Content uploaded by Dale A. Casamatta
Author content
All content in this area was uploaded by Dale A. Casamatta on Nov 06, 2014
Content may be subject to copyright.
Sensitivity of Two Disjunct Bacterioplankton Communities to
Exudates from the Cyanobacterium Microcystis aeruginosa
Ku¨tzing
D.A. Casamatta,
1
C.E. Wickstrom
2
1
Department of Environmental and Plant Biology, Ohio University, Athens, Ohio 45701, USA
2
Department of Biological Sciences, Kent State University, Kent, Ohio 44242, USA
Received: 6 December 1999; Accepted: 3 April 2000; Online Publication: 18 July 2000
ABSTRACT
Microcystis aeruginosa Ku¨tzing releases a variety of bioactive compounds during growth. This study
determined whether bacteria from communities co-occurring (M+) or not (M-) with this cosmo-
politan cyanobacterium respond similarly to its products. Fifty M+ bacteria from a M. aeruginosa
bloom site (Western Basin of Lake Erie) and 50 M- bacteria from a Microcystis-free site (East Twin
Lake, Portage Co., OH) were isolated and grown on Standard Methods Agar. Three levels of testing
were performed: chemotaxis, antibiotic response, and 48-h cell abundance. Chemotaxis was com-
pared using capillary tubes placed in contact with bacterial, Standard Methods Broth (SMB) sus-
pensions. The capillary choices were conditioned SMB, M. aeruginosa exudate, and BG-11. M+
bacteria showed significantly greater (Tukey’s test, p< 0.005) positive chemotaxis to M. aeruginosa
exudate compared to control conditions and to M-strains. The latter showed a negative chemotactic
response to M. aeruginosa exudate compared to control conditions. Antibiotic response was tested
by sensitivity disk assays, first using M. aeruginosa exudates, whole cells, and homogenized cells, and
then placing the disks on bacterial lawns of each strain. M+ bacteria were significantly more
resistant to inhibition than M- bacteria (chi-square test, p< 0.01). M. aeruginosa exudate, BG-11
algal medium, SMB, and distilled water effects on 48-h abundance of the strains were compared.
The M- community bacteria exhibited significantly lower growth yields (Tukey’s comparison of
means test, p< 0.005) in M. aeruginosa exudate than did the M+ strains. It is evident that those
bacteria co-occurring with M. aeruginosa are more likely to be attracted to it, able to withstand
exposure to it, and able to utilize its products without inhibition than are bacteria from commu-
nities without previous exposure to this cyanobacterium.
Correspondence to: D.A. Casamatta; Fax: (740) 593-1130; E-mail:
dc274389@ohio.edu
MICROBIAL
ECOLOGY
Microb Ecol (2000) 41:64–73
DOI: 10.1007/s002480000035
© 2000 Springer-Verlag New York Inc.
Introduction
As algae photosynthesize, both excreted and secreted extra-
cellular products (ECP) are released into the environment
[19, 8]. Typical ECP include, but are not limited to, carbo-
hydrates, nitrogenous substances, organic acids, lipids, phos-
phatases, and other enzymes [28]. The amount and nature of
exuded products depends on several factors. The taxon of
the alga dictates the compositions of compounds released
[e.g., 28]. Environmental conditions and algal age also may
significantly affect exudate release [24, 50, 51].
Bacteria are the primary assimilators of dissolved organic
carbon in aquatic ecosystems [5]. The bacterioplankton may
respond to different types of carbon compounds products by
altering their metabolic pathways and rates of growth and
activity [52], or the bacteria may be unable to use or with-
stand the compounds, resulting in their decline and eventual
loss from the community.
In an effort to examine some of the effects that a photo-
synthesizing alga may exert on surrounding bacterioplank-
ton, we studied the cyanobacterium Microcystis aeruginosa
Ku¨tzing. Microcystis spp. are cyanobacteria with coccoid cells
loosely arranged within polymorphic mucilaginous colonies.
This genus is possibly the most cosmopolitan cyanobacte-
rium found in freshwater [16] and typically has large num-
bers of bacteria within and/or attached to their mucilaginous
colonies [44].
Microcystis, a non-nitrogen-fixer, may be planktonic or
benthic depending on environmental conditions such as
light and nutrient levels [40]. Some strains of Microcystis are
a major source of freshwater biotoxins [3]. In particular,
they may produce bioactive secondary metabolites known as
microcystins [15, 16]. The release of these compounds from
Microcystis depends on several factors such as culture age,
light levels, nutrient levels, and temperature [49].
In addition to being toxic to some metazoans, Microcystis
may be toxic or inhibitory to other microbes. Allelopathic
interactions are widespread among algal taxa and may play
an important role in the structuring of algal communities
[31, 32]. Microcystis, in particular, has been shown to posses
allelopathic properties against several taxa of algae such as
Cryptomonas ovata, Chlorella sp., Ceratium sp., and Ana-
baena oscillariodes [33, 47, 48]. The impact of this cyano-
bacterium on the bacterioplankton, however, has not been
previously examined.
The purpose of this study is to examine some of the
effects of Microcystis aeruginosa (Ma) on two disjunct cul-
tured bacterioplankton communities. Three pairs of ques-
tions are addressed in this paper: (1) Do the cultured envi-
ronmental bacteria respond chemotactically to the presence
of M. aeruginosa extracellular products? Is positive chemo-
taxis more common among those bacteria that co-occur
with this species, and is negative chemotaxis more common
among those non-co-occurring species? (2) Are strains of
environmental bacteria inhibited by M. aeruginosa products?
More specifically, are the bacteria cultured from Lake Erie
inhibited by Lake Erie M. aeruginosa products, and are the
bacteria from M. aeruginosa regions collectively more resis-
tant to Microcystis inhibition than those bacteria isolated
from a region free of M. aeruginosa? (3) Does Lake Erie
Microcystis aeruginosa stimulate or inhibit the growth of bac-
teria isolated from Lake Erie, and does it stimulate naturally
co-occurring bacteria to a greater degree than it stimulates
bacteria that do not co-occur with it?
Materials and Methods
Cultures
A uni-algal culture of M. aeruginosa was obtained from Dr. Wayne
W. Carmichael (Wright State University, OH). This strain had been
isolated from the western basin of Lake Erie during the summer of
1996. An axenic isolate was established by streak plate isolation on
BG-11 agar medium [45]. Cultures were grown in BG-11 medium
under 20–40 µE m
-2
s
-1
coolwhite fluorescent illumination with an
18-h photoperiod at room temperature, 20–24°C.
Fifty bacterial isolates were established in June of 1996 from the
western basin of Lake Erie, near South Bass Island. Water taken by
Kemmerer bottle 1 m above a benthic zebra mussel bed which had
an associated population of M. aeruginosa was transported on ice
back to the lab. One-ml subsamples were plated on Standard Meth-
ods Agar (SMA) and plates were incubated at room temperature
for 4 days. To obtain as high a diversity of strains as possible, 50
isolates were selected based on visual examination of colony char-
acteristics, including colony morphology, color, and texture. Iso-
lates collected concurrently with M. aeruginosa were designated as
Ma+.
An additional 50 bacterial clones were obtained from East Twin
Lake (ETL), Portage Co., OH. Five water samples were taken with
a Kemmerer Bottle from a depth approximately 2 m below the
surface and 1 m above the sediments. Two, 1-ml aliquots from each
samples were plated on SMA and incubated at room temperature
for 4 days. These clones were isolated, examined for contamination,
and maintained on SMA as described for the Ma+ isolates. ETL was
determined to be free from M. aeruginosa at the time of collection
by microscopic examination of water samples and was reported to
be free of this blue-green for at least the past 10 years (D. Fleish-
man, personal communication; Kent State University). These iso-
lates, from a Microcystis-free area, were designated Ma- strains.
Bacterial Sensitivity to M. aeruginosa 65
M. aeruginosa Test Preparation
M. aeruginosa exudate (growth medium and associated algal prod-
ucts) was obtained from 4-week old stationary phase cultures of the
alga growing in BG-11. Two 1-liter cultures were centrifuged (In-
ternational Equipment Co.) at 4,000 revolutions per min for 10
min (relative centrifugal force = 1,897 g) and the supernatant was
decanted into sterile 250 ml Erlenmeyer flasks. To remove remain-
ing cells and cellular debris, the supernatant was filtered sequen-
tially through Whatman GF/C glass fiber, 1.0 µm, and a 0.2 µm
nitrocellulose (Millipore Corporation, Bedford, MA) filters with a
vacuum pump.
Intact and homogenized cell treatments were prepared by air-
drying two 1.2 g wet weight samples of M. aeruginosa for 48 h.
These cyanobacterial samples were obtained during the initial steps
of exudate collection. The dried samples were then resuspended in
30 ml distilled water to give a final concentration of 40 mg M.
aeruginosa ml
-1
. This concentration is similar to that used by others
in toxicity testing [e.g., 36, 37]. One of the intact cell samples was
then homogenized for 1 min with a Brinkmann Polytron Homog-
enizer. Aliquots of the homogenized sample were microscopically
inspected at 600 and 1,000× magnification to confirm cell rupture.
These preparations were stirred with a magnetic stirrer to resus-
pend cells and cell debris prior to disk saturation.
Chemotaxis
Bacterial chemotaxis in response to M. aeruginosa exudate was
tested as described by Adler [1] and modified in Mandimba et al.
[38]. Twenty-four h Standard Methods Broth (SMB) cultures of all
strains were inspected for motility by Zeiss phase-contrast micros-
copy. Only motile strains were subsequently tested for chemotaxis.
Thus, 32 clones from Lake Erie and 29 clones from East Twin Lake
were used in this part of the study.
Motile bacteria were inoculated into culture tubes containing 10
ml SMB and incubated for 48 h. At that time, a 0.5 ml aliquot of
the new bacterial culture was placed in the center of a sterile glass
petri dish bottom. Three sterile glass 100 µl capillary tubes were
filled with one of three fluids by capillary action. The first tube
contained M. aeruginosa exudate; the second and third tubes were
controls and contained BG-11 medium and conditioned SMB me-
dium, respectively. The latter control was derived from the 48-h
SMB cultures filtered through a 0.2 µm filter. Each pipette was
filled by passing one end through a flame and then drawing up the
individual attractants by capillary action through the unheated end.
One tube of each treatment was then rested at an angle on rectan-
gular Styrofoam blocks with one end of the capillary tube placed
into contact with the centrally located test bacterium culture drop-
let. A cotton ball saturated with sterile distilled water was added to
each plate to retard evaporation. The plates were then closed and
incubated at room temperature [38].
The capillary tubes were carefully removed from the plates at
the end of 1 h, and the ends that had been in contact with the
bacterial suspension were rinsed with autoclaved distilled water
into separate, sterile 1.5 ml microcentrifuge tubes. Capillary tube
contents were then flushed into the microcentrifuge tubes. Samples
were preserved and cell counts performed using a Petroff–Hausser
counting chamber. These tests were run in triplicate.
Antibacterial Testing
The potential antibacterial properties of M. aeruginosa exudate as
well as whole and homogenized M. aeruginosa cells were tested
against each of the 100 bacterial isolates. Isolates were grown in
SMB at 20–24°C for 24 h in preparation for lawn inoculation.
Eight-mm diameter filter paper disks punched from larger GF/C
filters were autoclaved and then dipped with sterile forceps into
beakers containing one of the three algal preparations: M. aerugi-
nosa exudate, whole cells, and homogenate. The saturated disks
were placed into separate sterile glass petri dishes and air dried at
room temperature. Three disks of each treatment were then posi-
tioned on the bacterial lawn of each strain. A BG-11 disk placed in
the center of the plate served as the control disk. Presence of a zone
of inhibition was recorded after 24 h; the control did not inhibit
any strains.
Growth Studies
In preparation for the growth studies, the 100 bacterial isolates
were transferred from SMA slants to sterile culture tubes of SMB
and incubated for 48 h. One-ml aliquots from the SMB culture of
each isolate were then inoculated into 12 sterile culture tubes con-
taining 5 ml SMB. To each set of three tubes, 5 ml of one of the
following were added: SMB, M. aeruginosa exudate, or sterile dis-
tilled water (DW).
After addition of the treatments to the bacterial suspensions, the
time-zero turbidity of each culture was measured at a wavelength of
750 nm with a Bausch and Lomb Spectronic 100 spectrophotom-
eter (Bausch and Lomb, Rochester, NY). Turbidity was again mea-
sured after 48 h and corrected for initial turbidity. Preliminary
trials with 15 isolates showed that the populations entered station-
ary phase after 48 h. The 48 h corrected response turbidity was used
as a measure of total growth response (biomass increase) to the
experimental and control treatments of the same isolate.
Results
Bacterial Chemotactic Response to Microcystis
aeruginosa Exudate
Ma+ strains showed no significant differences in number of
bacterial cells attracted between SMB and BG-11. However,
Ma exudate attracted a significantly greater number of bac-
teria (Fisher’s LSD test, p< 0.05) than either of the controls:
ca. twofold more bacteria accumulated in these capillary
tubes (Fig. 1). Ma- strains showed significant differences
among treatments. Conditioned SMB controls elicited sig-
nificantly greater (p< 0.05) response by these bacteria than
66 D.A. Casamatta, C.F. Wickstrom
did BG-11 with or without exudate. The latter attractants
were not significantly different from each other.
Post ANOVA Fisher’s LSD comparison of bacterial abun-
dances of sites and attractant choices showed significant dif-
ferences (p< 0.05) between the bacteria from the two sites
(Fig. 1). Conditioned SMB attracted similar numbers of bac-
teria (p> 0.05) among Ma+ and Ma- strains. BG-11, on the
other hand, attracted approximately twice as many bacteria
(p< 0.05) among the Ma+ strains than Ma- strains. The
most pronounced difference between sites, however, was
with Ma exudate. Ma+ strains showed an approximately
fivefold greater attraction (p< 0.05) than did Ma- strains.
Fisher LSD tests compared chemotactic responses of the
motile Ma+ strains of bacteria to Ma exudates and BG-11
after 1 h (Fig. 1). Each challenge was categorized according
to the significance, if any, of differences between those
means. In 63% of the strains (i.e., 20 out of 32), Ma exudates
elicited a significantly greater (Fisher’s LSD p< 0.05) posi-
tive chemotactic response than did BG-11. Nineteen percent
of the Ma+ strains were more attracted to the BG-11 control
than exudate, and 19% of these strains showed no prefer-
ence. When Ma- strains were similarly compared only 14%
(4 out of 29) favored exudate while 48% (14 of 29) re-
sponded positively to the control. No significant (p> 0.05)
preference was noted for 38% (11 strains) of Ma- strains
(Table 1).
Antibacterial Properties of Microcystis aeruginosa
Susceptibility of Ma+ and Ma- bacteria to M. aeruginosa
disks with exudates, whole cells, and homogenized cells was
tested. Controls consisted of disks saturated with the M.
aeruginosa medium, BG-11. Strains were scored as inhibited
if a visible zone of inhibition (ca. 1 mm) developed around
2 out of 3 disks. All of the bacteria developed lawns on the
SMA plates, and no strains of bacteria were inhibited by the
control BG-11 disks.
The presence or absence of zones of inhibition surround-
ing the variously prepared disks after 24 h were compared
between and within Ma+ and Ma- strains (Table 2). A Yates-
corrected chi-square test [54] revealed that significantly (p<
0.05) more Ma- strains (n= 24) were inhibited by M. aeru-
ginosa exudate than by either whole cells or homogenized
cells. Ma+ strains showed no significant variation in re-
sponse to exudate, whole, and homogenized cell treatments
with a mean inhibitory response of 17%. Only 20% of Ma+
strains were inhibited by Ma exudate while 48% of Ma-
strains showed exudate inhibition (Table 2). Inhibition by
the other treatments followed similar patterns, with sensi-
tivity to whole cells of 16% Ma+ vs 32% Ma- and 16% vs
42% inhibition by homogenized cells. In all treatments, at
least twice as many Ma- strains were susceptible to M. ae-
ruginosa products than were Ma+ strains.
Effects of M. aeruginosa on growth of isolates
The growth of Ma+ and Ma- strains in response to Standard
Methods Broth (SMB), distilled water (DW) controls, and
M. aeruginosa exudates is illustrated in Fig. 2. Analysis of
variance (ANOVA) revealed significant (p< 0.05) yield dif-
ferences within as well as between isolation site groupings.
Fisher’s LSD comparison of means showed some significant
differences (p< 0.05) between the treatments (i.e., growth in
Fig. 1. Effect of attractant on habitat-specific bacterial chemotax-
is. Bars indicate mean (+1 SE) abundances of bacteria from sites
with (Ma+) or without (Ma-) M. aeruginosa. Statistical categories,
A–C, are indicated above the bars; category A chemotactic re-
sponses are not significantly different from each other (Fisher’s
LSD comparison, p> 0.05), but are significantly different (p< 0.05)
from those in categories B and C. Similarly, categories B and C are
statistically unique.
Table 1. Distribution of Ma+ and Ma- strains among chemotaxis
categories
a
M. aeruginosa Total no.
of strains
Chemotaxis categories
Habitat X > C
b
X=C X<C
Present (Ma+) 32 20 (63%) 6 (19%) 6 (19%)
Absent (Ma-) 29 4 (14%) 11 (38%) 14 (48%)
a
Chemotaxis categories based on t-test comparison (␣=0.05;df=5)of
exudate versus BG-11 (control) bacterial abundances following strain che-
motaxis tests
b
X denotes exudate; C denotes control.
Bacterial Sensitivity to M. aeruginosa 67
Ma exudate, SMB, and distilled water), as well as between
the two collection sites (p< 0.05). Among Ma+ strains, only
DW showed significantly (Fisher’s LSD p< 0.05) lower
growth yields, amounting to approximately 75% of the other
treatments. All other Ma+ growth yields were statistically
similar. Ma- strains showed significantly lower growth yields
in exudate, BG-11, and DW compared to the SMB controls,
although they were not statistically different from each
other.
M. aeruginosa exudate elicited significantly greater
growth with Ma+ strains than with Ma- strains (Fisher’s LSD
p< 0.05), and the stimulated Ma+ strains were no different
from those growing solely on SMB. SMB and DW growth
was not statistically different with respect to collection sites
(p> 0.05).
Discussion
Under natural field conditions when bacteria encounter an
alga releasing extracellular products (ECP), the motile mem-
bers of the bacterial community may respond chemotacti-
cally to the exudates [12, 30, 7]. These bacteria may be
attracted to, repulsed from, or indifferent to the ECP. We
anticipated that those motile bacteria co-occurring with M.
aeruginosa would be more frequently attracted to its exu-
dates than bacteria from a Microcystis-free site. From this
response, the bacteria would be expected to show a predict-
able pattern of antibacterial susceptibilities and growth rates
depending on the history of their habitat.
Except for systems receiving major inputs of allochtho-
nous organic matter, most bacteria depend largely on phy-
toplankton as a source of carbon [10]. Photosynthesizing
algae are typically the most readily available in situ source of
organic nutrients to bacterioplankton in oligotrophic waters
[22]. As such, bacteria tend to be most metabolically active
in the vicinity of metabolizing algal cells, forming an algal–
bacterial consortium. As a result of their dependence on
algae as a source of carbon, bacterioplankton blooms typi-
cally co-occur with elevated algal numbers [9, 20].
Researchers have documented significantly greater num-
bers of active bacteria in association with Microcystis sheaths
than were found free-living in the same waters, suggesting
that Microcystis spp. colonies may serve as incubators for the
surrounding bacterial community [53]. In contrast, Brun-
berg [14] found that bacteria associated with Microcystis
colonies in the water column were less active than ambient
water column bacteria. Several reasons for lower relative
attached bacterial production were postulated, including
lower bacterial growth efficiency, slow diffusion of products
within Microcystis mucilage, and lower mortality.
Chemotactic Response of Bacteria to Microcystis Exudate
The ability to actively seek out or escape M. aeruginosa exu-
dates would provide an advantage to those bacteria that
either utilize or are inhibited by these compounds, respec-
tively. Our experiments indicated that there is a significant
difference between the chemotactic response of co-occurring
versus non-co-occurring bacterioplankton. We hypothesize
that some compound(s) released by photosynthesizing Mi-
crocystis activates a response mechanism in a large set of the
bacterioplankton community, eliciting either a positive (to-
ward the exudate) or negative (away from) response. This
response may have a significant effect on the dynamics of the
Table 2. Inhibition of Ma+ and Ma- bacterial strains by M. ae-
ruginosa cells and cell products
No. of Ma+ strains No. of Ma- strains
Disk treatment Inhibited
Not
inhibited Inhibited
Not
inhibited
Ma exudate 10 (20%) 40 (80%) 24 (48%) 26 (52%)
Whole cells 8 (16%) 42 (84%) 16 (32%) 34 (68%)
Homogenate 8 (16%) 42 (84%) 21 (42%) 29 (58%)
Total 26 124 61 89
Fig. 2. Growth response of Ma+ and Ma- bacteria to growth
medium amendments (M. aeruginosa exudate; Standard Methods
Both, SMB; and distilled water, DW). Mean 48-h optical densities
(+1 SE) are shown. Statistical categories (Fisher’s LSD multiple
comparison test) are marked A and B; category A growth yields are
not significantly different from each other (p> 0.05), but are sig-
nificantly different (p< 0.05) from those in category B.
68 D.A. Casamatta, C.F. Wickstrom
microbial assemblage, such as serving to alter nutrient par-
titioning within the microbial consortium [11]. Alterna-
tively, bacterial clustering may have little effect on nutrient
cycling [12], and clustering may be the result of unidirec-
tional nutrient flow from phytoplankton to bacterioplank-
ton.
From our experiment, we suspect that chemotactic bac-
teria will respond to Microcystis in natural settings. The bac-
terioplankton response will depend to a large extent on prior
contact. This may be in response to any number of potential
exudate compounds, such as carbon or some other micro-
nutrient. Reynolds [41] reports that buoyant Microcystis
colonies are capable of moving up to 3.2 m day
-1
, although
colonies often move much less. As the colony moves through
the water column, chemotactically active bacteria relying on
Microcystis exudates would be able to maintain contact with
their carbon source. Jackson [30] hypothesized that even
bacterioplankton with limited motility may benefit from ex-
posure to phytoplankton exudate. Our results are consistent
with this hypothesis. Not only were co-occurring bacteria
chemotactically attracted to Microcystis exudates, but some
non-co-occurring strains were as well. The exact nature of
the chemotactic substance remains beyond the scope of this
paper, but others have noted a wide range of compounds
that elicit a chemotactic response at varying concentrations
[i.e., 12].
Vaque et al. [46] hypothesized that one possible reason
bacterial cells become metabolically inactive in aquatic sys-
tems is to save energy until they encounter either algae or
other carbon sources, especially when resources are tempo-
rally or spatially patchy. Bacteria that are capable of positive
chemotactic response to Microcystis exudates would exhibit
an advantage over other bacteria as Microcystis both moves
through the water column and propagates new colonies.
Therefore, a positive chemotactic response would be favored
as a way of obtaining those carbon products that serve to
most stimulate the bacterial cells (Fig. 3). Conversely, a
negative chemotactic response may be favored as a way of
avoiding potentially harmful products. East Twin Lake bac-
terial isolates exhibited significantly greater negative or neu-
tral chemotactic response to M. aeruginosa exudates than did
bacteria co-occurring with the cyanobacterium.
Antibacterial Properties of Microcystis
M. aeruginosa exudates inhibited ca. twice as many Micro-
cystis-free bacteria as bacteria isolated from the Lake Erie
site. Inhibition by intact and homogenized cells followed a
similar pattern, although some variation in strain suscepti-
bility did occur. This amensalistic action by M. aeruginosa
may be a way to gain a competitive advantage over co-
occurring bacteria for contested nutrients, or maybe a by-
product of a metabolic pathway.
Microcystis often forms benthic mats that may float to the
surface as the population ages [13]. As senescence begins and
cells lyse, the surrounding bacterial community may also be
exposed to intracellular compounds, represented by the ex-
perimental homogenate. These compounds may be very dif-
ferent from the typical exudates released during photosyn-
thesis and, as such, may exert a separate selective pressure(s).
Those bacteria associated with Microcystis showed little
variation from the response to exudates or homogenized
cells. Nonassociated bacteria, however, showed greater varia-
tion to the different experimental treatments than did asso-
ciated bacteria. This implies some degree of differential ef-
fects by exudates and homogenized cells on nonacclimated
and/or nonselected bacteria.
A wide array of freshwater algae, including Bacillari-
ophyta, Pyrrophyta, and Chlorophyta, as well as Cyanobac-
teria, produce antibacterial substances [e.g., 23, 27]. The
results of the present study indicated that Microcystis pos-
sessed antibacterial properties that affected bacteria from M.
aeruginosa and M. aeruginosa-free habitats in the anticipated
manner (Fig. 3). Between Microcystis exudates, intact cells,
and homogenized cells, exudates inhibited the greatest num-
ber of environmental bacteria from both sites. During its
Fig. 3. Pathways of potential outcomes for populations within the
bacterioplankton community after encountering or introduction of
M. aeruginosa in an aquatic habitat. See text for discussion.
Bacterial Sensitivity to M. aeruginosa 69
growth phase, M. aeruginosa is expected to release exudates
as a result of cell metabolism and photosynthesis as shown
for other cyanobacteria [24]. The amount and composition
of these exudates, however, may depend on a variety of
factors, such as nutrient concentrations [2] and light levels
[49].
Our experiments showed that Microcystis had toxic prop-
erties to some bacterioplankton strains. In natural condi-
tions, interactions between bacterioplankton and Microcystis
may have additional components, and as a result the inter-
actions may not be limited to allelopathy. For example, bac-
teria are considered superior competitors for phosphorous, a
common limiting nutrient in freshwater systems [21]. This
may create a pool of nutrients around the metabolizing algal
cell, in which the bacteria sequester the phosphorus but are
in turn carbon limited. There is a great degree of recycling
and turnover in aquatic ecosystems, and so the Microcystis
mat will probably benefit from nutrient regeneration en-
abled by close proximity to the active bacterial community.
For example, the surrounding bacteria may be a nutrient
reserve that immediately surrounds the photosynthesizing
algal cells. Turnover of nutrients from one group may be
made readily available to the other. In our cultures, exudates
rather than active colonies were employed, so competition
for nutrients cannot be assessed, nor can the likelihood of
nutrient regeneration or feedback. However, the overall
metabolic and chemotactic response of the bacteria to the
exudates indicates that it is reasonable that the Microcystis
colonies may be metabolically coupled to bacteria in natural
systems.
Bacterial Growth with Microcystis Exudates
The growth experiments indicated that, as a group, the bac-
terial strains isolated from Lake Erie are stimulated to a
significantly (p< 0.05) greater degree by M. aeruginosa exu-
dates than the bacterial strains isolated from East Twin Lake.
Both sets of community isolates responded to SMB and dis-
tilled water in similar manners. Worm and Søndergaard [52]
report that bacteria associated with Microcystis show signifi-
cantly greater metabolic activity when coupled with Micro-
cystis than those bacteria found free-living in the lake. They
conclude that Microcystis exudates could drive the metabolic
activity of the associated bacterial community under carbon-
limited conditions.
Our experimental results concur with the above conclu-
sion, and we hypothesize that in natural conditions bacterial
nutrient remineralization can recycle nutrients to Microcystis
colonies, as others have suggested [18]. This coupling may
benefit all members of the consortium, especially if bacteria
attached to Microcystis are more efficient nutrient recyclers
than the surrounding bacterioplankton community. This
would benefit both partners, with the bacterioplankton pro-
viding nutrients (i.e., phosphorus, nitrogen, etc.) and the
photosynthesizing alga providing a readily usable carbon
source. The maintenance of bacterial epiphytes in M. aeru-
ginosa mucilage may also be commensalistic, with the bac-
teria obtaining nutrients from Microcystis exudates. Con-
versely, the bacteria may obtain exuded products and also
prove superior competitors to Microcystis for other nutrients
(e.g., phosphorus). As an amensalistic infochemical(s), M.
aeruginosa ECP may repulse potentially harmful bacteria
that would otherwise successfully compete for phycosphere
space with bacteria that might be superior nutrient recyclers.
Amensalism of this nature would allow preemptive coloni-
zation [4] of the colony by beneficial bacteria. The testing of
these ideas, however, were beyond the scope of this project.
Since bacterioplankton typically rely on algal exudates for
growth [6, 34], those bacteria co-occurring with Microcystis
should be able to utilize Microcystis ECP. The rationale is
that historical contact with Microcystis has selected against
those community bacteria inhibited by or unable to use this
ECP (Fig. 3), so they would be absent. Bacteria from com-
munities not associated with Microcystis may not be able to
utilize those compounds fully or may need to acclimate to
those specific carbon compounds when exposed to them.
Therefore, they would exhibit lowered growth if not inhib-
ited or destroyed outright. Murray et al. [39] show greater
growth yields among bacteria already involved in an algal–
bacterial consortium with diatoms than among those bacte-
ria newly inoculated into such an association.
The results of the growth experiments imply that Micro-
cystis may affect the surrounding bacterial community.
Herndl [29] suggests that algal–bacterial coupling is a
mechanism to increase nutrient regenerative processes, and
as a result increase overall microbial activity. Collectively,
the isolated Lake Erie bacterial community members were
able to use M. aeruginosa exudates just as efficiently as SMB,
whereas bacteria from ETL were not. Bacteria not inhibited
by Microcystis ECP were typically stimulated by that exudate.
Microcystis exudates may be a significant source of dissolved
organic carbon, especially under conditions where Microcys-
tis is the dominant alga. High levels of Microcystis ECP
70 D.A. Casamatta, C.F. Wickstrom
would confer a selective advantage to those bacteria capable
of this ECP utilization.
Figure 3 presents a conceptual model of M. aeruginosa
impact on bacterial communities. Those bacteria exhibiting
positive chemotactic response to M. aeruginosa exudates are
hypothesized to show no susceptibility to subsequent anti-
bacterial compounds and to respond positively to M. aeru-
ginosa exudates. Conversely, those bacteria showing negative
chemotaxis would be expected to exhibit greater levels of
antibacterial susceptibility and lower levels of exudate utili-
zation. Overall, those bacteria isolated from a M. aeruginosa
site showed a greater ability to move toward and utilize its
extracellular products than did the disjunct bacterial isolates.
Members of Lake Erie’s bacterioplankton community in Mi-
crocystis areas may depend on M. aeruginosa products and
may have been accordingly selected through historical ex-
posures to this species. If a bloom of M. aeruginosa occurred
in East Twin Lake, many of its bacterial populations would
suffer, and the bacterioplankton community would likely
undergo appropriate successional changes.
A wide range of chemical cues are employed in structur-
ing the microbial community, both within and among taxa
[e.g., 25, 35, 43, 17]. We have shown that co-occurring bac-
teria are more responsive to Microcystis exudates than non-
co-occuring strains. One possible mechanism for this could
be the presence of chemical signals acted upon by the bac-
terioplankton community, a process that has been shown to
be important in many taxonomic groups of bacteria [e.g.,
26]. It may be these infochemicals that the bacterioplankton
are responding to. We also noted an antibiotic effect of
Microcystis mainly on non-co-occurring bacteria, but also on
some associated strains. Although this was not specifically
tested, we hypothesize that as stressful conditions occur,
such as from nutrient limitation, a greater level of inhibitory
compounds may be released, as is observed with bacteriocins
[42]. In natural settings, metabolically coupled bacteria and
Microcystis, or any phytoplankton taxa, may have a greater
level of productivity than either alone [11]. Our experiments
suggest that this may be a valid assumption, as one half of
this association, the bacterioplankton community, did re-
spond positively to Microcystis exudates.
There are several possible mechanisms for associations
between Microcystis and bacterioplankton in natural settings.
One excellent possibility deals with nutrient recycling be-
tween the photosynthesizing alga and the associated bacteria.
The nature of this interaction, however, is subject to much
speculation [e.g., 12, 11, 20, 22]. Our experiments showed
that Microcystis exudate has at least three distinct avenues of
impact on the bacterioplankton communities—chemotaxis,
as a source of growth nutrients, and antibiosis—and that
community responses to these three were dependent on
their exposure history.
Acknowledgments
The authors extend their thanks to Dr. Wayne Carmichael
for the generous contribution of Microcystis aeruginosa. In
addition, the authors thank two anonymous reviewers for
constructive comments with an earlier version of this manu-
script. This work was supported by the Department of Bio-
logical Sciences, Kent Sate University, and was partially sup-
ported by grant #97-18 from the Lake Erie Protection Fund.
References
1. Adler J (1973) A method for measuring chemotaxis and use of
the method to determine optimum conditions for chemotaxis
by Escherichia coli. J Gen Microbiol 74:77–91
2. Aguiar DG, Azevedo SMFO (1998) Effect of different nutrient
concentrations on growth and hepatotoxin production by Mi-
crocystis aeruginosa (Cyanobacteria). Verh Internat Verein
Limnol 26:1657–58
3. Arment AA, Carmichael WW (1996) Evidence that microcys-
tin is a Thiotemplate product. J Phycol 32:591–97
4. Atlas RM, Bartha R (1998) Microbial Ecology: Fundamentals
and Applications, 4th ed. Benjamin/Cummings Science Pub-
lishing, California
5. Azam F, Smith DC, Steward GF, Hagstro¨m A
˚(1993) Bacteria–
organic matter coupling and its significance for oceanic car-
bon cycling. Microb Ecol 28:167–179
6. Becquevort S, Rousseau V, Lancelot C (1998) Major and com-
parable roles for free-living and attached bacteria in the deg-
radation of Phaeocystis-derived organic matter in Belgian
coastal waters of the North Sea. Aquat Microb Ecol 14:39–48
7. Bell WH, Mitchell R (1972) Chemotactic and growth re-
sponses of marine bacteria to algal extracellular products. Biol
Bull 143:265–277
8. Bell WH, Sakshaug E (1980) Bacterial utilization of algal ex-
tracellular products. 2. A kinetic study of natural populations.
Limnol Oceanogr 25:1021–1033
9. Bird DF, Kalff J (1984) Empirical relationships between bac-
terial abundance and chlorophyll concentration in fresh and
marine waters. Can J Fish Aquat Sci 41:1015–1023
10. Billen G (1990) Delayed development of bacterioplankton
with respect to phytoplankton: a clue for understanding their
trophic relationships. Arch Hydrobiol Beih 34:191–201
11. Bowen JD, Stolzenbach KD, Chisholm SW (1993) Simulating
Bacterial Sensitivity to M. aeruginosa 71
bacterial clustering around phytoplankton cells in a turbulent
ocean. Limnol Oceanogr 38:36–51
12. Brett MT, Lubnow FS, Villar-Argaiz M, Mu¨ller-Solger A,
Goldman CR (1999) Nutrient control of bacterioplankton and
phytoplankton dynamics. Aquat Ecol 33:135–145
13. Brookes JD, Ganf GG, Burch MD (1998) Buoyancy regulation
of Microcystis aeruginosa. Verh Internat Verein Limnol
26:1670–1673
14. Brunberg AK (1999) Contribution of bacteria in the mucilage
of Microcystis spp. (Cyanobacteria) to benthic and pelagic bac-
terial production in a hypereutrophic lake. FEMS Microb Ecol
29:13–22
15. Carmichael WW (1986) Algal toxins. Advances in Botanical
Research, vol 12. Academic Press, San Diego, pp 47–74
16. Carmichael WW (1996) Toxic Microcystis and the environ-
ment. In: Watanabe M (ed) Toxic Microcystis.. CRC Press,
New York, pp 1–11
17. Chapuis C, Fladrois JP (1994) Mathematical model of the
interactions between Micrococcus spp and Pseudomonas aeru-
ginosa on agar surface. J Appl Bacteriol 77:727–732
18. Chrzanowski TH, Sterner RW, Elser JJ (1995) Nutrient en-
richment and nutrient regeneration stimulates bacterioplank-
ton growth. Microb Ecol 29:221–230
19. Cole JJ (1982) Interactions between bacteria and algae in
aquatic ecosystems. Ann Rev Ecol Syst 13:291–314
20. Cole JJ, Findlay S, Pace M (1988) Bacterial production in fresh
and saltwater ecosystems: a cross-system overview. Mar Ecol
Prog Ser 43:1–10
21. Currie DJ, Kalff J (1984) A comparison of the abilities of
freshwater algae and bacteria to acquire and retain phospho-
rus. Limnol Oceanogr 29:298–310
22. Egan B (1987) Marine microbial adhesion and its conse-
quences. In: Sleigh M (ed) Microbes in the Sea. Ellis Horwood
Pub, Chichester, pp 220–238
23. Flores E, Wolk CP (1986) Production by filamentous, nitro-
gen-fixing cyanobacteria, of a bacteriocin and other antibiotics
that kill related strains. Arch Microbiol 145:215–219
24. Fogg GE (1965) Algal cultures and phytoplankton ecology.
University of Wisconsin, Press, Wisconsin, pp 1–175
25. Gimmler H, Carandang JSR (1998) Do vitamins play a role in
commensalistic relationships between microorganisms coex-
isting in extremely acid habitats? Acta Manilana 46:9–21
26. Gray KM (1997) Intracellular communication and group be-
havior in bacteria. Trends Microbiol 5:184–188 DOI: 10.1016/
S0966-842X(97)01002-0
27. Gupta AB, Shrivastava GC (1965) On antibiotic properties of
some freshwater algae. Hydrobiologia 25:285–288
28. Helleburst JA (1974) Extracellular products. In: Stewart W
(ed) Algal Physiology and Biochemistry. Blackwell Scientific
Pub, Oxford, pp 838–859
29. Herndl GJ (1988) Ecology of amorphous aggregations (marine
snow) in the North Adriatic Sea. II. Microbial density and
activity in marine snow and its implication to overall pelagic
processes. Mar Ecol Prog Ser 48:265–275
30. Jackson GA (1987) Simulation of bacterial attraction and ad-
hesion to falling particles in an aquatic environment. Limnol
Oceanogr 34:514–530
31. Keating KI (1977) Allelopathic influences on blue-green se-
quence in a eutrophic lake. Science 196:885–887
32. Keating KI (1978) Blue-green algal inhibition of diatom
growth: Transition from mesotrophic to eutrophic commu-
nity structure. Science 199:971–973
33. Lam CW, Silvester WB (1979) Growth interactions among
blue-green (Anabaena oscillariodes, Microcystis aeruginosa) and
green (Chlorella sp) algae. Hydrobiologia 63:135–143
34. Lancelot C, Billen G (1985) Carbon–nitrogen relationships in
nutrient metabolism of coastal marine ecosystems. Adv Aquat
Microbiol 3:263–321
35. Lurling M (1999) Grazer-induced coenobial formation in
clonal cultures of Scenedesmus obliquus (Chlorococcales, Chlo-
rophyceae). J Phycol 35:19–23
36. Lustigman B, Brown C (1991) Antibiotic production by ma-
rine algae isolated from the New York/New Jersey Coast. Bull
Environ Contam Toxicol 46:329–335
37. Lustigman B, Lee L, Thees N, Masucci J (1992) Production of
antibacterial substances by macroalgae of the New York/New
Jersey Coast, USA. Bull Environ Contam Toxicol 49:743–749
38. Mandimba G, Heulin T, Bally R, Guckert A, Balandreau J
(1986) Chemotaxis of free-living nitrogen-fixing bacteria to-
wards maize mucilage. Plant Soil 90:129–139
39. Murray RE, Cooksey KE, Priscu JC (1986) Stimulation of
bacterial DNA synthesis by algal exudates in attached algal–
bacterial consortia. Appl Environ Microbiol 52:77–82
40. Paerl HW (1988) Nuisance phytoplankton blooms in coastal,
estuarine, and inland waters. Limnol Oceanogr 33:823–847
41. Reynolds CS (1973) Growth and buoyancy of Microcystis ae-
ruginosa Ku¨tz emed. Elenkin in a shallow eutrophic lake. Proc
R Soc Lond 184:29–50
42. Riley MA, Gordon DM (1999) The ecological role of bacte-
riocins in bacterial competition. Trends Microbiol 7:129–133
DOI: 10.1016/S0966-842X(99)01459-6
43. Rodriguez AV, Manca De Nadra MC (1995) Effect of pH and
hydrogen peroxide produced by Lactobacillus hilgardii on Pe-
diococcus pentosaceus growth. FEMS Microbiol Lett 128:59–62
DOI: 10.1016/0378-1097(95)00083-H
44. Shirai M, Matumara K, Ohtake A, Takamura Y, Aida T, Na-
kano M (1989) Development of a solid medium for growth
and isolation of axenic Microcystis strains (Cyanobacteria).
Appl Environ Microbiol 55:2569–2571
45. Stanier RY, Kunisawa R, Mandel M, Cohen-Bazire G (1971)
Purification and properties of unicellular blue-green algae (or-
der Chroococcales). Bacteriol Rev 35:171–181
46. Vaque D, Duarte CM, Marrase C (1989) Phytoplankton colo-
nization by bacteria: encounter probability as a limiting factor.
Mar Ecol Prog Ser 54:137–140
47. Vance BD (1965) Composition and succession of cyanophy-
cean water blooms. J Phycol 1:81–86
48. Vincent WF, Silvester WB (1979) Growth of blue-green algae
72 D.A. Casamatta, C.F. Wickstrom
in the Manukae (New Zealand) oxidation ponds. II. Experi-
mental studies on algal interaction. Water Res 13:717–723
49. Watanabe M (1996) Isolation, cultivation, and classification of
bloom-forming Microcystis in Japan. In: Watanabe M (ed)
Toxic Microcystis.. CRC Press, New York, pp 13–34
50. Watanabe M, Oishi S (1985) Effects of environmental factors
on the toxicity of a Cyanobacterium (Microcystis aeruginosa)
under culture conditions. Appl Environ Microbiol 49:1342–
1344
51. Watt WD (1965) Release of dissolved organic material from
the cells of phytoplankton populations. Proc Roy Soc B
164:521–551
52. Wehr JD, Petersen J, Findlay S (1999) Influence of three con-
trasting carbon sources on planktonic bacterial metabolism in
a mesotrophic lake. Microb Ecol 37:23–35
53. Worm J, Søndergaard M (1998) Dynamics of heterotrophic
bacteria attached to Microcystis spp (Cyanobacteria). Aquat
Microb Ecol 14:19–28
54. Zar JH (1984) Biostatistical Analysis, 2nd ed. Prentice-Hall,
Englewood Cliffs, New Jersey
Bacterial Sensitivity to M. aeruginosa 73
