Blackwell Publishing LtdOxford, UKEMIEnvironmental Microbiology 1462-2912© 2006 The Authors; Journal compilation © 2006 Society for Applied Microbiology and Blackwell Publishing Ltd? 20068610741084Original Article
Algae-bacteria interactionsH.-P. Grossart, G. Czub and M. Simon
Environmental Microbiology (2006) 8(6), 1074–1084
doi:10.1111/j.1462-2920.2006.00999.x
Algae–bacteria interactions and their effects on
aggregation and organic matter flux in the sea
Hans-Peter Grossart,*† Gertje Czub‡ and
Meinhard Simon
Institute for Chemistry and Biology of the Marine
Environment, University of Oldenburg, PO Box 2503,
D-26111 Oldenburg, Germany.
Summary
Aggregation of algae, mainly of diatoms, is an important process in marine pelagic systems, often terminating phytoplankton blooms and leading to the
sinking of particulate organic matter in the form of
marine snow. This process has been studied extensively, but the specific role of heterotrophic bacteria
has largely been neglected, mainly because field
studies and most experimental work were performed
under non-axenic conditions. We tested the hypothesis that algae–bacteria interactions are instrumental
in aggregate dynamics and organic matter flux. A
series of aggregation experiments has been carried
out in rolling tanks with two marine diatoms typical
of temperate regions (Skeletonema costatum and
Thalassiosira rotula) in an axenic treatment and one
inoculated with marine bacteria. Exponentially growing S. costatum and T. rotula exhibited distinctly different aggregation behavior. This was reflected by
their strikingly different release of dissolved organic
matter (DOM), transparent exopolymer particles (TEP)
and protein-containing particles (CSP), as well as
their bacterial biodegradability and recalcitrance.
Cells of S. costatum aggregated only little and their
bacterial colonization remained low. Dissolved organic matter, TEP and CSP released by this alga were
largely consumed by free-living bacteria. In contrast,
T. rotula aggregated rapidly and DOM, TEP and
CSP released resisted bacterial consumption. Experiments conducted with T. rotula cultures in the
stationary growth phase, however, showed rapid bacterial colonization and decomposition of algal cells.
Received 20 December, 2005; accepted 20 December, 2005. *For
correspondence. E-mail [email protected]; Tel. (+49) 33082
699 91; Fax (+49) 33082 699 17. Present address: †Institute of Freshwater Ecology and Inland Fisheries, Department of Limnology of
Stratified Lakes, Alte Fischerhuette 2, D-16775 Neuglobsow, Germany; ‡Department of Applied Environmental Science, Stockholm
University, SE-10691 Stockholm, Sweden.
Our study highlights the importance of heterotrophic
bacteria to control the development and aggregation
of phytoplankton in marine systems.
Introduction
There is ample evidence that phytoplankton blooms, in
particular composed of diatoms, form aggregates (marine
snow) which are one of the major components of sinking
particulate organic carbon (POC) (Simon et al., 2002;
Thornton, 2002). Transparent exopolymer particles (TEP)
have been identified as an important agent for aggregation (Passow, 2002a). Various studies have shown that
TEP is produced by planktonic algae, but also by bacteria
and from dissolved precursor material (Zhou et al., 1998;
Passow, 2002b; Engel et al., 2004). The role of another
class of microparticles containing protein [Coomassie brilliant blue stainable particles (CSP); Long and Azam,
1996] in aggregation processes needs to be clarified. The
specific role of heterotrophic bacteria in aggregation processes, i.e. the formation of TEP, other polymeric sticky
material, and CSP, and the effect of bacteria on the adhesive properties and coagulation of this material is still
unclear. This is because the great majority of studies have
been carried out with non-axenic planktonic algae, thus
masking the specific bacterial impact. Four scenarios are
conceivable: (i) the organic material may be recalcitrant
to bacterial decomposition; (ii) bacteria may consume it to
varying extents, thus reducing the aggregation potential;
(iii) bacteria may modify this material and change its adhesive properties; and (iv) the colonization of the algal cell
surface by specific bacterial communities may directly
affect the alga’s adhesive properties. These scenarios
may be characteristic of an algal species, but may also
vary with its physiological state, i.e. depend on the quality
of the secreted material and the associated bacteria.
The role of heterotrophic bacteria in decomposing
phytoplankton-derived POC and phytodetrital aggregates
is much better understood. It has been shown that bacteria colonize planktonic algae, mainly when they become
more senescent and aggregate (e.g. Smith et al., 1995),
and that they solubilize phytodetrital aggregates (Smith
et al., 1992; Grossart and Ploug, 2001), i.e. transforming
particulate (POC) to dissolved organic carbon (DOC). In
addition to POC solubilization, the high proteolytic activity
of bacteria associated with diatoms greatly enhances sil-
© 2006 The Authors
Journal compilation © 2006 Society for Applied Microbiology and Blackwell Publishing Ltd
Algae-bacteria interactions 1075
ica dissolution and thus reduces the POC sinking flux
(Bidle and Azam, 1999).
Hence, our present understanding of the role of heterotrophic bacteria in algae–bacteria interactions appears
to be ambivalent. Whereas their role in phytodetrital POC
decomposition appears to be rather well understood, we
still lack a detailed understanding of the bacterial decomposition of specific phytoplankton-derived DOC compounds and how this affects aggregation processes.
Further, we still know little about whether specific algal
species attract distinct bacterial communities or populations, which may also have implications for phytoplankton–bacteria interactions (Schäfer et al., 2002; Bates
et al., 2004; Rooney-Varga et al., 2005). The latter question has been addressed in an accompanying study using
the same experimental set-up (Grossart et al., 2005).
The aim of the present study was to examine specific
interactions between heterotrophic bacteria and two
marine diatoms, Skeletonema costatum and Thalassiosira
rotula, and their implications for phytoplankton aggregation and sinking flux. We chose these chain-forming
species as they are common in the phytoplankton of temperate coastal marine regions. To test whether bacterial
consumption and secretion of specific DOC compounds,
such as dissolved amino acids and dissolved carbohydrates, affect the formation of TEP, CSP and aggregates,
we conducted experiments with axenic and non-axenic
cultures of these two diatoms.
Results
Aggregation patterns
The results exhibited distinct differences between the two
diatoms with respect to aggregation processes. Whereas
exponentially growing (exp) S. costatum aggregated only
weakly, exponentially growing T. rotula readily formed
aggregates (Table 1). The first aggregates > 2 mm in size
in cultures of S. costatum did not occur until 40 h of incubation, and even an increase of the initial cell concentration by one order of magnitude to 104 cells ml−1 only
slightly reduced the time until aggregates formed (35 h).
Image analysis revealed that the area of S. costatum
aggregates remained small compared with T. rotula,
although both types of aggregates were of similar length.
Aggregation of T. rotula was faster in low light conditions
than in high light conditions, yielding aggregates of
> 2 mm already within 4 h. However, in the former case
aggregates remained smaller (Table 1). Under high light
conditions, aggregates increased in size for about 60% of
the incubation time and the algae and bacteria kept growing, except in the stationary-phase experiment. Aggrega-
Table 1. Aggregation dynamics of Skeletonema costatum and Thalassiosira rotula cultures growing exponentially and in the stationary phase.
Thalassiosira rotula
Skeletonema costatum
Aggregation experiment
Algal growth stage
High light conditions
Time to form aggregates > 2 mm Ø
Aggregate area
(mm2 l−1)
Maximum aggregate length (mm)
Diatom growth
Bacterial growth
CSP
(103 µm2 ml−1)
TEP
(103 µm2 ml−1)
Low light conditions
Time to form aggregates > 2 mm Ø
Aggregate area
(mm2 l−1)
Maximum aggregate length (mm)
Diatom growth
Bacterial growth
CSP
(103 µm2 ml−1)
TEP
(103 µm2 ml−1)
1
Exponential
2
Exponential
3
Exponential
4
Exponential
5
Stationary
6
Exponential
(axenic)
7
Exponential
nd
nd
nd
nd
nd
nd
nd
nd
nd
nd
nd
nd
nd
nd
10.5
++/o
40–280
> 20
+
++
–
45–195
++
50–320
14
++/o
20–650
18
+++
+++
+
8–18
+/–
6–16
15
+
35–320
14
o/–
+/–
++/–
150–1200
o
20–25
9
+/o
35–180
8
o/–
nd
+/–
25–95
o
10–20
12
++/o
210–380
12
o/+
++
+/–
35–205
(+)
10–25
40
(+)
4–20
10
nd
nd
nd
35
(+)/+
10–70
4
nd
nd
nd
nd
nd
0.5
o/–
80–270
4
–/o
++
–
10–300
–
10–40
4
(–)
10–32
2
+/–
++
++
6–110
+/–
5–17
3
o/+
22–170
6
o/–
o
+/–
20–175
+/–
7–42
19
(+)/o
20–170
6
o/–
nd
–/+
20–75
+/–
2–20
5
++/o
70–440
12
o/+
++
+/–
30–140
–
2–20
nd, not determined; +, increase; o, no changes; –, decrease; ±, first increase then decrease; ( ), only slight changes.
Skeletonema costatum and Thalassiosira rotula cultures were inoculated with natural seawater bacteria and incubated under either high or low
light conditions. The initial concentration of algal cells was approximately 1000 cells ml−1, except in experiment 2, which started with approximately
10 000 cells ml−1. Experiment 7 is a continuation of experiment 6 which was inoculated with seawater bacteria after 68 h.
© 2006 The Authors
Journal compilation © 2006 Society for Applied Microbiology and Blackwell Publishing Ltd, Environmental Microbiology, 8, 1074–1084
1076 H.-P. Grossart, G. Czub and M. Simon
Without bacteria
With bacteria
+ bacteria
2 -1
Area (mm l )
450
300
150
137:00
132:00
126:10
120:00
114:20
105:00
92:00
81:00
75:10
69:00
68:00
60:00
54:00
48:20
42:00
33:20
24:10
17:50
6:00
12:00
0:00
0
Fig. 1. Video images of aggregates (top) in both axenic and non-axenic approaches as well as average aggregate area (bottom) throughout the
incubation of exponentially growing Thalassiosira rotula. Scale bar equals 10 mm. At 68 h natural seawater bacteria were added to the axenic
diatom culture (arrow). Error bars are given for parallel experiments.
tion of axenic T. rotula at high light was relatively fast
compared with the non-axenic experiments but aggregate
size remained smaller. In contrast, at low light conditions
axenic T. rotula aggregated substantially slower than nonaxenic ones. When bacteria were added to the axenic
culture after 68 h aggregation was immediately enhanced,
leading to strikingly larger aggregates within 10 h (Table 1,
Fig. 1). Numbers and area of TEP and CSP did not exhibit
clear-cut results with respect to the treatments and growth
stages of T. rotula (Table 1).
Therefore, we re-examined the formation of TEP and
CSP in relation to algal and bacterial abundance as well
as concentration and quality of dissolved organic matter
and bacterial activities in axenic versus non-axenic cultures of T. rotula and S. costatum.
Phytoplankton growth dynamics
Algal numbers of axenic and non-axenic cultures
exhibited pronounced differences between algal species
and growth stage (Fig. 2A and B). Numbers of axenic
S. costatum were higher than of non-axenic S. costatum
until the late exponential phase. The non-axenic treatment, however, reached higher algal cell numbers at the
end of the exponential growth phase and enhanced num-
bers persisted throughout the stationary and declining
phase until the end of the experiment. In contrast, numbers of axenic (exp) T. rotula were lower than those in the
non-axenic treatment during the active growth phase
from day 4 to 7 but remained higher thereafter. Numbers
of axenic stationary (stat) T. rotula remained almost constant throughout the whole incubation period whereas
those in the non-axenic treatment dramatically decreased
already after 1 day, indicating a rapid bacterial degradation of the alga. In none of the axenic cultures were
bacteria detectable by epifluorescence microscopy,
demonstrating sterile conditions throughout the whole
incubation.
Dynamics of microparticles
In both diatom cultures, concentrations of TEP continuously increased during exponential growth (Fig. 3A). The
non-axenic (exp) T. rotula culture had higher numbers of
TEP on days 7 and 9 than the axenic T. rotula. In contrast,
the axenic (exp) S. costatum culture had substantially
higher concentrations of TEP than the non-axenic culture
after 2 days. In the (stat) T. rotula culture, TEP abundance
increased over time in the axenic treatment but decreased
in the non-axenic one (Fig. 3A). In axenic f/2 medium TEP
© 2006 The Authors
Journal compilation © 2006 Society for Applied Microbiology and Blackwell Publishing Ltd, Environmental Microbiology, 8, 1074–1084
Algae-bacteria interactions 1077
A
Skeletonema costatum
B
Thalassiosira rotula
Algae (103 mL-1)
150
1500
Free Bacteria
(106 mL-1)
60
30
0
0
40
40
D
Skc exp+
f/2
30
Thr stat+
Thr exp+
30
20
20
10
10
0
0
F
6
Attached Bacteria
(106 mL-1)
BP (µg C L-1 h-1)
Skc expSkc exp+
500
E
G
90
1000
C
Thr expThr exp+
Thr statThr stat+
120
Skc exp+
f/2
Thr stat+
Thr exp+
30
4
20
2
10
0
0
100
100
H
Skc exp+
f/2
75
Thr stat+
Time(d) vs AttThr stat+
Time(d) vs AttThr exp+Thr exp+
75
50
50
25
25
0
0
0
2
4
6
8
10
12
0
2
4
6
8
10
12
Time (d)
Fig. 2. Abundances of Skeletonema costatum (Skc, A), Thalassiosira rotula (Thr, B), free-living (C and D) and attached bacteria (E and F) and
bacterial biomass production (G and H) in the course of experiments with the diatoms in exponential (exp) and stationary (stat) growth and in
axenic (–) and non-axenic (+) cultures and in f/2 medium. Note the different scales of some corresponding y-axes. Error bars are given for
replicates of the same experiment.
© 2006 The Authors
Journal compilation © 2006 Society for Applied Microbiology and Blackwell Publishing Ltd, Environmental Microbiology, 8, 1074–1084
1078 H.-P. Grossart, G. Czub and M. Simon
A
3
B
Skc exp
2
Skc exp
2
1
1
0
3
0
3
Thr exp
3
Thr stat
2
-1
5
0
1
5
-1
1
Thr exp
2
CSP (10 ml )
2
TEP (10 ml )
3
0
3
Thr stat
2
1
1
0
0
3
3
f/2
f/2
axenic
non-axenic
2
2
1
1
0
0
0
2
4
6
8
10
12
Time (d)
0
2
4
6
8
10
12
Time (d)
Fig. 3. Abundance of (A) transparent exopolysaccharide (TEP) and (B) proteinaceous particles (CSP) in axenic and non-axenic cultures of
Skeletonema costatum and Thalassiosira rotula and in f/2 medium. Error bars are given for replicates of the same experiment. For legends see
Fig. 2.
abundances were low and at times slightly higher in the
non-axenic treatment.
Concentrations of CSP in both (exp and stat) T. rotula
cultures showed similar patterns as TEP (Fig. 3B). However, CSP abundance in the axenic (stat) T. rotula culture
remained lower than TEP, whereas numbers of CSP
decreased more strongly in the non-axenic culture than
TEP. In contrast to TEP, numbers of CSP in both axenic
and non-axenic (exp) S. costatum cultures increased until
day 9. Thereafter, numbers further increased in the axenic
treatment but decreased in the non-axenic one. In the f/2
medium, CSP abundance remained very low in both treatments except for days 4–7.
Dynamics of bacterial numbers and activities
Numbers of free-living and attached bacteria showed pronounced differences with respect to algal species and
growth stage (Fig. 2C–F). Numbers of free-living bacteria
in the (exp) T. rotula culture were similar to those in f/2
medium until day 7 with a peak on day 3. This suggests
that growth was not due to algal-born substrates but due
to DOC in the medium. When subtracting bacterial numbers in the f/2 medium (control) from those of the (exp)
T. rotula culture, both numbers of free-living and attached
bacteria remained low and only slightly increased towards
the end. In contrast, free-living as well as attached bacteria in the (stat) T. rotula culture reached a pronounced
maximum on day 3 and rapidly decreased thereafter.
Numbers of free-living bacteria in the (exp) S. costatum
culture strongly increased until day 7 and decreased
thereafter, showing a similar but time-extended pattern
as the (stat) T. rotula culture. In contrast, numbers of
attached bacteria remained low and only slightly
increased on days 9 and 11 when algal numbers declined,
indicating enhanced bacterial colonization of senescent
algae.
Bacterial production (BP) co-varied with bacterial numbers both of (exp) S. costatum and of (stat) T. rotula with
a pronounced peak on days 7 and 3 respectively
(Fig. 2G–H). The initial increase of BP in (exp) T. rotula
until day 3 co-varied with free-living bacterial numbers in
© 2006 The Authors
Journal compilation © 2006 Society for Applied Microbiology and Blackwell Publishing Ltd, Environmental Microbiology, 8, 1074–1084
Algae-bacteria interactions 1079
Dynamics of dissolved organic matter
Concentrations of DOC were almost similar in the axenic
and the non-axenic cultures of (exp) T. rotula (Fig. 4).
They increased towards the end of the experiment, covarying only with enhanced AMP activities. In contrast,
DOC concentrations in non-axenic cultures of (stat)
T. rotula and (exp) S. costatum were much lower compared with the respective axenic treatments, indicating a
rapid bacterial solubilization and decomposition of the
algal-born organic matter. Dissolved organic carbon concentration in the axenic (exp) S. costatum culture constantly increased until the end of the experiment. In the f/
2 medium, DOC concentrations remained nearly constant
in the axenic treatment but declined in the non-axenic
treatment after day 3.
Concentrations of dissolved free (DFAA) and dissolved
combined amino acids (DCAA) were generally lower in all
non-axenic treatments (Fig. 5A and B). In contrast to
DOC, amino acid concentrations remained low at the end
of the experiment in the non-axenic treatment of the (exp)
T. rotula culture. Organic carbon bound in dissolved amino
acids constituted between 1% and 10% of DOC with highest proportions in the axenic (stat) T. rotula culture. Dissolved free amino acids released by both axenic algae
were dominated by aspartic and glutamic acid, constituting 20–29 and 24–39 mol% respectively. In the (exp)
T. rotula culture, bacteria preferentially consumed the
acidic DFAA, as indicated by greatly reduced mol% of
these amino acids to <13 mol% in the non-axenic treatments towards the end of the incubations. Mol% composition of DCAA was similar for both axenic and non-axenic
algal cultures.
Dynamics of dissolved free neutral (DFCHO) and dis-
1500
1000
500
Skc exp
0
1500
axenic
non-axenic
1000
500
DOC (µM)
this culture and with those in the f/2 medium. Whereas BP
remained at a substantially higher level in the (exp)
T. rotula culture, it decreased in the f/2 medium.
Aminopeptidase (AMP) activity generally co-varied with
bacterial numbers and BP rates in the (stat) T. rotula culture, but maximum rates of AMP (9.2 µmol l−1 h−1) followed
the other peaks by 1 day. In both (exp) diatom cultures,
AMP activity increased to 6.5 µmol l−1 h−1 until day 9 when
numbers of algae and bacteria had already declined.
However, from days 2 to 7 AMP activity in the S. costatum
culture was higher than in (exp) T. rotula.
β-Glucosidase activity in the (stat) T. rotula culture
co-varied with that of AMP. In the (exp) S. costatum
culture, β-glucosidase activity strongly increased up to
0.87 µmol l−1 h−1 between days 3 and 7 and remained high
thereafter. The increase, however, was delayed relative to
that of AMP. In the (exp) T. rotula culture β-glucosidase
activity remained low (< 0.3 µmol l−1 h−1), except towards
the end of the incubation period.
Thr exp
0
1500
1000
500
Thr stat
0
1500
1000
500
f/2
0
0
2
4
6
8
10
12
Time (d)
Fig. 4. Concentrations of dissolved organic carbon (DOC) in both
axenic and non-axenic cultures of Skeletonema costatum and Thalassiosira rotula, and in f/2 medium. Error bars are given for replicates
of the same experiment. For legends see Fig. 2.
solved combined neutral monosaccharides (DCCHO)
were rather similar to those of DFAA and DCAA with
generally lower concentrations in the non-axenic algal cultures ( Fig. 6A and B). Organic carbon bound in dissolved
carbohydrates constituted between 1% and 13% of DOC
with highest values in both axenic (stat) T. rotula and
axenic (exp) S. costatum cultures. Glucose dominated the
DFCHO and DCCHO released by both (exp) axenic algae
(50–78 and 48–54 mol%) but constituted only 20 mol% in
the (stat) axenic T. rotula culture. Other monosaccharides
occurring in both pools included fucose, ribose, rhamnose, arabinose, galactose and mannose. Glucose
was always consumed preferentially, as indicated by the
strongly reduced mol% of glucose in the DFCHO pool of
the non-axenic relative to the axenic treatments towards
© 2006 The Authors
Journal compilation © 2006 Society for Applied Microbiology and Blackwell Publishing Ltd, Environmental Microbiology, 8, 1074–1084
1080 H.-P. Grossart, G. Czub and M. Simon
A
6
B
Axenic
Non-axenic
Thr exp
4
30
Thr exp
20
2
10
0
6
0
30
Thr stat
Thr stat
20
DCAA (µM)
DFAA (µM)
4
2
0
6
Skc exp
4
2
10
0
30
Skc exp
20
10
0
0
6
30
f/2
f/2
4
20
2
10
0
0
0
2
4
6
8
10
12
Time (d)
0
2
4
6
8
10
12
Time (d)
Fig. 5. Concentrations of (A) dissolved free (DFAA) and (B) combined amino acids (DCAA) in both axenic and non-axenic incubations. For
legends see Fig. 2.
the end of the incubation. In the DCCHO of the (exp)
S. costatum culture, glucose and galactose were preferentially consumed as indicated by their reduced mol% in
the non-axenic relative to the axenic treatments towards
the incubation end. In the (stat) T. rotula culture, however,
all neutral DCCHO were consumed in rather equal
proportions.
Discussion
The two investigated diatoms represent two completely
different lifestyles and types of interactions with bacteria.
Whereas in S. costatum bacterial degradation of exudates
prevents aggregation until a late growth stage, cells of
T. rotula aggregate much more rapidly and irrespective of
their bacterial colonization, but the presence of bacteria
does enhance aggregation. Hence, bacteria prevent
aggregation and subsequent sedimentation in the first
case but favour these processes in the second one. These
experiments with axenic and non-axenic diatom cultures
clearly demonstrate that specific interactions of bacteria
with the algae are instrumental for growth dynamics,
aggregation, and sinking of these or other diatoms with
similar properties during the course of a bloom. Specific
algae–bacteria interactions have been largely neglected
so far, but are presumably as important as inorganic nutrient supply and grazing in controlling the development and
fate of diatom blooms (Kiørboe et al., 1994; Passow and
Alldredge, 1995; Smith et al., 1995; Tiselius and Kuylenstierna, 1996). These interactions may vary to a certain
extent, depending on the light conditions, the changing
composition of the bacterial community during the various
growth stages (Grossart et al., 2005) and varying temperatures. They may explain the variability of aggregation
properties of diatoms observed in several studies (e.g.
Kiørboe and Hansen, 1993; Passow and Alldredge, 1995).
So far, growth and aggregation experiments have been
carried out with non-axenic cultures (Alldredge et al.,
1993; Passow et al., 1994). Hence, it has been impossible
to elucidate the role of heterotrophic bacteria in the aggregation process. Kiørboe and colleagues (1994), for example, found that a diatom bloom in a Danish fjord was
© 2006 The Authors
Journal compilation © 2006 Society for Applied Microbiology and Blackwell Publishing Ltd, Environmental Microbiology, 8, 1074–1084
Algae-bacteria interactions 1081
A
0.9
Thr exp
B
Axenic
Non-axenic
0.6
20
Thr exp
15
10
0.3
5
0.0
0
0.9
20
Thr stat
DCCHO (µM)
DFCHO (µM)
0.6
0.3
0.0
0.9
Thr stat
15
Skc
Skc exp
exp
0.6
0.3
10
5
0
20
Skc exp
15
10
5
0.0
0
0.9
20
f/2
f/2
f/2
15
0.6
10
0.3
5
0.0
0
0
2
4
6
8
10
12
0
2
4
6
8
10
12
Time (d)
Time (d)
Fig. 6. Concentrations of (A) dissolved free (DFCHO) and (B) combined neutral carbohydrates (DCCHO) in both axenic and non-axenic
incubations. For legends see Fig. 2.
dominated by S. costatum, and five other diatoms, including T. rotula, constituted minor proportions. The adhesive
properties of the former diatom were much lower than
those of most of the other species, but interactions with
other diatoms enhanced the adhesive properties of
S. costatum. The critical concentration of S. costatum for
aggregation, however, was much higher than that of the
other algae, explaining its numerical dominance until the
termination of this bloom by aggregation and sedimentation. According to our experiments we assume that
free-living bacteria consumed the DOC released by
S. costatum, whereas the other diatoms were colonized
faster by bacteria, thus enhancing aggregation and presumably their subsequent sedimentation. Because in
early spring the water temperature in the fjord was much
lower than in our experiments (2–4°C versus 15°C) and
bacterial growth and physiological activity are low at this
temperature, the situation in this field study resembled the
situation in our axenic cultures.
Our observations complement previous studies which
address the specific role of bacteria in the decomposition
of diatoms and their silicate frustules as well as in the
formation of diatom aggregates (Smith et al., 1992; 1995;
Bidle and Azam, 1999). Our results, however, show that
bacteria are also important in the initial phase of diatom
blooms and can contribute to controlling the development
of these blooms by more subtle and specific interactions
with the diatoms present. Further, bacteria may also
determine whether diatoms are subject to zooplankton
grazing as single cells or colonies, or whether they aggregate and sink and thus escape zooplankton grazing
(Jackson, 2001).
Experimental procedures
Experimental design
Two axenic marine diatoms (T. rotula, CCMP 1647 and
S. costatum, CCMP 1332) were obtained from the ProvasoliGuillard National Center for Culture of Marine Phytoplankton
(CCMP, ME, USA). The algae were incubated in batch cultures in Guillard’s f/2 medium at 15°C in 2.5 l rolling tanks
(5 r.p.m.) and illuminated in a 12:12 h light:dark cycle. For
further experiments, one replicate of each a 12-day-old
(exponentially growing) and a > 8-week-old (stationary
© 2006 The Authors
Journal compilation © 2006 Society for Applied Microbiology and Blackwell Publishing Ltd, Environmental Microbiology, 8, 1074–1084
1082 H.-P. Grossart, G. Czub and M. Simon
phase) diatom culture was inoculated with 10 ml of seawater, containing the natural bacterial community. The seawater was collected in June 2000 from the German Wadden
Sea (53°42′N, 7°50′E). Another replicate of each culture
was incubated under the same conditions, but axenically. As
a further control we inoculated the sterile f/2 medium with
10 ml of natural seawater. Due to experimental limitations
we have only incubated exponentially growing S. costatum
in duplicate. All measured parameters developed in an
almost identical manner in both replicates, indicating a
rather similar development of the algal and bacterial communities when using identical incubation protocols (Grossart
et al., 2005).
Aggregation experiments
Axenic cultures of S. costatum and T. rotula both at exponential (exp) and stationary (stat) growth were incubated at
concentrations of approximately 1000 cells ml−1 in sterile
polyvinyl tanks (1.2 l) rotating horizontally at 6 r.p.m. The
tanks were illuminated for 24 h with either full light
(100 µmol m−2 s−1) or low light (4.0 µmol m−2 s−1). All algal
cultures were inoculated with a natural bacterial community
(Wadden Sea samples) and incubated for at least 65 h at 15–
20°C. Numbers and size classes of formed aggregates were
determined by video and image analysis and growth of diatoms and bacteria as well as concentrations of TEP and CSP
by microscopy (see below).
In an additional experiment, we incubated an axenic exponentially growing T. rotula culture in full light under sterile
conditions for 68 h. Thereafter, the culture was inoculated
with natural seawater bacteria and incubated for another
69 h.
Image analysis
Rolling tanks were illuminated with side-arm focus lamps to
capture sample images with a Sony 900CCD handy cam
(680K gross pixels) in a volume of 45 µm3. In order to analyse
the images, a 250 × 250-pixel window (1 pixel = 60 µm) was
chosen and the images were grey-scale transformed to highlight all aggregates within each window. The transformed
images were used to assess the aggregate abundance, the
equivalent spherical diameter (ESD) and the equivalent
spherical area (mm2) by computer-assisted image analysis
(analySIS V 3.0, Soft Imaging System, Muenster, Germany).
The total aggregate abundance in the experimental tanks
was corrected for the aggregate abundance in the corresponding control tanks.
Sampling
Samples for algal and bacterial abundance, dissolved substrates, bacterial activities, TEP and CSP were collected
under sterile conditions (laminar flow) at regular intervals
and were immediately processed for further tests and analyses. The volume withdrawn was replaced by an equal volume of sterile medium to avoid air bubbles inside the rolling
tanks.
Microparticles
Duplicate subsamples (2 ml each) were filtered onto polycarbonate membranes (0.2 µm pore size) under vacuum (< 10
mbar pressure) to enumerate TEP and CSP. Transparent
exopolymer particle samples were stained with 0.22 µm prefiltered 0.02% Alcian Blue prepared in 0.06% glacial acetic
acid (pH 2.5; Alldredge et al., 1993). Similarly, CSP samples
were stained with 1 ml of 0.04% Coomassie brilliant blue
(G-250) according to Long and Azam (1996). Samples were
filtered dry, placed over a drop of oil on a frosted slide (Cytoclear TM, Poretics, USA), and enumerated by using a Zeiss
Axioplan microscope under bright field illumination and at a
magnification of 100–200×. The abundance of TEP and CSP
in the experimental tanks was corrected for that in the control
tanks.
Algal and bacterial abundance
After staining with DAPI (4′6′diamidino-2-phenolindole) freeliving and attached bacteria from 1 to 5 ml subsamples were
counted on 0.2 and 5.0 µm Nuclepore membranes, respectively, by epifluorescence microscopy (Axioplan, Zeiss, Germany) at 1000× magnification (Porter and Feig, 1980). Algae
were counted by simultaneously using light and epifluorescence microscopy at 400–1000× magnification. A minimum
of 10 replicates was counted for each sample.
Dissolved organic carbon
Samples (10 ml) were collected in glass ampoules after filtration through 0.2 µm polycarbonate membranes (Nuclepore, USA). Samples were acidified with 100 µl of 85%
H3PO4, flame sealed, and stored until analysis at 4°C in the
dark. Dissolved organic carbon was analysed by high temperature combustion (Shimadzu TOC-5000). The standard
deviation between three injections per sample was usually
< 1%. The instrument blank (8–12 µM C) was measured
using UV-irradiated Milli-Q water and was subtracted from
each sample.
Neutral carbohydrates
Samples (10 ml) were filtered using 0.2 µm pore size polycarbonate filters (Nuclepore) and stored frozen at −20°C
until analysis. Concentrations of DFCHO were analysed by
HPLC using a Carbopac PA 10 column (Dionex, USA) and
pulsed amperometric detection (Mopper et al., 1992). NaOH
(20 mM) was used as eluent. Before analysis, samples were
desalted by ion-exchange chromatography according to
Borch and Kirchman (1997). DCCHO were analysed by
HPLC as DFCHO after 20 h of hydrolysis with 0.09 N HCl at
100°C.
Amino acids
Samples (10 ml) were filtered through 0.22 µm pore size low
protein binding acrodisc filters (Pall Corporation) and stored
frozen at −20°C until analysis. Concentrations of DFAA were
© 2006 The Authors
Journal compilation © 2006 Society for Applied Microbiology and Blackwell Publishing Ltd, Environmental Microbiology, 8, 1074–1084
Algae-bacteria interactions 1083
analysed by HPLC after ortho-phthaldialdehyde derivatization (Lindroth and Mopper, 1979). Dissolved combined amino
acids were hydrolysed with 6 N HCl at 155°C for 1 h and
analysed as DFAA.
Bacterial production and hydrolytic enzyme activities
Bacterial production was determined by incorporation of
14
[C]-leucine (14C-Leu; Simon and Azam, 1989). Triplicates
and a formalin-killed control were incubated with 14C-Leu
(292 mCi mmol−1, Hartmann Analytik, Braunschweig, Germany) at a final concentration of 50 nM which ensured saturation of leucine uptake systems. Incubation was performed
in the dark at 15°C for 1 h. Further processing, including
fixation, filtration and radioassay followed the procedure outlined in Grossart and colleagues (2004). BP was calculated
according to Simon and Azam (1989).
Aminopeptidase and β-glucosidase activities were measured using L-leucine-methyl coumarinyl amide (Leu-MCA)
and methyl-umbelliferyl-β-D-glucoside (β-D-Gluc-MUF) as
substrate analogues according to Hoppe (1993). For each
substrate, triplicates and a formalin-killed control were incubated at 15°C in the dark for 1 h. Final concentrations of
substrate analogues were 100 µM which ensured maximum
hydrolysis as determined by saturation kinetics. Fluorescence of both fluorochromes was measured in a TD 700
fluorometer (Turner Design, USA) at 300–400 nm (excitation)
and 410–610 nm (emission).
Acknowledgements
We thank Birgit Kürzel and Rolf Weinert for neutral monosaccharide analysis and Lars Borchers, Isabel Schmalenbach
and Florian Levold for sample collection. This work was supported by the Deutsche Forschungsgemeinschaft within the
Research Group BioGeoChemistry of the Wadden Sea
(FG 432-TP5).
References
Alldredge, A.L., Passow, U., and Logan, B.E. (1993) The
abundance and significance of a class of large, transparent
organic particles in the ocean. Deep-Sea Res I 40: 1131–
1140.
Bates, S.S., Gaudet, J., Kaczmarska, I., and Ehrman, J.E.
(2004) Interaction between bacteria and the domoicacid-producing diatom Pseudonitzschia multiseries (Hasle)
Hasle; can bacteria produce domoic acid autonomously?
Harmful Algae 3: 11–20.
Bidle, K.D., and Azam, F. (1999) Accelerated dissolution of
diatom silica by marine bacterial assemblages. Nature
397: 508–512.
Borch, N.H., and Kirchman, D.L. (1997) Concentration
and composition of dissolved combined neutral sugars
(polysaccharides) in seawater determined by HPLC-PAD.
Mar Chem 57: 85–95.
Engel, A., Thoms, S., Riebesell, U., Rochelle-Newall, E., and
Zondervan, I. (2004) Polysaccharide aggregation as a
potential sink of marine dissolved organic carbon. Nature
428: 929–932.
Grossart, H.-P., and Ploug, H. (2001) Microbial degradation
of organic carbon and nitrogen on diatom aggregates.
Limnol Oceanogr 46: 267–277.
Grossart, H.-P., Brinkhoff, T., Martens, T., Duerselen, C.,
Liebezeit, G., and Simon, M. (2004) Tidal dynamics of
dissolved and particulate matter and bacteria in the German Wadden Sea in spring and fall. Limnol Oceanogr 49:
2212–2222.
Grossart, H.-P., Levold, F., Allgaier, M., Simon, M., and
Brinkhoff, T. (2005) Marine diatom species harbour distinct
bacterial communities. Environ Microbiol 7: 860–873.
Hoppe, H.G. (1993) Use of fluorogenic model substrates for
extracellular enzyme activity (EEA) measurements of bacteria. In Handbook of Methods in Aquatic Microbial Ecology. Kemp, P.F., Sherr, B.F., Sherr, E.B., and Cole, J.J.
(eds). Boca Raton, CA, USA: Lewis Publishers, pp. 423–
432.
Jackson, G.A. (2001) Effect of coagulation on a model planktonic food web. Deep-Sea Res I 48: 95–123.
Kiørboe, T., and Hansen, J.L.S. (1993) Phytoplankton aggregate formation: observations of patterns and mechanisms
of cell sticking and the significance of exopolymer material.
J Plankton Res 15: 993–1018.
Kiørboe, T., Lundsgaard, C., Olesen, M., and Hansen, J.L.S.
(1994) Aggregation and sedimentation processes during a
spring phytoplankton bloom: a field experiment to test
coagulation theory. J Mar Res 52: 297–323.
Lindroth, P., and Mopper, K. (1979) High performance liquid
chromatographic determinations of subpicomole amounts
of amino acids by precolumn fluorescence derivatization
with o-phtaldialdehyde. Anal Chem 51: 1667–1674.
Long, R.A., and Azam, F. (1996) Abundant protein-containing
particles in the sea. Aquat Microb Ecol 10: 213–221.
Mopper, K., Kieber, R.J., and Zhou, X. (1992) Determination
of sugars in unconcentrated seawater and other natural
waters by liquid-chromatography and pulsed amperometric
detection. Environ Sci Technol 26: 133–138.
Passow, U. (2002a) Transparent exopolymer particles (TEP)
in aquatic environments. Prog Oceanogr 55: 287–333.
Passow, U. (2002b) Production of transparent exopolymer
particles (TEP) by phyto- and bacterioplankton. Mar Ecol
Prog Ser 238: 1–12.
Passow, U., and Alldredge, A.L. (1995) Aggregation of a
diatom bloom in a mesocosm – the role of transparent
exopolymer particles (TEP). Deep-Sea Res II 42: 99–109.
Passow, U., Alldredge, A.L., and Logan, B.E. (1994) The role
of particulate carbohydrate exudates in the flocculation of
diatom blooms. Deep-Sea Res I 41: 335–357.
Porter, K.G., and Feig, Y.S. (1980) The use of DAPI for
identification and enumeration of bacteria and blue-green
algae. Limnol Oceanogr 13: 389–398.
Rooney-Varga, J.N., Giewat, M.W., Savin, M.C., Sood, S.,
LeGresley, M., and Martin, J.L. (2005) Links between phytoplankton and bacterial community dynamics in a coastal
marine environment. Microb Ecol 49: 163–175.
Schäfer, H., Abbas, B., Witte, H., and Muyzer, G. (2002)
Genetic diversity of ‘satellite’ bacteria present in cultures
of marine diatoms. FEMS Microbiol Ecol 42: 25–35.
Simon, M., and Azam, F. (1989) Protein content and protein
synthesis rates of planktonic marine bacteria. Mar Ecol
Prog Ser 51: 201–213.
© 2006 The Authors
Journal compilation © 2006 Society for Applied Microbiology and Blackwell Publishing Ltd, Environmental Microbiology, 8, 1074–1084
1084 H.-P. Grossart, G. Czub and M. Simon
Simon, M., Grossart, H.-P., Schweitzer, B., and Ploug,
H. (2002) Microbial ecology of organic aggregates in
aquatic ecosystems. Aquat Microb Ecol 28: 175–
211.
Smith, D.C., Simon, M., Alldredge, A.L., and Azam, F. (1992)
Intense hydrolytic enzyme-activity on marine aggregates
and implications for rapid particle dissolution. Nature 359:
139–142.
Smith, D.C., Steward, G.F., Long, R.A., and Azam, F. (1995)
Bacterial mediation of carbon fluxes during a diatom bloom
in a mesocosm. Deep-Sea Res II 42: 75–97.
Thornton, D.C.O. (2002) Diatom aggregation in the sea:
mechanisms and ecological implications. Eur J Phycol 37:
149–161.
Tiselius, P., and Kuylenstierna, M. (1996) Growth and decline
of a diatom spring bloom: phytoplankton species composition, formation of marine snow and the role of heterotrophic
dinoflagellates. J Plankton Res 18: 133–155.
Zhou, J., Mopper, K., and Passow, U. (1998) The role of
surface-active carbohydrates in the formation of transparent exopolymer particles by bubble adsorption of seawater.
Limnol Oceanogr 43: 1860–1871.
© 2006 The Authors
Journal compilation © 2006 Society for Applied Microbiology and Blackwell Publishing Ltd, Environmental Microbiology, 8, 1074–1084