R E S E A R C H A R T I C L E

Verticalmigration, nitrateuptakeanddenitri¢cation: survivalmechanismsofforaminifers (Globobulimina turgida)under lowoxygen conditionsKaroliina A. Koho1, Elisa Pina-Ochoa2, Emmanuelle Geslin3,4 & Nils Risgaard-Petersen2

1Faculty of Geosciences, Utrecht University, Utrecht, The Netherlands; 2Center for Geomicrobiology, Institute of Biological Sciences, Aarhus University,

Aarhus C, Denmark; 3Laboratory of Recent and Fossil Bio-Indicators, Angers University, Angers, France; and 4LEBIM, Ile d’Yeu, France

Correspondence: Karoliina A. Koho, Faculty

of Geosciences, Utrecht University,

Aardwetenschappen Building, Budapestlaan

4, 3584 CD Utrecht, The Netherlands. Tel.:

131 0 30 253 5170; fax: 131 0 30 253 5302;

e-mail: koho@geo.uu.nl

Received 16 July 2010; revised 1 November

2010; accepted 2 November 2010.

Final version published online 13 December

2010.

DOI:10.1111/j.1574-6941.2010.01010.x

Editor: Riks Laanbroek

Keywords

benthic foraminifera; denitrification; nitrate

uptake; migration; pseudopodia.

Abstract

15NO3� isotope labelling experiments were performed to investigate foraminiferal

nitrate uptake strategies and the role of pseudopodial networks in nitrate uptake.

Globobulimina turgida were placed below the nitrate penetration depth in

homogenized sediment cores incubated in artificial seawater containing 15NO3�. A

nylon net prevented the vertical migration of foraminifera to strata containing

nitrate and oxygen, but allowed potential access to such strata by extension of

pseudopods. No 15NO3� was found in G. turgida in these cores, suggesting that

foraminifera cannot extend their pseudopods for nitrate uptake through several

millimetres of sediment, but must physically migrate upwards closer to nitrate-

containing strata. However, foraminiferal migration patterns in control cores with

no nylon net were erratic, suggesting that individuals move in random orientations

until they find favourable conditions (i.e. free nitrate or oxygen). A second

experiment showed that foraminifera actively collect nitrate both in the presence

and in the absence of oxygen, although uptake was initiated faster if oxygen was

absent from the environment. However, no systematic influence of the size of the

intracellular nitrate pool on nitrate uptake was observed, as specimens containing

a large range of intracellular nitrate (636–19 992 pmol per cell) were measured to

take up 15NO3� at comparable rates.

Introduction

Benthic foraminifera, single-celled eukaryotes, inhabit a

large range of marine environments including those low in

oxygen (e.g. Bernhard & Reimers, 1991; Bernhard, 1992;

Gooday et al., 2000). However, the mechanism of low

oxygen tolerance in foraminifera was not comprehended

until recently, when evidence for anaerobic respiration

through the intracellular nitrate pool to dinitrogen gas was

shown in the species Globobulimina turgida [Risgaard-

Petersen et al., 2006 (After a more thorough taxonomic

analysis, Globobulimina pseudospinescens described in refer-

ence Risgaard-Petersen et al., 2006 should be determined as

G. turgida. This distinction was already applied in Pina-

Ochoa et al., 2010a.)]. Furthermore, the recent discovery of

the widespread occurrence of nitrate accumulation and

denitrification among foraminifera has shown these capa-

cities to be present in several different foraminiferal groups

(miliolids, rotaliids, textulariids) and Gromia (Pina-Ochoa

et al., 2010a). Benthic foraminifera are commonly recorded

in the top 10 cm of sediment and their species-specific

vertical distribution, or microhabitat, has been related to

sediment redox chemistry (mainly oxygen) and benthic food

or (labile) organic carbon availability (e.g. Corliss, 1985;

Jorissen et al., 1994; Koho et al., 2008a). Some species,

referred to as deep infaunal foraminifera (e.g. Globobulimi-

na), have been found to occupy niches deep in the sediment,

often below the oxygen penetration depth (Bernhard & Sen

Gupta, 1999; Gooday et al., 2000). However, in very eu-

trophic settings, deep infaunal species are often found solely

close to the sediment–water interface (Koho et al., 2008a).

Thus, the concept of foraminiferal microhabitat is not

fixed and must be adapted depending on the sedimentary

conditions.

FEMS Microbiol Ecol 75 (2011) 273–283 c� 2010 Federation of European Microbiological SocietiesPublished by Blackwell Publishing Ltd. All rights reserved

MIC

ROBI

OLO

GY

EC

OLO

GY

Foraminifera can adjust to changes in geochemistry

(e.g. Duijnstee et al., 2003; Geslin et al., 2004) and/or food

supply (e.g. Ernst et al., 2005; Koho et al., 2008b) by

migrating and moving within the sediment. To move,

foraminifera make use of their extensive reticulose pseudo-

podial network, which can change rapidly in shape in

response to environmental inputs (Travis et al., 2002). The

pseudopods are also involved in vital cell tasks, such as

feeding, prey capture, digestion, exchange and transport of

gases, symbiont recognition and some aspects of respiration

and reproduction (Lipps, 1983; Travis & Bowser, 1991;

Bernhard & Bowser, 1992; Goldstein, 1999; Cedhagen &

Frimanson, 2002). Furthermore, enhanced pseudopodial

activity has been observed around the site of calcification in

benthic foraminifera (L. de Nooijer, pers commun.), sug-

gesting that pseudopods may be involved in taking up

dissolved components in seawater during the building of

foraminiferal tests.

In addition to the physical advantages of the pseudo-

podial network, the intracellular nitrate pool allows

foraminifera a chemical advantage, facilitating respiration

under otherwise ‘hostile conditions’ when the desired elec-

tron acceptors (O2, NO3�) are absent in the environment.

Risgaard-Petersen et al. (2006) estimated that on average

G. turgida can survive anoxia for �25 days if respiring

through the intracellular nitrate pool only. This was

confirmed by a laboratory study of Pina-Ochoa et al.

(2010b), which also showed that some specimens with

very high intracellular nitrate pools can survive anoxia for

56 days.

This study was conducted to advance our understanding

of the role of benthic foraminifera in the nitrogen cycle,

and importantly, to add to the small body of existing

literature on the topic (Risgaard-Petersen et al., 2006;

Høgslund et al., 2008; Glud et al., 2009; Pina-Ochoa et al.,

2010a, b). The first specific aim was to test, by means of a

laboratory experiment, whether it is necessary for foramini-

fera to migrate and reside in the nitrate-rich sediment strata

in order to replenish their nitrate storage, or whether they

can use their pseudopodial network like ‘straws’ to collect

nitrate over long vertical distances (� 5 mm), while residing

well below the nitrate penetration depth. In the experimen-

tal cores, the vertical movement of foraminifera was re-

stricted with a nylon net, preventing migration to strata

containing either free nitrate or oxygen. Thus, to survive,

foraminifera needed to respire on their internal nitrate

supply, or if possible, use their fine structured reticulopodial

network to collect nitrate from above the net. The second

aim of the study was to investigate the relative rates of

nitrate uptake in foraminifera under oxic and anoxic condi-

tions and to establish whether the size of the existing

intracellular nitrate pool influences the rate of further

uptake.

Materials and methods

Sediment collection, processing and isolation ofliving foraminifera

Sediment was collected, using the Olausson-grabber de-

signed to collect intact, cubic samples of the soft bottom

seafloor, onboard of R/V Oscar von Sydow from the deep

basin of Gullmar Fjord, Sweden (Alsback, depth 119 m;

58119.40N; 111 32.70E). The cubic samples were subse-

quently subsampled for the top 10–15 cm of sediment using

plastic tubes of various dimensions (on average +6 cm).

Intact cores were then transported to the laboratory in

cool boxes. The cores were placed in an aquarium filled with

in situ seawater. The aquarium was kept in a climate-

controlled room at �10 1C and the water was aerated with

an aquarium pump, until sediment was needed for the

experiments.

At the start of the experiment, the top 2 cm of the

sediment cores were sliced (at 1 cm intervals) and sieved

over 4 150mm mesh. In addition, the fine fraction

(o 63mm) was stored and saved for later use in the experi-

mental set-up. The sieving was carried out using in situ

seawater. Normally, Globobulimina spp. occur relatively

deep in sediment (e.g. Corliss, 1985, 1991; Fontanier et al.,

2002; Koho et al., 2008a). However, in Gullmar Fjord, this

species appears most abundant in the upper 2 cm of

sediment (Risgaard-Petersen et al., 2006) as a result of the

relatively shallow redox zonation.

Specimens assumed to be alive in the sieve residue

(bearing a good coloration, protoplasm well defined, sedi-

ment gathered around the apertura) were gently picked

into a Petri dish containing Artificial Sea Water (ASW,

Red Sea Salt; salinity �34–35) and a thin layer of fine-

grained sediment (o 38mm; sieved from the same experi-

mental cores). Only the specimens that moved (within

�16 h) and thus produced a burrow or a trace on the

sediment surface were considered living and used further in

the experiment.

Experimental design: set-up and sampling

Special splitable cores (Fig. 1) were designed and built for

the experiments; the main objective of the set-up was to

restrict the migration of foraminifera, with the aid of a nylon

net, into the upper sediment unit containing 15N-labelled

nitrate.

In order to establish the nitrate penetration depth in the

experimental set-up, a test core was made. A core (same

design as that used for the experiment) was filled with fine-

grained, homogenized sediment (o 63 mm; sediment from

sieving residue). The core was then placed in an aerated

aquarium containing 50 mM 15NO3�. Following a 4 24-h

incubation period, the nitrate penetration depth was

FEMS Microbiol Ecol 75 (2011) 273–283c� 2010 Federation of European Microbiological SocietiesPublished by Blackwell Publishing Ltd. All rights reserved

274 K.A. Koho et al.

measured using a Biosensor for NOX� (Larsen et al., 1997)

(see Microsensor techniques: NOx and O2 for more details).

The penetration depth was �0.6 cm (Fig. 2).

After the nitrate penetration depth was established in the

test core, two control cores without a nylon net (C1, C2) and

two experimental cores with a nylon net (N1 or ‘net 1’, N2 or

‘net 2’) were set up (Fig. 1). The lower part of the core was

filled with homogenized, fine-grained sediment (o 63 mm;

sieve residue) and 50 living G. turgida were placed on top.

The nylon net (mesh size 60 mm) was secured on top with

the middle core ring, containing two rubber O-rings. The

relatively large mesh size of 60 mm should not restrict the

movement of the fine-structured pseudopods, which have a

diameter of �1mm (Bowser & Travis, 2002). The upper part

of the core tube was placed on top and secured with an

additional O-ring. 9.6 mL of sediment was then poured on

top of the net to create a 1-cm-thick sediment blanket, thus

containing the nitrate reduction zone. The control cores

were made in the same way, with the exception of the nylon

net. To avoid sediment resuspension and disturbance, cores

were then slowly filled with ASW with the aid of a pipette

(salinity 35, temperature �15 1C, 50 mM 15NO3�) and placed

in an aerated aquarium for 4 weeks. The aquarium con-

tained 50mM 15NO3� at the start of the experiment.

At the end of the experiment, the nitrate and oxygen

penetration depths in the control core were measured using

Fig. 1. Experimental set-up.

Fig. 2. Down-core NO3�profile at the start and end of the experiment. In addition, the O2 pore water profile was measured at the end of the experiment

to clarify the end NO3� profile.

FEMS Microbiol Ecol 75 (2011) 273–283 c� 2010 Federation of European Microbiological SocietiesPublished by Blackwell Publishing Ltd. All rights reserved

275Survival mechanisms of foraminifers

microsensors (see Microsensor techniques: NOx and O2).

Oxygen penetration depth was�0.3 cm and nitrate�0.4 cm

(measured profiles shown in Fig. 2). Hereafter, all cores were

sampled downcore for foraminifera. The control cores (C1,

C2) were sliced at 0.5 cm resolution (top 2 cm) and 1 cm

resolution (2–5 cm). The experimental cores (N1, N2) were

sampled in a similar manner; however, the top 1 cm above

the net was disregarded. The sediment slices were sieved

through a 4 63mm mesh using nitrate-free ASW and

foraminifera were then picked and counted.

As the nitrate penetration depth was only established at

the start and at the end of the experiment, a possible study

limitation may result from the heterogeneity of the sediment

and pore water chemistry within and between the cores.

This issue was addressed at the start of the experiment by

thoroughly homogenizing the sediment used to make the

cores. Furthermore, as only the very fine-grained and

homogenized sediment was applied, the porosity should be

similar in all cores, and thus diffusion of nitrate from

overlaying water into the sediment should be identical in

all cores. Bioturbation was also minimized as all macro- and

meiofauna (4 63mm), except the added foraminifera, were

removed. Finally, the net was placed at a ‘safe’ distance away

from the measured nitrate penetration depth to allow some

potential fluctuation in nitrate penetration to take place, but

still within the reach of the foraminifera, which can extend

their pseudopodial network at least 10 times their own

diameter (Travis & Bowser, 1991).

A fluorescent technique was used to check for the viability

of foraminifera at the end of the experiment, due to its

reliable and fast nature, to identify living individuals in

experimental set-ups (Bernhard et al., 1995; Pina-Ochoa

et al., 2010b). The fluorogenic probe used in this experiment

was fluorescein diacetate (FDA), which was diluted in a

dimethyl sulphoxide solution and ASW in order to obtain a

final solution of 100 mM FDA. About half of the foraminifera

found in cores C1, C2 and N1 were incubated in the ASW-

FDA solution for 3–4 h, and checked for viability. No FDA

check was carried out on core N2 due to time limitations.

After the incubation, foraminifera were picked out and

placed in ASW with no FDA. Each specimen was checked

individually for the fluorescence using an Axiovert 200 M

Apotome epifluorescence microscope (Carl Zeiss, Jena,

Germany).

Finally, foraminifera were placed individually in Eppen-

dorf tubes (Eppendorf AG, Germany) and frozen at

� 80 1C. All the specimens identified as living using the

FDA staining in control cores 1, 2 and the net 1 core were

measured for their internal nitrate pool (see Intracellular

nitrate content) and 15N content (see 15NO3� content in

foraminifera). In the net 2 core, 22 specimens were chosen

randomly for analyses. Before the measurements, all speci-

mens were defrosted and immersed immediately in 100mL

of milli-Q water to extract the intracellular nitrate. Each

foraminifer was crushed to maximize the extraction. Sam-

ples were split, so that 70mL was used for establishing the15N content (see 15NO3

� content in foraminifera) and 30mL

for measuring the intracellular nitrate pool (see Intracellular

nitrate content).

The average living depth (ALD) of foraminifera (G.

turgida) in all the cores was calculated according to Jorissen

et al. (1995). The ALD value was calculated separately for

the total and live (FDA-stained) foraminifera.

Microsensor techniques: NOx and O2

Pore water nitrate content was determined at the start (test

core) and at the end of the experiment (core C1, Fig. 2) with

a NOx� biosensor (Larsen et al., 1997) equipped with

electrophoretic sensivity control to optimize the sensitivity

of the sensor (Kjær et al., 1999). The sensor is based on

bacterial reduction of nitrate and nitrite to N2O, which is

subsequently detected by an electrochemical N2O sensor.

Tip diameter was 120 mm, with a detection limit of 0.5 mM

and a sensitivity of �0.3 pA mM�1. The sensor had a slow

response time of 80 s at 90%. It was two-point calibrated

with ASW with 50 mM nitrate and nitrate-free ASW.

The pore water oxygen profile was measured at the end of

the experiment in the core C1 to clarify the nitrate profile. A

Clark-type oxygen microsenor (Revsbech, 1989) with a tip

diameter of 15 mm, detection limit o 0.5 mM with a re-

sponse time of 1 s at 90%, was two-pointed calibrated in air-

saturated water and anoxic alkaline ascorbate.

The oxygen and NOx� sensors were mounted on a

computer-controlled micromanipulator. Both nitrate and

oxygen profiles were measured downcore at 500mm resolu-

tion.

Nitrate uptake by foraminifera under oxic andanoxic conditions

Nitrate uptake in single foraminifera under oxic and anoxic

conditions was investigated in a separate experiment. Two

series of six Exetainers (Labco Ltd, UK) (six oxic, six anoxic

flushed with He), each containing five living G. turgida, were

filled with ASW (salinity 35) containing �50 mM K15NO3.

Exetainers were sampled at various times (80 min, 7, 23.1,

30, 50.5, 101 h) and foraminifera were picked. Furthermore,

at each sample time, ASW in the Exetainer was sampled to

monitor the nitrate concentration (Fig. 3). These were

filtered over 0.2 mm and frozen at � 20 1C until analysis.

Each foraminifer was rinsed three times in nitrate-free

ASW, placed in an Eppendorf tube, immersed in liquid N2

for � 30 s, followed by the addition of 100 mL of milli-Q

water. Samples were frozen at � 20 1C until analyses of

intracellular nitrate (see Intracellular nitrate content) and15NO3

� content (see 15NO3� content in foraminifera).

FEMS Microbiol Ecol 75 (2011) 273–283c� 2010 Federation of European Microbiological SocietiesPublished by Blackwell Publishing Ltd. All rights reserved

276 K.A. Koho et al.

Intracellular nitrate content

Foraminiferal specimens from the both experiments were

analysed for nitrate content using the VCl3 reduction

method (Braman & Hendrix, 1989) on a chemilumines-

cence detector (Model CLD 86, Eco Physics AG, Duernten,

Switzerland) as described previously by Risgaard-Petersen

et al. (2006) and Høgslund et al. (2008), and with modifica-

tions as specified in Pina-Ochoa et al. (2010a).

15NO3� content in foraminifera

The 15N-atom% of the nitrate extracted from each indivi-

dual foraminifer was measured as described in Risgaard-

Petersen et al. (1993), but modified for small volumes. The

basis of this method is to convert the 15N-labelled NO3�

extracted from foraminifera into 15N-labelled N2 gas by

means of a denitrifying bacterial enrichment culture. In

short, two media for denitrifying bacteria (Pseudomonas

aeroginosa) were prepared in sterile autoclaved serum bot-

tles, containing tryptone soy broth (TSB), 1 mM NH41 and

10 mM NO3� (Medium 1), and TSB and 1 mM NH4

1

(Medium 2). Bacteria were grown in Medium 1 for 2 days,

after which the medium was centrifuged and the super-

natant was removed. The enriched culture (now nitrate free)

was inoculated into Media 2. Two hundred microlitres of

Media 2 was pipetted into 6 mL Exetainers, which were

subsequently flushed with He gas for 5 min to reduce the N2

background. Seventy microlitres of sample extract was then

injected into the Exetainer and flushed immediately with He

for an additional 30 s to remove any traces of atmospheric

nitrogen. The abundance of 15N2 (i.e. 29N2 and 30N2) was

measured by combined GC-MS (RoboPrep-G1 in line with

Tracermass, Europa Scientific) as described by Risgaard-

Petersen & Rysgaard (1995).

Results and discussion

Survival of G. turgida under experimentalconditions

In both the control and the net cores of the migration

experiment, significant portions of the population were still

alive at the end of the experiment (Fig. 4). However, the

proportion of the living specimens was generally higher in

the control cores than in the net cores. Of the FDA-checked

specimens, 71% and 59% were still alive in control 1 (n = 24)

and control 2 (n = 22), whereas only 45% were alive in the

net 1 core (n = 22). In the net 2 core, where viability was

based on specimens containing intracellular nitrate, 55% of

specimens were found alive. These survival rates of G.

turgida under laboratory conditions (especially the control

situation) are similar to those observed in the experiment of

Pina-Ochoa et al. (2010b). It could be argued that such

survival rates after 4 weeks in the net cores were somewhat

higher than expected, as it has been estimated that on

average G. turgida can survive for around 25 days if respiring

from the intracellular nitrate pool only (Risgaard-Petersen

et al., 2006; Pina-Ochoa et al., 2010b).

Intracellular nitrate pool and 15N uptake incontrol vs. net cores

High total intracellular nitrate concentrations (14NO3�115NO3

�)

were measured for G. turgida in both control and net cores

(Fig. 5). The total nitrate concentration in a single living

G. turgida individuals ranged from 0 to 32 541 pmol N per cell.

Furthermore, 87.5% and 92% of FDA stained, living indivi-

duals in the control 1 (n = 18) and the control 2 (n = 14) cores,

respectively, contained intracellular nitrate. In contrast, in the

net 1 core, only 46% (n = 13) were found to contain nitrate. In

the net 2 core, 55% of measured specimens contained nitrate

(n = 22); however, because in this core, the presence of nitrate

was the criterion used to identify living individuals, viability was

not independently checked. These concentrations corresponded

to an average concentration in the control cores of 3929� 4590,

8999� 9023 pmol N per cell (C1 and C2, respectively) and in

the net cores of 4908� 7079 and 5259� 5042 pmol N per cell

(N1 and N2, respectively); note that only the ‘real’ zeros (i.e.

individuals identified as living based on FDA staining, but with

no detectable nitrate, were taken into account in the above

calculations. The measured intracellular nitrate concentrations

are well within the range of values reported for G. turgida

(Risgaard-Petersen et al., 2006; 0–72 000 pmol N per cell, with

an average of 18 000� 5000 n = 20). The presence of large

intracellular nitrate pools explains the high survival rates

observed in all cores. Additionally, it could be speculated that

Fig. 3. Seawater nitrate concentration measured in the Exetainers dur-

ing the short-term nitrate uptake experiment (both oxic and anoxic

concentrations are shown). The analytical error was 3.5%.

FEMS Microbiol Ecol 75 (2011) 273–283 c� 2010 Federation of European Microbiological SocietiesPublished by Blackwell Publishing Ltd. All rights reserved

277Survival mechanisms of foraminifers

foraminifera in the net cores were able to scavenge leaking

nitrate from the dead or dying individuals, contributing to the

high survival rates of those remaining. This theory requires

further investigation.

The dubious occurrence of living G. turgida in the net

core 1 with no measurable intracellular nitrate is puzzling

and will require further study. Ultrastructural investigations

and supravital staining methods of the reticulopodia of

foraminifera may shed more light on the role of mitochon-

drial respiration in foraminiferal pseudopods.

Fifteen of the 32 living individuals in the control cores

were found to contain 15NO3� (Fig. 5). 15NO3

� concentrations

ranged from 0 to 20 130 pmol N per cell, corresponding to

an average concentration of 1078� 2488 and

2946� 5694 pmol N per cell in C1 and C2, respectively.

Among 15NO3�-containing specimens, 37% had a 15N-label-

ling of their extracted NO3�pool 4 30 15N-atom%. Further-

more, most of the 15N-labelled individuals, except two in the

control 1 and one specimen in the control 2 core, were

found in the sediment depth interval 0.5–1 cm, thus close to

the nitrate penetration depth in sediment. In contrast, in the

net cores, no 15NO3�was detected in any of the specimens.

The absence of 15NO3� in foraminifera from the net cores,

and the clear presence of 15NO3� in foraminifera from the

control cores, indicates that foraminifera can replenish their

nitrate pool only if nitrate is available in the surrounding or

very close by (within a few millimetres) environment.

Foraminifera residing deep in the sediment well below the

diffusive boundary layer of nitrate do not seem to be able to

stretch their pseudopods over larger distances (several milli-

metres) into the zone of free nitrate. However, due to the

slight upward migration/fluctuation of pore water nitrate

front during the experiment, it cannot be ruled out that this

limited the foraminifera in the net cores from replenishing

their pools.

Vertical distribution of G. turgida andimplications for migration behaviour

The vertical distribution of G. turgida differed between the

set-ups, being generally shallower in the control cores than

in the net cores (Fig. 4). This is also clearly illustrated by the

ALDs of the total G. turgida population, which were 1.1 and

1.2 cm in the control 1 and control 2 cores, in contrast to 1.5

and 1.9 cm in the net 1 and net 2 cores. The ALDs of the

living populations were very similar to the values calculated

for the total populations, being 1.0 cm for both control cores

and 1.6 cm for the net 1 core and 2.0 cm for the net 2 core.

As all foraminifera were placed at 1 cm depth in sediment

at the start of the experiment, under both control and

experimental conditions, G. turgida populations were ob-

served to move (Fig. 4). Interestingly, downwards migration

was observed in both the net and the control cores.

Fig. 4. The vertical distribution of Globobulimina turgida in sediment. Left-

hand plots show the total numbers found in each sediment depth interval. The

nitrate penetration depth at the start of the experiment is indicated with a

dashed line. Right-hand plots (control 1, 2 and net 1) show the per cent found

alive based on specimens stained with the FDA. The live (%) for net 2 core (grey

filling) represents minimum values and it is based on the number of specimens

containing intracellular nitrate (= live) vs. not containing nitrate (= dead).

FEMS Microbiol Ecol 75 (2011) 273–283c� 2010 Federation of European Microbiological SocietiesPublished by Blackwell Publishing Ltd. All rights reserved

278 K.A. Koho et al.

Although upwards migration was also present in the control

cores, without the restriction of the net, no G. turgida were

recorded in the very superficial sediments of either control

core, where oxygen was available for respiration. Upwards

migration of foraminifera has been reported previously in

several experimental studies investigating the influence of

pore water oxygen content on the vertical distribution of

benthic foraminifera (Moodley et al., 1998; Gross, 2000;

Duijnstee et al., 2003; Geslin et al., 2004); in general,

following a change in sedimentary redox conditions, many

foraminifera migrate towards more oxygenated conditions.

This behaviour has been less clearly defined for deep

infaunal species, such as G. turgida. Nevertheless, field

studies of Fontanier et al (2003), Silva et al. (1996) and

Koho et al. (2008a) have noted that G. turgida is typically

found close to oxygen penetration depth in sediment, within

or just below the denitrification zone. In our experiment, G.

turgida was found within and below the denitrification zone,

but apparently did not choose to migrate to the zone of

oxygen penetration in the shallowest sediment.

The downward migration observed in all cores is intri-

guing. We speculate that foraminifera continue moving in

random directions until they find favourable conditions

(with free pore water NO3� and/or O2). This would explain

the distributions in the control cores where foraminifera

were able to move freely. Thus, some foraminifera were able

to find their way to the nitrate-containing sediment layer,

while others remained deeper than the nitrate penetration

depth throughout the experiment (indicated by the lack of15N in deep-dwelling specimens, Fig. 5). A few of the nitrate-

labelled specimens seemed to have moved downward after

uptake, whereas the majority ‘preferred’ to remain shallower

in the sediment. Interestingly, the highest intracellular

nitrate pools in the net cores were found in the deep-

residing specimens, suggesting that individuals with very

high NO3� pools can residue further from the nitrate source.

Short-term nitrate uptake by G. turgida underoxic and anoxic conditions

Nitrate uptake capacities of single G. turgida specimens

under oxic and anoxic conditions were studied by following

the intracellular nitrate content and the 15NO3� atom content

in a time-series experiment (Fig. 6). The individuals incu-

bated under oxic conditions between 1 h and 4 days showed

(total) intracellular nitrate pools ranging between 1422 and

32 885 pmol, and under anoxic (He-flushed) conditions

between 634 and 21 234 pmol. An intracellular nitrate pool

of zero was measured in eight foraminifera from the anoxic

incubations and nine specimens in the oxic incubations. The

zero values were not considered further as the viability of the

specimens was not checked during sampling. All the intra-

cellular nitrate pool values were within the range of previous

measurements of Globobulimina (Risgaard-Petersen et al.,

Fig. 5. The intracellular nitrate and isotope signature of Globobulimina turgida in control and net cores. The intracellular nitrate pool of zero was only

taken into account if the specimen was identified live using FDA staining. In core net 2, where FDA staining was not done, zero values were taken as an

indication that the specimen was dead at the time of sampling.

FEMS Microbiol Ecol 75 (2011) 273–283 c� 2010 Federation of European Microbiological SocietiesPublished by Blackwell Publishing Ltd. All rights reserved

279Survival mechanisms of foraminifers

2006; Glud et al., 2009; Pina-Ochoa et al., 2010a, b).

Furthermore, no clear trend and no statistical differences

were observed in the (total) intracellular nitrate content

through time either in the oxic or the anoxic incubations

(Kruskal–Wallis test: P = 0.118 and 0.393, respectively).15N-labelled nitrate was measured in the intracellular

nitrate pool of foraminifera incubated under both oxic and

anoxic conditions (Fig. 6). In the anoxic treatment nitrate,

uptake of 720 pmol was already measured after 7 h (in one

specimen). Furthermore, after 23 h of incubation, four out

of five specimens contained labelled nitrate (ranging be-

tween 483 and 938 pmol). At the subsequent sample times

(30 and 50.5 h), G. turgida were completely devoid of

intracellular nitrate; thus, it is possible that G. turgida were

dead or dying in the incubation. At the end of the experi-

ment (after 101 h), however, one specimen was found again

to contain 249 pmol of 15N-labelled nitrate. Under oxic

conditions, uptake was observed to start later than that

under the anoxic conditions. After a 30-h incubation period,

two G. turgida contained labelled nitrate (928 and

2085 pmol) and at each subsequent sample time, one speci-

men was found to contain 15N-labelled nitrate

(�2000 pmol). The somewhat delayed uptake under the

oxic conditions may be due to preferential respiration with

oxygen rather than nitrate. It is unlikely that the delayed

nitrate uptake was triggered by decreasing oxygen concen-

trations in the Exetainers. The oxic respiration rates for

foraminifera are in the order of � 1000 pmol per ind. day�1

(Pina-Ochoa et al., 2010a) and the Exetainers were filled

with oxygen-saturated (� 250 mM) ASW. In addition, the

Exetainers contained no sediment; thus, any bacterial re-

spiration should have been of minor influence.

No systematic trend was seen in the uptake of labelled

nitrate either in oxic or in anoxic incubations. Labelled 15N

was detected in specimens containing both low (from

636 pmol per cell) and high (19 992 pmol per cell) ‘original’

concentrations of nitrate measured as 14NO3�. Further-

more, a considerable number of specimens, across a similar

range of intracellular pool concentrations (ranging

634–32 886 pmol per cell), did not take up any of the label.

Some of the scatter in the uptake data may be due to variable

feeding histories between individual G. turgida specimens

within the experiment. Nitrate collection and cellular sto-

rage are energy intensive; hence, variable levels of respirable

fuel between individuals may dictate the energy available for

this behaviour. This hypothesis is supported by the previous

laboratory experiment of Pina-Ochoa et al. (2010b), in

which a large scatter was measured in the ATP content of

G. turgida, and the observations of Linke (1992), where

ingestion of fresh phytodetritus lead to a 10-fold increase in

the cellular ATP levels.

Nitrate uptake vs. consumption in G. turgida

The results from the oxic vs. the anoxic 15N uptake experi-

ment showed nitrate uptake rates of G. turgida ranging

Fig. 6. Intracellular nitrate and N-isotope signature of Globobulimina turgida in the short-term uptake experiment. As no viability check was carried out

at the sample times, an intracellular nitrate pool of zero was taken as an indication that the specimen was most likely dead.

FEMS Microbiol Ecol 75 (2011) 273–283c� 2010 Federation of European Microbiological SocietiesPublished by Blackwell Publishing Ltd. All rights reserved

280 K.A. Koho et al.

between 457 and 1668 pmol N per ind. day�1 for oxic and

between 59 and 2467 pmol N per ind. day�1 for anoxic

conditions. The maximum uptake rates for both set-ups

were measured at the early stages of the incubations, and the

minimum rates close to the end. The general decrease in the

uptake rates with time may reflect the unnatural and adverse

(no sediment and no food) conditions in the experiment, as

nitrate was still readily available in both incubations at the

end of the experiment (44mM in oxic and 45 mM in anoxic

Exetainer, Fig. 3).

In comparison with the consumption rates of G. turgida

(263–453 pmol N per ind. day�1, Pina-Ochoa et al., 2010b),

the maximum potential rates of nitrate accumulation were

around five times higher than the consumption rates under

both oxic and anoxic conditions. Furthermore, labelled

nitrate accumulation of 928–2199 pmol for oxic conditions

and 249–938 pmol for anoxic conditions would be enough

to feed denitrification between 2–8.3 and 0.5–2 days,

respectively. Similar results were observed for the foramini-

fer Nonionella cf. stella from the oxygen minimum zone off

the Chilean coast, where, compared with the denitrification

rates, the potential rates of nitrate accumulation by N. cf.

stella (864 pmol N per ind. day�1) were an order of magni-

tude higher, and a steady state of nitrate storage of about

230 pmol N per individual fed denitrification for 2.5 days

(Høgslund et al., 2008). Therefore, it appears that foramini-

fera can accumulate nitrate at ease, even if they may need to

denitrify simultaneously.

Although the maximum uptake rates were higher in the

anoxic than in oxic incubations, the total nitrate accumula-

tion was greater under the oxic conditions. The higher total

capacity of G. turgida to collect nitrate under oxic condi-

tions may be related to the auxiliary nature of denitrification

metabolism in foraminifera. As oxygen respiration rates are

generally between three and 13 times higher than denitrifi-

cation rates (Pina-Ochoa et al., 2010a), and have a signifi-

cantly higher energy yield (Strohm et al., 2007),

foraminifera can concentrate more nitrate with less effort.

Furthermore, under oxygen-free conditions, G. turgida may

not able to accumulate as much nitrate because they are

simultaneously respiring through denitrification.

Conclusions

� Results from our long-term experiment demonstrate that

G. turgida living well below the nitrate penetration depth

were not capable of stretching their pseudopodial net-

work into the above diffusive boundary layer to collect

nitrate. Thus, to restore their intracellular nitrate storage,

they must physically migrate upward through the sedi-

ment strata, which is either within or very close (� few

millimetres) to the zone of free pore water nitrate. The

importance of the pseudopodia in foraminiferal nitrate

respiration (i.e. through the presence of mitochondria

over the pseudopodial network and its periphery) needs

to be further clarified.

� The vertical distribution of G. turgida in the control and

experimental cores showed that foraminifera migrated

upward or downward in the sediment while searching for

its ‘favoured’ geochemical environment. Living G. turgida

present in control cores seemed to gather in the top 1 cm

of the sediment (within and below the denitrification

zone); however, despite the facultative anaerobic nature of

foraminifera, no G. turgida were found in the uppermost,

well-oxygenated sediment. Foraminifera in the net cores

were found deeper and the deep-living individuals ap-

peared to have the highest nitrate storage supplies.

� Nitrate uptake capacities of G. turgida under anoxic

conditions were faster than under oxic conditions, prob-

ably related to the auxiliary nature of denitrification

metabolism. Furthermore, labelled nitrate was measured

in specimens containing both high and low intracellular

nitrate levels, thus suggesting that nitrate uptake was not

triggered by a threshold intracellular nitrate concentra-

tion. The maximum potential nitrate uptake rates were

five times higher than the consumption rates under both

conditions, being enough to feed denitrification between

0.5–2 and 2–8 days under oxic and anoxic conditions,

respectively.

Acknowledgements

We are grateful to N.P. Revsbech and P. Sørensen for

providing microsensors. The crew of R/V Oscar von Sydow

at the Kristineberg Marine Field Station (Sweden) are also

thanked. Tom Jilbert is thanked for proof reading the

manuscript. This research was financially supported by the

European Union Marie Curie Fellowship (FP7-IEF-220894),

Netherlands Organisation for Scientific Research (NWO),

Darwin Center for Biogeosciences (grant no. 1092), the

Danish National Research Foundation and the German

Max Plank Society. Two anonymous reviewers are acknowl-

edged, whose comments improved the previous version of

the manuscript considerably.

Authors’contribution

K.A.K. and E.P.-O. contributed equally to this paper.

References

Bernhard JM (1992) Benthic foraminiferal distribution and

biomass related to pore-water oxygen-content: central

California continental slope and rise. Deep-Sea Res I 39:

585–605.

FEMS Microbiol Ecol 75 (2011) 273–283 c� 2010 Federation of European Microbiological SocietiesPublished by Blackwell Publishing Ltd. All rights reserved

281Survival mechanisms of foraminifers

Bernhard JM & Reimers CE (1991) Benthic foraminiferal

population fluctuations related to anoxia: Santa Barbara Basin.

Biochem 15: 127–149.

Bernhard JM & Bowser SS (1992) Bacterial biofilms as a trophic

resource for certain benthic foraminifera. Mar Ecol Prog Ser 83:

263–272.

Bernhard JM & Sen Gupta BK (1999) Foraminifera of oxygen

depleted environments. Modern Foraminifera (Gupta S, ed),

pp. 201–216. Kluwer Academic Publishers, Dordrecht.

Bernhard JM, Newkirk SG & Bowser SS (1995) Towards a non-

terminal viability assay for foraminiferan protists. J Eukaryot

Microbiol 42: 357–367.

Bowser SS & Travis JL (2002) Reticulopodi: structure and

behavioral basis for the suprageneric placement of

granuloreticulosan protests. J Foram Res 32: 44–447.

Braman RS & Hendrix SA (1989) Nanogram nitrite and nitrate

determination in environmental and biological-Materials by

vanadium(III) reduction with chemi-luminescence detection.

Anal Chem 61: 2715–2718.

Cedhagen T & Frimanson H (2002) Temperature dependence of

pseudopodial organelle transport in seven species of

foraminifera and its functional consequences. J Foram Res 32:

434–439.

Corliss BH (1985) Microhabitats of benthic foraminifera within

deep-sea sediments. Nature 314: 435–438.

Corliss BH (1991) Morphology and microhabitat preferences of

benthic foraminifera from the northwest Atlantic Ocean. Mar

Micropaleontol 17: 195–236.

Duijnstee I, Ernst S & Van der Zwaan B (2003) Effect of anoxia on

the vertical migration of benthic foraminifera. Mar Ecol Progr

Ser 246: 85–94.

Ernst SR, Bours R, Duijnstee IAP & van der Zwaan GJ (2005)

Experiment effects of an organic matter pulse and oxygen

depletion on a benthic foraminiferal shelf community. J Foram

Res 35: 177–197.

Fontanier C, Jorissen FJ, Licari L, Alexandre A, Anschutz P &

Carbonel P (2002) Live benthic foraminiferal faunas from the

Bay of Biscay: faunal density, composition, and microhabitats.

Deep-Sea Res I 49: 751–785.

Fontanier C, Jorissen FJ, Chailloua G, David C, Anschutz P &

Lafon V (2003) Seasonal and interannual variability of benthic

foraminiferal faunas at 550 m depth in the Bay of Biscay. Deep-

Sea Res Part I 50: 457–494.

Geslin E, Heinz P, Jorissen F & Hemleben C (2004) Migratory

responses of deep-sea benthic foraminifera to variable oxygen

conditions: laboratory investigations. Mar Micropaleontol 53:

227–243.

Glud RN, Thamdrup B, Stahl H, Wenzhoefer F, Glud A, Nomaki

H, Oguri K, Revsbech NP & Kitazato H (2009) Nitrogen

cycling in a deep ocean margin sediment (Sagami Bay, Japan).

Limnol Oceanogr 54: 723–734.

Goldstein ST (1999) Foraminifera: a biological overview. Modern

Foraminifera (Sen Gupta BK, ed), pp. 7–55. Kluwer,

Dordrecht.

Gooday AJ, Bernhard JM, Levin LA & Suhr SB (2000)

Foraminifera in the Arabian Sea oxygen minimum zone and

other oxygen-deficient settings: taxonomic composition,

diversity, and relation to metazoan faunas. Deep-Sea Res Part II

47: 25–54.

Gross O (2000) Influence of temperature, oxygen and food

availability on the migrational activity of Bathyal benthic

foraminifera: evidence by microcosm experiments. Hydrobiol

426: 123–137.

Høgslund S, Revsbech NP, Cedhagen T, Nielsen LP & Gallardo VA

(2008) Denitrification, nitrate turnover, and aerobic

respiration by benthic foraminiferans in the oxygen minimum

zone off Chile. J Exp Mar Biol Ecol 359: 85–91.

Jorissen FJ, Buzas MA, Culver SJ & Kuehl SA (1994) Vertical

distribution of living benthic foraminifera in submarine

canyons off New Jersey. J Foram Res 24: 28–36.

Jorissen FJ, De Stigter HC & Widmark JGV (1995) A conceptual

model explaining benthic foraminiferal microhabitats. Mar

Micropaleontol 22: 3–15.

Kjær T, Hauer Larsen L & Revsbech NP (1999) Sensitivity control

of ion-selective biosensors by electrophoretically mediated

analyte transport. Analy Chim Acta 391: 57–63.

Koho KA, Garcıa R, de Stigter HC, Epping E, Koning E,

Kouwenhoven TJ & van der Zwaan GJ (2008a) Sedimentary

labile organic carbon and pore water redox control on species

distribution of benthic foraminifera: a case study from Lisbon-

Setubal Canyon (southern Portugal). Progr Oceanogr 79:

55–82.

Koho KA, Langezaal AM, van Lith YA, Duijnstee IAP & van der

Zwaan GJ (2008b) The influence of a simulated diatom bloom

on deep-sea benthic foraminifera and the activity of bacteria: a

mesocosm study. Deep-Sea Res I 55: 696–719.

Larsen LH, Kjær T & Revsbech NP (1997) A microscale NO3�

biosensor for environmental applications. Anal Chem 69:

3527–3531.

Linke P (1992) Metabolic adaptations of deep-sea benthic

foraminifera to seasonally varying food input. Mar Ecol Prog

Ser 81: 51–63.

Lipps JH (1983) Biotic interactions in benthic foraminifera. Biotic

Interactions in Recent and Fossil Benthic Communities (Tevesz

MJS & McCall PL, eds), pp. 331–376. Plenum Press, New York.

Moodley L, Schaub BEM, Van der Zwaan GJ & Herman PMJ

(1998) Tolerance of benthic foraminifera (Protista: Sarcodina)

to hydrogen sulphide. Mar Ecol Progr Ser 169: 77–86.

Pina-Ochoa E, Høgslund S, Geslin E, Cedhagen T, Revsbech NP,

Nielsen LP, Schweizer M, Jorissen F, Rysgaard S & Risgaard-

Petersen N (2010a) Widespread occurrence of nitrate storage

and denitrification among Foraminifera and Gromiida. P Natl

Assoc S 107: 1148–1153.

Pina-Ochoa E, Koho KA, Geslin E & Risgaard-Petersen N (2010b)

Survival and life strategy of foraminifer, Globobulimina

turgida, through nitrate storage and denitrification. Mar Ecol

Prog Ser 417: 39–49.

Revsbech NP (1989) An oxygen microelectrode with a guard

cathode. Limnol Oceanogr 34: 474–478.

FEMS Microbiol Ecol 75 (2011) 273–283c� 2010 Federation of European Microbiological SocietiesPublished by Blackwell Publishing Ltd. All rights reserved

282 K.A. Koho et al.

Risgaard-Petersen N & Rysgaard S (1995) Nitrate reduction in

sediments and waterlogged soils measured by 15N techniques.

Methods in Applied Soil Microbiology and Biochemistry (Alef K

& Nannipieri P, eds), pp. 287–296. Academic Press Inc.,

London.

Risgaard-Petersen N, Rysgaard S & Revsbech NP (1993) A

sensitive assay for determination of N-14/N-15 isotope

distribution in No3. J Microbiol Meth 17: 155–164.

Risgaard-Petersen N, Langezaal AM, Ingvardsen S et al. (2006)

Evidence for complete denitrification in a benthic foraminifer.

Nature 443: 93–96.

Silva KA, Corliss BC, Rathburn AE & Thunnell RC (1996)

Seasonality of living benthic foraminifera from the San Pedro

Basin, California Borderland. J Foram Res 26: 71–93.

Strohm TJ, Griffin B, Zumft WG & Schink B (2007) Growth

yields in bacterial denitrification and nitrate ammonification.

Appl Environ Microbiol 73: 1420–1424.

Travis JL & Bowser SS (1991) The motility of foraminifera.

Biology of Foraminifera (Lee JJ & Anderson GC, eds), pp.

91–155. Academic Press, London.

Travis JL, Welnhofer EA & Orokos DD (2002) Automonous

reorganization of foraminiferan reticulopodia. J Foram Res 32:

425–433.

FEMS Microbiol Ecol 75 (2011) 273–283 c� 2010 Federation of European Microbiological SocietiesPublished by Blackwell Publishing Ltd. All rights reserved

283Survival mechanisms of foraminifers