al. 2006), cold-eddies and upwelling sys-

tems (Feely et al. 2008), which have lower pH and high

pCO

levels compared to ambient water and are thus

highly suitable to study effects of ocean acidification,

although the data interpretation is challenging due to

factors such as advection or migrations.

There is evidence for acclimation to the pH

regula-

tion by a Vibrio strain (Labare et al. 2010). Adaptation,

i.e. adjustment to environmental change by genetic

change, is likely faster in microbes than in multi-cellu-

lar marine organisms. This is due to their short genera-

tion time of a few days, which allows for thousands of

generations by 2100, hence increasing the accumula-

tion of mutations, and, at least for prokaryotes, due to

more efficient lateral gene transfer. Most experiments

have been conducted over short periods (days to

weeks), and there is a strong need to carry out longer-

term experiments to detect if adaptation or acclimation

occurs. Genomics, transcriptomics, proteomics and

assessment of the expression of specific marker genes

for crucial functions are among the most promising

methods that are or soon will be available to tackle

these problems.

So far, most studies have investigated individual spe-

cies. More research is needed at multi-species and

community scales. Losses of diversity in the sense of

extinction of species are unlikely for free-living micro-

organisms. However, a large body of research supports

the idea that free-living microbial taxa exhibit biogeo-

graphic patterns (Martiny et al. 2006). Recently, the so-

called ‘rare biosphere’ was detected, i.e. bacterial phy-

lotypes which only occur in low abundance (Sogin et

al. 2006) and may serve as seed banks available for

adaptation to environmental changes (such as increas-

ing pCO

) at the species level. Such questions can be

addressed by large-scale sequence approaches such as

high-throughput DNA sequencing. The decrease in

costs is making this approach affordable.

Until now, no standard protocols have been available

to manipulate the carbonate chemistry during pertur-

bation experiments and no guidelines have been avail-

able for data reporting. Consequently, some of the

experiments published to date are difficult to interpret,

for example due to inadequate pCO

levels or the lack

of information in the data reporting, which seriously

hampers comparative studies and meta-analyses. The

publication of the ‘Guide for best practices on ocean

acidification research and data reporting’ (Riebesell et

al. 2010) will hopefully lead to better experimental set

ups and reporting in future publications.

Joint et al. (in press) pointed out that ocean pH is

variable on short time scales in surface waters and also

as a function of depth. They noted that microbial pro-

cesses continue at depths where pH reaches values

projected in the surface ocean in 2100, asked whether

302

Liu et al.: Ocean acidification effects on microbes

‘microbial assemblages will continue to function at the

lower pH values that are projected for the near future’,

and suggested a null hypothesis that biogeochemical

processes other than calcification will not be funda-

mentally different in a high-CO

ocean. Meta-analysis

is the right tool to test the null hypothesis that ocean

acidification will have no effect on microbial processes.

Although it has not proven to be of great use for many

variables because of the low sample sizes, notable

exceptions are nitrogen fixation, cyanobacterial photo-

synthesis and, to a lesser extent, elemental ratios. This

review and analysis therefore suggests that it is

unlikely that any microbial process will cease to func-

tion due to ocean acidification and that the null

hypothesis of Joint et al. (in press) can be rejected. The

rates of several processes will be affected by ocean

acidification, some positively, others negatively.

Another outcome of our meta-analysis is that the

response of almost all parameters to ocean acidifica-

tion is heterogenous among studies, suggesting the

occurrence of confounding effects. There is no doubt

that the launch of major national and international pro-

jects on ocean acidification will considerably increase

the number of datasets available over the next few

years and will lead to more solid conclusions on the

effect of ocean acidification on microbial processes.

Acknowledgements. The organisers of the 11th Symposium

on Aquatic Microbial Ecology held in Piran, Slovenia, in 2009

kindly invited us to participate. Thanks to A. M. Nisumaa for

invaluable help with data compilation and production of

Fig. 1. This work is a contribution to the European Project on

Ocean Acidification (EPOCA), which received funding from

the European Community’s Seventh Framework Programme

(FP7/2007-2013) under grant agreement no. 211384, the ANR

projects AQUAPHAGE and MAORY of the French Research

Ministry and the FR-EPOCA project of the Institut Polaire

Français Paul Émile Victor (IPEV).

LITERATURE CITED

Allgaier M, Riebesell U, Grossart HP (2008) Coupling of het-

erotrophic bacteria to phytoplankton bloom development

at different pCO

levels: a mesocosm study. Biogeo-

sciences 5:1007–1022

Azam F, Fenchel T, Field JG, Gray JS, Meyer Reil LA,

Thingstad F (1983) The ecological role of water-column

microbes in the sea. Mar Ecol Prog Ser 10:257–263

Badger MR, Price GD (2003) CO

concentrating mechanisms

in cyanobacteria: molecular components, their diversity

and evolution. J Exp Bot 54:609–622

Badger MR, Price GD, Long BM, Woodger FJ (2006) The envi-

ronmental plasticity and ecological genomics of the

cyanobacterial CO

concentrating mechanism. J Exp Bot

57:249–265

Barcelos e Ramos J, Biswas H, Schulz KG, LaRoche J, Riebe-

sell U (2007) Effect of rising atmospheric carbon dioxide

on the marine nitrogen fixer Trichodesmium. Glob Bio-

geochem Cycles 21:GB2028 doi:2010.1029/2006GB002898

Beardall J, Giordano M (2002) Ecological implications of

microalgal and cyanobacterial CCMs and their regulation.

Funct Plant Biol 29:335–347

Beardall J, Stojkovic S, Larsen S (2009) Living in a high CO

world: impacts of global climate change on marine phyto-

plankton. Plant Ecol Divers 2:191–205

Bellerby RGJ, Schulz KG, Riebesell U, Neill C and others

(2008) Marine ecosystem community carbon and nutrient

uptake stoichiometry under varying ocean acidification

during the PeECE III experiment. Biogeosciences 5:

1517–1527

Booth IR (1985) Regulation of cytoplasmic pH in bacteria.

Microbiol Rev 49:359–378

Borenstein M, Hedges LV, Higgins JPT, Rothstein HR (2009)

Introduction to meta-analysis. John Wiley & Sons, Chich-

ester

Burkhardt S, Riebesell U (1997) CO

availability affects ele-

mental composition (C:N:P) of the marine diatom Skele-

tonema costatum. Mar Ecol Prog Ser 155:67–76

Burkhardt S, Riebesell U, Zondervan I (1999) Effects of

growth rate, CO

concentration, and cell size on the stable

carbon isotope fractions in marine phytoplankton. Geo-

chim Cosmochim Acta 63:3729–3741

Burkhardt S, Amoroso G, Riebesell U, Sültemeyer D (2001)

and HCO

–

uptake in marine diatoms acclimated

to different CO

concentrations. Limnol Oceanogr 46:

1378–1391

Burns R, MacDonald CD, McGinn PJ, Campbell DA (2005)

Inorganic carbon repletion disrupts photosynthetic accli-

mation to low temperature in the cyanobacterium Syne-

chococcus elongatus. J Phycol 41:322–334

Caldeira K, Wickett ME (2003) Anthropogenic carbon and

ocean pH. Nature 425:365

Cavicchioli R (2002) Extremophiles and the search for

extraterrestrial life. Astrobiology 2:281–292

Czerny J, Barcelos e Ramos J, Riebesell U (2009) Influence of

elevated CO

concentrations on cell division and nitrogen

fixation rates in the bloom-forming cyanobacterium Nodu-

laria spumigena. Biogeosciences 6:1865–1875

Delille B, Harley J, Zondervan I, Jacquet S and others (2005)

Response of primary production and calcification to

changes of pCO

during experimental blooms of the coc-

colithophorid Emiliania huxleyi. Glob Biogeochem Cycles

19:GB2023 doi:10.1029/2004GB002318

Dickson A (2010) The carbon dioxide system in seawater:

equilibrium chemistry and measurements. In: Riebesell U,

Fabry VJ, Hansson L, Gattuso JP (eds), Guide to best prac-

tices for ocean acidification research and data reporting.

Publications Office of the European Union, Luxembourg,

p 17–40

Dupont S, Dorey N, Thorndyke M (2010) What meta-analysis

can tell us about vulnerability of marine biodiversity to

ocean acidification? Estuar Coast Shelf Sci 89:182–185

Egge JK, Thingstad TF, Larsen A, Engel A, Wohlers J,

Bellerby RGJ, Riebesell U (2009) Primary production dur-

ing nutrient-induced blooms at elevated CO

concentra-

tions. Biogeosciences 6:877–885

Engel A (2000) The role of transparent exopolymer particles

(TEP) in the increase in apparent particles stickiness (α)

during the decline of a diatom bloom. J Plankton Res

22:485–497

Engel A (2002) Direct relationship between CO

uptake and

transparent exopolymer particle production in natural

phytoplankton. J Plankton Res 24:49–53

Engel A, Delille B, Jacquet S, Riebesse U, Rochelle-Newall E,

Terbrüggen A, Zondervan I (2004) Transparent exopoly-

mer particles and dissolved organic carbon production by

303

➤

➤➤

➤

➤➤

➤

Aquat Microb Ecol 61: 291–305, 2010

Emiliania huxleyi exposed to different CO

concentra-

tions: a mesocosm experiment. Aquat Microb Ecol 34:

93–104

Engel A, Zondervan I, Aerts K, Beaufort L and others (2005)

Testing the direct effect of CO

concentration on a bloom

of the coccolithophorid Emiliania huxleyi in mesocosm

experiments. Limnol Oceanogr 50:493–507

Engel A, Schulz K, Riebesell U, Bellerby RGJ, Delille B,

Schartau M (2008) Effects of CO

on particle size distribu-

tion and phytoplankton abundance during a mesocosm

bloom experiment (PeECEII). Biogeosciences 5:509–521

Feely RA, Sabine CL, Hernandez-Ayon JM, Ianson D, Hales B

(2008) Evidence for upwelling of corrosive ‘acidified’

water onto the continental shelf. Science 320:1490–1492

Fu FX, Warner ME, Zhang Y, Feng Y, Hutchins DA (2007)

Effects of increased temperature and CO

on photo-

synthesis, growth, and elemental ratios in marine Syne-

chococcus and Prochlorococcus (Cyanobacteria). J Phycol

43: 485–496

Fu FX, Mulholland MR, Garcia NS, Beck A and others (2008)

Interactions between changing pCO

, N

fixation, and

Fe limitation in the marine unicellular cyanobacterium

Crocosphaera. Limnol Oceanogr 53:2472–2484

Gates S (2002) Review of methodology of quantitative reviews

using meta-analysis in ecology. J Anim Ecol 71:547–557

Gattuso JP, Lavigne H (2009) Technical note: approaches and

software tools to investigate the impact of ocean acidifica-

tion. Biogeosciences 6:2121–2133

Giordano M, Beardall J, Raven JA (2005) CO

concentrating

mechanisms in algae: mechanisms, environmental modu-

lation, and evolution. Annu Rev Plant Biol 56:99–131

Grossart HP, Allgaier M, Passow U, Riebesell U (2006a) Test-

ing the effect of CO

concentration on dynamics of marine

heterotrophic bacterioplankton. Limnol Oceanogr 51:1–11

Grossart HP, Czub G, Simon M (2006b) Algae–bacteria inter-

actions and their effects on aggregation and organic mat-

ter flux in the sea. Environ Microbiol 8:1074–1084

Hall-Spencer JM, Rodolfo-Metalpa R, Martin S, Ransome E

and others (2008) Volcanic carbon dioxide vents show

ecosystem effects of ocean acidification. Nature 454:96–99

Hein M, Sand-Jensen K (1997) CO

increases oceanic primary

production. Nature 388:526–527

Hendriks IE, Duarte CM, Alvarez M (2010) Vulnerability of

marine biodiversity to ocean acidification: a meta-analy-

sis. Estuar Coast Shelf Sci 86:157–164

Huesemann MH, Skillman AD, Crecelius EA (2002) The inhi-

bition of marine nitrification by ocean disposal of carbon

dioxide. Mar Pollut Bull 44:142–148

Hutchins DA, Fu FX, Zhang Y, Warner ME and others (2007)

control of Trichodesmium N

fixation, photosynthesis,

growth rates, and elemental ratios: implications for

past, present, and future ocean biogeochemistry. Limnol

Oceanogr 52: 1293–1304

Hutchins DA, Mulholland MR, Fu FX (2009) Nutrient cycles

and marine microbes in a CO

-enriched ocean. Oceanog-

raphy 22:128–145

Inagaki F, Kuypers MMM, Tsunogai U, Ishibashi JI and others

(2006) Microbial community in a sediment-hosted CO

lake of the southern Okinawa Trough hydrothermal sys-

tem. Proc Natl Acad Sci USA 103:14164–14169

Joint I, Doney SC, Karl DM (in press) Will ocean acidification

affect marine microbes? ISME J, doi:10.1038/ismej.

2010.79

Kim JM, Lee K, Shin K, Kang JH and others (2006) The effect

of seawater CO

concentration on growth of natural

phytoplankton assemblage in a controlled mesocosm

experiment. Limnol Oceanogr 51: 1629–1636

Kirchman DL (2008) Microbial ecology of the oceans, 2nd

edn. John Wiley & Sons, Hoboken, NJ

Kirchman DL, Moran XA, Ducklow H (2009) Microbial

growth in the polar oceans — role of temperature and

potential impact of climate change. Nat Rev Microbiol 7:

451–459

Kranz SA, Sültemeyer D, Richter KU, Rost B (2009) Carbon

acquisition by Trichodesmium: the effect of pCO

and

diurnal changes. Limnol Oceanogr 54:548–559

Labare MP, Bays JT, Butkus MA, Snyder-Leiby T and others

(2010) The effects of elevated carbon dioxide levels on a

Vibrio sp. isolated from the deep-sea. Environ Sci Pollut

Res Int 17:1009–1015

Lance V (2009) Carbon productivity responses to increased

dissolved inorganic carbon concentrations in surface ocean:

exploring the feasibility of an in situ carbon addition exper-

iment. Workshop report. LDEO in situ CO

Workshop, 23–

24 March 2009, Lamont-Doherty Earth Observatory, New

York, NY. Available at www.ldeo.columbia.edu/~vlance/

CO2WorkshopReportonWEB_16Jun09.pdf

Larsen JB, Larsen A, Thyrhaug R, Bratbak G, Sandaa RA

(2008) Response of marine viral populations to a nutrient

induced phytoplankton bloom at different pCO

levels.

Biogeosciences 5:523–533

Lavigne H, Gattuso JP (2010) seacarb: seawater carbonate

chemistry with R. R package version 2.3.3. Available at:

http://CRAN.R-project.org/package=seacarb

Levitan O, Rosenberg G, Setlik I, Setlikova E and others

(2007) Elevated CO

enhances nitrogen fixation and

growth in the marine cyanobacterium Trichodesmium.

Glob Change Biol 13: 531–538

Logan BE, Passow U, Alldredge AL, Grossart HP, Simon M

(1995) Rapid formation and sedimentation of large aggre-

gates is predictable from coagulation rates (half-lives) of

transparent exopolymer particles (TEP). Deep-Sea Res II

42:203–214

Mahaffey C, Michaels AF, Capone DG (2005) The conundrum

of marine N

fixation. Am J Sci 305:546–595

Mari X (2008) Does ocean acidification induce an upward flux

of marine aggregates? Biogeosciences 5:1023–1031

Martiny JBH, Bohannan BJM, Brown JH, Colwell RK and oth-

ers (2006) Microbial biogeography: putting microorgan-

isms on the map. Nat Rev Microbiol 4:102–112

Mühling M, Fuller NJ, Somerfield PJ, Post AF and others (2006)

High resolution genetic diversity studies of marine Syne-

chococcus isolates using rpoC1-based restriction fragment

length polymorphism. Aquat Microb Ecol 45: 263–275

Nelson NB, Seigel DA, Michaels AF (1998) Seasonal dynam-

ics of colored dissolved organic matter in the Sargasso

Sea. Deep-Sea Res I 45:931–957

Nisumaa AM, Pesant S, Bellerby RGJ, Middelburg JJ and

others (2010) EPOCA/EUR-OCEANS data compilation on

the biological and biogeochemical responses to ocean

acidification. Earth Syst Sci Data 2:167–175

Orr JC, Fabry VJ, Aumont O, Bopp L and others (2005)

Anthropogenic ocean acidification over the twenty-first

century and its impact on calcifying organisms. Nature

437:681–686

Partensky F, Hess WR, Vaulot D (1999) Prochlorococcus, a

marine photosynthetic prokaryote of global significance.

Microbiol Mol Biol Rev 63:106–127

Passow U (2002) Transparent exopolymer particles (TEP) in

aquatic environments. Prog Oceanogr 55:287–333

Paulino AI, Egge JK, Larsen A (2008) Effects of increased

atmospheric CO

on small and intermediate sized osmo-

trophs during a nutrient induced phytoplankton bloom.

Biogeosciences 5:739–748

304

➤

➤➤

➤

➤➤

➤

➤➤

➤

➤➤

➤

➤➤

➤

➤➤

➤

Liu et al.: Ocean acidification effects on microbes

Piontek J, Lunau M, Händel N, Borchard G, Wurst M, Engel

A (2010) Acidification increases microbial polysaccharide

degradation in the ocean. Biogeosciences 7:1615–1624

Riebesell U (2004) Effects of CO

enrichment on marine

phytoplankton. J Oceanogr 60:719–729

Riebesell U, Schulz KG, Bellerby RGJ, Botros M and others

(2007) Enhanced biological carbon consumption in a high

ocean. Nature 450:545–548

Riebesell U, Bellerby RGJ, Grossart HP, Thingstad F (2008)

Mesocosm CO

perturbation studies: from organism to

community level. Biogeosciences 5:1157–1164

Riebesell U, Fabry V, Hansson L, Gattuso JP (2010) Guide to

best practices for ocean acidification research and data

reporting. Publications Office of the European Union, Lux-

embourg

Rochelle-Newall EJ, Fisher TR, Fan CL, Glibert PM (1999)

Dynamics of chromophoric dissolved organic matter and

dissolved organic carbon in experimental mesocosms. Int

J Remote Sens 20:627–641

Rochelle-Newall E, Delille B, Frankignoulle M, Gattuso JP

and others (2004) Chromophoric dissolved organic matter

in experimental mesocosms maintained under different

pCO

levels. Mar Ecol Prog Ser 272:25–31

Rose JM, Feng YY, Gobler CJ, Gutierrez R, Hare CE, Leblanc

K, Hutchins DA (2009) Effects of increased pCO

and tem-

perature on the North Atlantic spring bloom. II. Microzoo-

plankton abundance and grazing. Mar Ecol Prog Ser

388:27–40

Sabine CL, Feely RA, Gruber N, Key RM and others (2004)

The oceanic sink for anthropogenic CO

. Science 305:

367–371

Santana-Casiano JM, González-Dávila M, Rueda MJ, Llinás

O, González-Dávila EF (2007) The interannual variability

of oceanic CO

parameters in the northeast Atlantic sub-

tropical gyre at the ESTOC site. Glob Biogeochem Cycles

21:GB1015 doi:10.1029/2006GB002788

Schulz KG, Riebesell U, Bellerby RGJ, Biswas H and others

(2008) Build-up and decline of organic matter during

PeECE III. Biogeosciences 5:707–718

Schwarzer G (2010) meta: meta-analysis with R. R package

version 1.1-8. Available at: http://CRAN.R-project.org/

package=meta

Sogin ML, Morrison HG, Huber JA (2006) Microbial diversity

in the deep sea and the underexplored ‘rare biosphere’.

Proc Natl Acad Sci USA 103:12115–12120

Solomon S, Qin D, Manning M, Alley RB and others (2007)

Technical summary. In: Solomon S, Qin D, Manning M,

Chen Z and others (eds) Climate change 2007: the physi-

cal science basis. Contribution of Working Group I to the

4th Assessment Report of the Intergovernmental Panel on

climate change. Cambridge University Press, Cambridge

Steinacher M, Joos F, Frölicher TL, Plattner GK, Doney SC

(2009) Imminent ocean acidification projected with the

NCAR global coupled carbon cycle-climate model. Bio-

geosciences 6:515–533

Suffrian K, Simonelli P, Nejstgaard JC, Putzeys S, Carotenuto

Y, Antia AN (2008) Microzooplankton grazing and phyto-

plankton growth in marine mesocosms with increased

levels. Biogeosciences 5:1145–1156

Takeuchi K, Fujioka Y, Kawasaki Y, Shirayama Y (1997)

Impacts of high concentration of CO

on marine organ-

isms; a modification of CO

ocean sequestration. Energy

Convers Manag 38(Suppl 1):S337–S341

Tanaka T, Thingstad TF, Løvdal T, Grossart HP and others

(2008) Availability of phosphate for phytoplankton and

bacteria and of glucose for bacteria at different pCO

lev-

els in a mesocosm study. Biogeosciences 5:669–678

Tortell PD, Rau GH, Morel FMM (2000) Inorganic carbon

acquisition in coastal Pacific phytoplankton communities.

Limnol Oceanogr 45:1485–1500

Tortell PD, DiTullio GR, Sigman DM, Morel FMM (2002) CO

effects on taxonomic composition and nutrient utilization

in an Equatorial Pacific phytoplankton assemblage. Mar

Ecol Prog Ser 236:37–43

Vega Thurber R, Willner-Hall D, Rodriguez-Mueller B,

Desnues C and others (2009) Metagenomic analysis of

stressed coral holobionts. Environ Microbiol 11:2148–2163

Veloza AJ, Chu FLE, Tang KW (2006) Trophic modification of

essential fatty acids by heterotrophic protists and its

effects on the fatty acid composition of the copepod Acar-

tia tonsa. Mar Biol 148:779–788

Wilhelm SW, Suttle CA (1999) Viruses and nutrient cycles in

the sea. Bioscience 49:781–788

Yamada N, Suzumura M, Tsurushima N, Harada K (2008)

Impact on bacterial activities of ocean sequestration of

carbon dioxide into bathypelagic layers. In: OCEANS’08

MTS/IEEE Kobe-Techno-Ocean’08 - Voyage toward the

Future, OTO’08, 8–11 April, Kobe, p 1–3

305

Submitted: March 29, 2010; Accepted: August 12, 2010 Proofs received from author(s): October 4, 2010

➤

➤➤

➤

➤➤

➤

➤➤

➤

➤➤

➤