ArticlePDF Available

Factors Controlling Extremely Productive Heterotrophic Bacterial Communities in Shallow Soda Pools

Authors:

Abstract and Figures

Dilute soda lakes are among the world's most productive environments and are usually dominated by dense blooms of cyanobacteria. Up to now, there has been little information available on heterotrophic bacterial abundance, production, and their controlling factors in these ecosystems. In the present study the main environmental factors responsible for the control of the heterotrophic bacterial community in five shallow soda pools in Eastern Austria were investigated during an annual cycle. Extremely high cyanobacterial numbers and heterotrophic bacterial numbers up to 307 x 10(9) L(-1) and 268 x 10(9) L(-1) were found, respectively. Bacterial secondary production rates up to 738 micro g C L(-1) h(-1) and specific growth rates up to 1.65 h(-1) were recorded in summer and represent the highest reported values for natural aquatic ecosystems. The combination of dense phytoplankton blooms, high temperature, high turbidity, and nutrient concentration due to evaporation is supposed to enable the development of such extremely productive microbial populations. By principal component analysis containing the data set of all five investigated pools, two factors were extracted which explained 62.5% of the total variation of the systems. The first factor could be interpreted as a turbidity factor; the second was assigned to as concentration factor. From this it was deduced that bacterial and cyanobacterial abundance were mainly controlled by wind-induced sediment resuspension and turbidity stabilized by the high pH and salinity and less by evaporative concentration of salinity and dissolved organic carbon. Bacterial production was clustered with temperature in factor 3, showing that bacterial growth was mainly controlled by temperature. The concept of describing the turbid water columns of the shallow soda pools as "fluid sediment" is discussed.
Content may be subject to copyright.
Factors Controlling Extremely Productive Heterotrophic Bacterial
Communities in Shallow Soda Pools
A. Eiler,
1
A.H. Farnleitner,
2
T.C. Zechmeister,
3
A. Herzig,
4
C. Hurban,
5
W. Wesner,
5
R. Krachler,
5
B. Velimirov,
1
A.K.T. Kirschner
1
1
Institute of Medical Biology, Vienna University, Waehringerstr. 10, A-1090 Vienna, Austria
2
Institute of Chemical Engineering, Technical University, A-1060 Vienna, Austria
3
Institute of Bacteriology, Mycology, and Hygeine, University of Veterinary Medicine, A-1210 Vienna,
Austria
4
Biological Research Institute Burgenland, A-7142 Illmitz, Austria
5
Institute of Inorganic Chemistry, Vienna University, A-1090 Vienna, Austria
Received: 20 June 2002; Accepted: 23 December 2002; Online publication: 13 May 2003
A BSTRACT
Dilute soda lakes are among the world’s most productive environments and are usually domi-
nated by dense blooms of cyanobacteria. Up to now, there has been little information available
on heterotrophic bacterial abundance, production, and their controlling factors in these eco-
systems. In the present study the main environmental factors responsible for the control of the
heterotrophic bacterial community in five shallow soda pools in Eastern Austria were investi-
gated during an annual cycle. Extremely high cyanobacterial numbers and heterotrophic bac-
terial numbers up to 307 · 10
9
L
)1
and 268 · 10
9
L
)1
were found, respectively. Bacterial
secondary production rates up to 738 lgCL
)1
h
)1
and specific growth rates up to 1.65 h
)1
were
recorded in summer and represent the highest reported values for natural aquatic ecosystems.
The combination of dense phytoplankton blooms, high temperature, high turbidity, and nutrient
concentration due to evaporation is supposed to enable the development of such extremely
productive microbial populations. By principal component analysis containing the data set of all
five investigated pools, two factors were extracted which explained 62.5% of the total variation of
the systems. The first factor could be interpreted as a turbidity factor; the second was assigned to
as concentration factor. From this it was deduced that bacterial and cyanobacterial abundance
were mainly controlled by wind-induced sediment resuspension and turbidity stabilized by the
high pH and salinity and less by evaporative concentration of salinity and dissolved organic
carbon. Bacterial production was clustered with temperature in factor 3, showing that bacterial
growth was mainly controlled by temperature. The concept of describing the turbid water col-
umns of the shallow soda pools as ‘‘fluid sediment’’ is discussed.
Correspondence to: A.K.T. Kirschner; E-mail: alexander.kirschner@univie.ac.at
Microb Ecol (2003) 46:43–54
DOI: 10.1007/s00248-002-2041-9
2003 Springer-Verlag New York Inc.
Introduction
Dilute soda lakes represent the most alkaline naturally
occurring aquatic ecosystems with pH values of up to 12
[13]. These ecosystems are characterized by the presence
of large amounts of NaHCO
3
and Na
2
CO
3
and in many
cases by low concentrations of Mg
2+
and Ca
2+
because of
the insolubility of those cations as carbonate minerals
under alkaline conditions. Conditions suitable for the
formation of soda lakes are found in arid and semiarid
climate zones [11, 40], where intense evaporative con-
centration rates exceed inow rates such that salts accu-
mulate [13]. Soda lakes are regarded as being among the
worlds most productive environments, as the access to
dissolved inorganic carbon in form of HCO
3
)
and CO
3
2)
for
primary producers is unlimited [6]. In most cases, alka-
liphilic cyanobacteria are the dominating primary pro-
ducers. Aerobic heterotrophic bacteria have been reported
to reach extremely high numbers of 10
7
to 10
8
cells mL
)1
[12, 16], and bacterial secondary production can be as-
sumed to exceed values measured thus far for natural
aquatic ecosystems. Seasonal variation in bacterioplank-
ton biomass and production of alkaline aquatic ecosys-
tems are poorly examined, as well as their controlling
environmental factors. Plate count numbers of organo-
trophic bacteria were shown to remain more or less con-
stant over the year in two deep African soda lakes, despite
marked seasonal changes in salinity due to periods of
heavy rain [14]. But to our knowledge, there is no infor-
mation available on the seasonal changes of total bacte-
rioplankton in shallow soda lakes.
The soda pools investigated here are located in Eastern
Austria, within the area of the national park Neusiedler
See-Seewinkel (4745¢49¢,1647¢53¢), above the largest
mineral water deposit in Europe [24]. The mineral solutes
ascending with the groundwater ux [24] formed these
shallow soda pools (max. depth: 50 cm) with pH values
ranging from 8.5 to 11. Na
+
is the dominating cation, and
HCO
3
)
,CO
3
2)
,Cl
)
, and SO
4
2)
represent the major anions.
The salinity of the pools can vary strongly over the sea-
sons, but also within years, ranging from 0.2% to >4.0%
(w/v) before complete evaporation, as they are often ex-
posed to periods of severe aridity, typical for the Panno-
nian climate in eastern Austria. Loss of salts can then
occur by wind deation. Chemophysical parameters such
as salinity, alkalinity, and temperature may have a strong
impact on the seasonal changes of bacterial activity in
these pools. High solar radiation and regional winds re-
suspending a substantial portion of the sediment are
considered additional important external factors.
The aim of the study was to elucidate the special en-
vironmental features characterizing these environments
and enabling the development of such abundant and
productive bacterial populations in the ve pools over a
seasonal cycle. Therefore, a variety of ecological parame-
ters were measured along with heterotrophic bacterial
numbers and secondary production, and the main factors
controlling the bacterial community in the pools were
extracted by multivariate comparisons.
Materials and Methods
Study Site and Sampling
Samples were collected from ve soda pools in biweekly in-
tervals during summer and once every 6 weeks during the
winter season from May 2000 to May 2001 (Fig. 1). According
to the trophic classication system proposed by Forsberg and
Ryding [9], the investigated soda pools are extremely hyper-
trophic ecosystems (Table 1). The Oberer Stinkersee (OS) is the
pool with the highest total salt concentrations and a wind-in-
dependent permanent turbidity, whereas the Unterer Stinkersee
(US) has a varying wind dependent turbidity and a varying
concentration of humic substances leading to dark brown water
color in periods of massive decay of plant material. The Illm-
itzer Zicksee (ZL) is characterized by a low turbidity caused by
the ne sediment covered by an intense biolm preventing
resuspension of the sediment particles. The Lange Lacke (LL)
and the Wo
¨
rthenlacke (WL) exhibit a varying wind dependent
turbidity. Despite their proximity they feature large differences
in their chlorophyll a content and their amount of total sus-
pended solids (Table 1).
Chemophysical Parameters
Conductivity (WTW, LF 330), temperature, pH (Seibold Wien,
GHM) and oxygen (Seibold Wien, SHO/2) were measured in situ
three times along 20-m transects toward the center of the pools.
Along the transects three 1-L water samples were collected and
integrated for further analysis. Wind velocity was arbitrarily es-
timated according to the Beaufort wind scale (http://www.zet-
net.co.uk/sigs/weather/Met_Codes/-beaufort.htm).
For the determination of total suspended solids (TSS), a de-
ned volume of sample water was ltered through premufed
glass-ber lters (GF/C; Whatman; England) and dried to con-
stant weight. To obtain the inorganic and organic fraction, the
samples were further combusted in a mufe furnace. Total
phosphorus was determined photometrically after dissolution of
the unltered sample with potassium peroxydisulfate, using the
molybdenium-blue method according to Strickland and Parsons
[35]. After 1:10 dilution with distilled water, the ion concentra-
44 A. Eiler et al.
tions in the samples were measured with ion chromatography.
All columns and chemicals were supplied by Dionex (Sunnyvale,
CA). For anions, AS-4A columns with AG-4A precolumns and for
cations CS-12A columns with CG-12A precolumns were used. As
displacing liquid for anions a sodium carbonate buffer (6 mM
Na
2
CO
3
+ 12 mM NaHCO
3
) and for cations an 18 mM meth-
anesulfonic acid solution was used. The ions were detected with a
conductivity detector after removing the anions with the ap-
propriate suppressors (AMMS-II) and for cations (CSRS-Ultra
4mm). Alkalinity (ALK; mmol L
)1
) was calculated by the fol-
lowing equation:
ALK ¼½Na
þ
þ½K
þ
þ2½Mg
2þ
þ2½Ca
2þ
ð½Cl
þ2½SO
2
4
Þ
¼½HCO
3
þ2½CO
2
3
ð1Þ
On two occasions the carbonate and bicarbonate concentra-
tions were measured with HCl titration to calculate the dissoci-
ation constant (pK) of the [CO
3
2)
] () [HCO
3
)
] equilibrium (2):
pK ¼logð½H
þ
½CO
2
3
½HCO
3
1
Þð2Þ
The mean of the calculated pK values from each site (n = 10)
was inserted in the following equations to calculate the carbonate
and bicarbonate concentrations of the other sampling dates:
½CO
2
3
¼ALK ð2 þ 10
pHþpK
Þ
1
ð3Þ
½HCO
3
¼ALK ð1 þ 2 10
pKþpH
Þ
1
ð4Þ
Salinity (SAL; mg L
)1
) was determined by the addition of the
measured and calculated salt concentrations:
SAL ¼½Na
þ
þ½K
þ
þ½Mg
2þ
þ½Ca
2þ
þ½Cl
þ½SO
2
4
þ½HCO
3
þ½CO
2
3
ð5Þ
Dissolved Organic Carbon and Chlorophyll a
For the determination of dissolved organic carbon (DOC), sub-
samples of the integrated water samples were ltered through
precombusted Whatman GF/C lters. DOC was determined using
Table 1. Basic characterization of the ve investigated soda pools
OS US ZL LL WL
Area (km
2
) 0.49 0.19 0.7 2.24 0.29
Total phosphorus (mg L
)1
) 2.1 (0.94.3) 1.1 (0.23.9) 0.6 (0.22.4) 1.1 (0.33.5) 1.0 (0.22.6)
Chlorophyll a (lgL
)1
) 62 (4133) 88 (0402) 8 (048) 197 (21430) 52 (0135)
Salinity (g L
)1
) 7.4 (4.231.3) 4.4 (2.816.8) 5.4 (2.717.8) 3.8 (2.48.4) 5.4 (2.48.3)
pH 9.78 (9.410.05) 9.47 (9.0110.19) 9.76 (9.3610.89) 9.46 (9.109.85) 9.23 (8.889.82)
Tot. susp. solids (mg L
)1
) 1254 (3212480) 769 (152980) 81 (22467) 344 (76969) 290 (38125)
Sediment Fine sand Fine sand Clay Silt Silt
a
Values represent annual average and range (in parentheses) of 16 measurements. OS, Oberer Stinker; US, Unterer Stinker; ZL, Zicklacke; LL, Lange
Lacke; WL, Wo
¨
rthen Lacke
Fig. 1. Location of the ve investigated
saltwater pools within the national park
Neusiedler SeeSeewinkel. Sampling sites
are marked by arrows. Villages are black.
Bacterial Communities in Soda Pools 45
a Shimadzu TOC 5000 carbon analyzer (Shimadzu Corporation,
Tokyo, Japan). Chlorophyll a was extracted with 90% ethanol
(1 h at 80C) and was measured spectrophotometrically (Hitachi
U-2000) [30].
Bacterial and Cyanobacterial numbers
For the determination of bacterial and cyanobacterial numbers,
15-mL samples were preserved with formalin (4% v/v nal con-
centration) for direct epiuorescene microscopy. Each sample
consisted of three 5-mL subsamples collected along the 20-m
transects. Cells were stained with DAPI (0.01% v/v nal con-
centration) and ltered through black polycarbonate membrane
lters (Millipore, Ireland) of 0.2 lm pore size [29]. Bacterial (UV
excitation: 340380 nm, barrier lter: 430 nm) and cyanobacterial
numbers (red light excitation: 515560 nm, barrier lter: 580 nm)
were counted with a Leitz Diaplan microscope. Bacterial cell
volumes (V; lm
3
) of at least 100 cells were determined for rep-
resentative samples by eyepiece micrometer to calculate the av-
erage cellular carbon content (C; fg C cell
)1
) after Norland [27].
Cyanobacteria in representative samples were taxonomically
classied after Komarek and Anagnostidis [23].
Bacterial Secondary Production
To measure bacterioplanktonic secondary production in the ve
saltwater pools the [
3
H]leucine incorporation method, initially
developed by Kirchman et al. [18] and Simon and Azam [34], was
used and the protocol of Kirschner and Velimirov [20] was fol-
lowed. A wide range of leucine concentrations ranging from 30 to
240 nM was tested twice at each sampling site. The resulting
incorporation velocities were iteratively tted to the hyperbola
function of the MichaelisMenten enzyme kinetics by using
nonlinear regression analysis (Delta Graph 4.0, Delta Point Inc.,
USA). The plots were used for the calculation of the theoretical
maximal uptake velocity (V
max
) and of the half-saturation con-
stant (K
m
). The mean of all calculated K
m
values (n = 10) was
used to calculate V
max
for all other sampling dates from the [
3
H]
leucine uptake velocity (V) at a substrate concentration (S) of 60
nM [
3
H]leucine according to Equation (6) [8].
V
max
¼ V ðK
m
þ SÞS
1
ð6Þ
[
3
H]Leucine incorporation rates (leu
inc
) were converted to
carbon production (BSP) according to Simon and Azam [34] by
using the following equations:
BPP ¼ leu
inc
2 ð100=7:3Þ131:2 ð7Þ
where BPP = bacterial protein production; 2 = isotope dilu-
tion as recommended by the authors; 100/7.3 = 100/mol% of
leucine in protein; 131.2 = molecular weight of leucine.
BSP ¼ BPP 0:86 1:18 ð8Þ
where 0.86 and 1.18 represent the factors requested for con-
verting bacterial protein production into BSP, taking into ac-
count that 86% of the weight of the aminoacids is the carbon
moiety and that 18% of the amino acids are not detected by [34].
The specic growth rate (l;[h
)1
]) and the doubling time (g; [h])
were determined by the following equations (9, 10):
l ¼½lnðBN
0
þ BN
bsp
ÞlnðBN
0
Þ 2 ð9Þ
where BN
0
= bacterial numbers at time 0, corresponding to
the bacterial numbers determined for this sampling point via the
epiuorescence technique, and BN
bsp
-bacterial secondary pro-
duction expressed in cells L
)1
measured within the 0.5 h incu-
bation interval. The doubling times were then calculated after:
g ¼ lnð2Þl
1
ð10Þ
Microautoradiography
Because of the high concentration of cyanobacteria, microauto-
radiography was used to determine whether cyanobacteria were
capable of leucine uptake under conditions similar to the de-
termination of BSP (180 nM
3
H-leucine, 30 min, in situ tem-
perature). On two occasions during summer, 1-mL samples from
each sampling site were incubated with 30 lL[
3
H]leucine and
stopped with 60 lL TCA (5% v/v nal concentration). After
staining with DAPI (0.01% v/v nal concentration) the samples
were diluted 10·50· to obtain optimal cell densities for mi-
croscopic examination. The protocol by Fuhrman and Azam [10]
was followed, which allows simultaneous visualization of devel-
oped silver grains and DAPI-stained bacteria, avoiding interfer-
ence of the silver grains with viewing the bacteria. Bacteria and
cyanobacteria were viewed by epiuorescence microscopy, and
clusters of silver grains (5 grains) were viewed by phase contrast
microscopy.
Statistical Analysis
Data were analyzed according to Zar [39]. A nonparametric Tukey
test was applied for two independent samples and a Kruskal
Wallis test for more than two independent samples to compare the
ve saltwater pools. Principal component analysis (PCA, varimax
rotated with Kaiser normalization) was performed to evaluate the
parameters responsible for the seasonal uctuations of the inves-
tigated environmental factors in the saltwater pools. For multiple
linear stepwise regression analyses and PCA analyses, data not
meeting the requirements of homoscedaticity and normal distri-
bution (ShapiroWilks test) were log
10
-transformed after adding 1
to the variable. Further nonparametric Spearman rank correla-
tions were performed for the data of each pool. For calculation of
annual averages, monthly means were used. All statistical analyses
were performed using the software SPSS 10.0 for Windows.
Results
Physical and Chemical Parameters
The water temperature during the investigation period
varied from 1.2C to 34.4C, following the typical pattern
46 A. Eiler et al.
observed in lakes located in the temperate climate zone.
The conductivity in the ve habitats increased with the
summer evaporation to maximal values of 32 mS cm
)1
(ZL) and decreased with the autumn and winter rainfalls
and the subsequent groundwater raising (Fig. 2A). The
conductivity was highly correlated with salinity
(r
s
= 0.89, p < 0.001) and alkalinity (r
s
= 0.63, p < 0.001)
and was a good indicator for the evaporative concentra-
tion. Salinity varied from 2.4 g L
)1
to 31.3 g L
)1
(Table 1)
and was dominated by chloride, sulfate, carbonate, and
bicarbonate anions and sodium cations. The proportion of
the carbonate and bicarbonate anions comprised up to
60% of total salinity (data not shown) and was responsible
for the observed high pH values up to 10.89 in the pools.
The high alkalinity (Fig. 2B) was in turn responsible for
the relatively low calcium and magnesium concentrations
with mean values of 47 mg L
)1
and 148 mg L
)1
, respec-
tively. The peak of conductivity and alkalinity in spring
2000 at US was observed when the part of the US which
was used as sampling site for the rst four sampling dates
desiccated and thereafter an adjacent sampling site was
chosen.
DOC and Chlorophyll a
DOC values and conductivity were highly signicantly
correlated (r
s
= 0.80, p < 0.001). The highest DOC value
coincided with the highest conductivity value in ZL before
complete evaporation (Fig. 2C). Chlorophyll a values
ranged from 0 to 430 ng L
)1
with the lowest concentration
in the ZL (annual average: 8.2 lgL
)1
) and the highest
concentration in LL (annual average: 197 lgL
)1
; Table 1).
Cyanobacterial and Bacterial Numbers
Cyanobacterial numbers ranged from 0 to 307 · 10
9
cells
L
)1
with highest values observed at OS and LL in early
August 2000 and early May 2001, respectively. US exhib-
ited two peaks with values of more than 170 · 10
9
cells L
)1
in July and September (Fig. 3A). ZL and WL featured the
lowest median cyanobacterial numbers (Table 2). During
the winter months cyanobacterial numbers were low (<10
· 10
9
cells L
)1
) in all pools except for LL in February,
where numbers of 54 · 10
9
cells L
)1
were observed. Cy-
anobacteria were dominated by unicellular Synechococcus
species and lamentous forms belonging to the genus
Cyanospira. Colony-forming species were frequently ob-
served after long periods of stable weather conditions
(personal observation). However, the phytoplankton con-
sisted not only of cyanobacteria, whose numbers explained
36% of the chlorophyll a variation, but also of diatoms and
euglenoids.
In all pools two peaks of bacterial numbers occurred in
early and late summer with values up to 270 · 10
9
cells L
)1
(Fig. 3B). Low numbers (<40 · 10
9
cells L
)1
) were ob-
served after a strong decrease in July and during the whole
winter season. A subjection of the dataset to a Kruskal
Wallis test revealed that the ve soda pools were statisti-
Fig. 2. Annual cycle (8 May 2000 to 28 May 2001) of conduc-
tivity (A), alkalinity (B), and DOC (C) in the ve investigated
soda pools: Oberer Stinkersee (OS), Unterer Stinkersee (US),
Illmitzer Zicklacke (ZL), Lange Lacke (LL), and Wo
¨
rthen Lacke
(WL). Conductivity values represent the mean of three meas-
urements; standard errors are smaller than the symbols used.
Missing September values for ZL are due to complete evaporation
of the pool.
Bacterial Communities in Soda Pools 47
cally different in bacterial (v
2
= 27, p < 0.001) and cya-
nobacterial numbers (v
2
= 41, p < 0.001). Bacterial and
cyanobacterial numbers were consistently greater in the
soda pools with higher concentrations of total suspended
solids (Tables 1 and 2) as indicated by Tukey tests and by
highly signicant positive correlations of TSS with bacte-
rial numbers (r
s
= 0.76, p < 0.001) and cyanobacterial
numbers (r
s
= 0.78, p < 0.001), which were also sig-
nicantly intercorrelated (r
s
= 0.85, p < 0.001).
Bacterial Carbon Production (BSP)
From all performed saturation experiments (n = 10) a
mean half-saturation constant (K
m
) of 31.8 nM leucine
(SE: 2.5 nM) was calculated and further used to calculate
the maximal leucine uptake (V
max
) from the leucine in-
corporation rates (V)at60nM(S) for all other sampling
dates. In all experiments production rates reached satu-
ration at a concentration of 180 nM (data not shown).
Rates of bacterial production ranged over three orders of
magnitude during the investigation period, from 1 to 738
lgL
)1
h
)1
(Table 2). The annual maximal BSP rates were
measured at OS, US, and LL in the middle of July, coin-
ciding with a strong decrease in bacterial numbers and
temperature. However, highly signicant positive corre-
lations of BSP with bacterial numbers (r
s
= 0.59, p <
0.001) and temperature (r
s
= 0.58, p < 0.001) were found,
as well as with total suspended solids (r
s
= 0.47, p < 0.001)
and DOC (r
s
= 0.45, p < 0.001). At all sites high rates >30
lgCL
)1
h
)1
of bacterial secondary production were
measured from April to October, when water temperature
was mostly above 20C. With one exception (January, LL)
low rates <30 lgCL
)1
h
)1
were recorded from November
to March. Over the year, the rates of bacterial production
were not statistically different between the ve study sites,
as indicated by KruskalWallis test (v
2
= 4.5, p > 0.1).
The specic growth rates of the bacterioplankton in the
ve pools varied strongly over the seasons from 0.004 to
1.65 h
)1
(Table 2) corresponding to doubling times of 8
days and 25 min, respectively. The microautoradiography
experiments showed that the majority of silver grains in
the leucine autoradiograms were associated with bacteria
(data not shown). In all incubations with [
3
H]leucine only
a small fraction (<0.5%) of silver grains was associated
with cyanobacteria (Fig. 4), suggesting that even a high
cyanobacterial biomass hardly biased the measurements of
bacterial secondary production in the ve saltwater pools.
‘‘Labeled’’ bacteria were determined once for each habitat,
ranging from 40% (OS) to 90% (WL) of all DAPI-stained
bacteria (data not shown). These results represent mini-
mum estimates of the percentage of active cells in the ve
studied pools, because leucine was probably not utilized
by all bacteria.
Principal Component Analysis
In order to elucidate the main environmental factors de-
scribing the systems of the ve soda pools, a PCA of the
Fig. 3. Annual cycle (8 May 2000 to 28 May 2001) of cyano-
bacterial numbers (A), bacterial numbers (B), and bacterial
secondary production (C) in the ve investigated soda pools:
Oberer Stinkersee (OS), Unterer Stinkersee (US), Illmitzer Zick-
lacke (ZL), Lange Lacke (LL), and Wo
¨
rthen Lacke (WL). For (A)
and (B), values represent the mean of an integrated sample ± one
standard error of1020 microscopic elds. For (C) values rep-
resent the mean of triplicate samples ± one standard deviation.
Missing September values for ZL are due to complete evaporation
of the pool.
48 A. Eiler et al.
whole data set was performed. Four principal components
(PC) with an eigenvalue >1 were extracted (Table 3, Fig.
5). PC 1 explained 45.2% of the observed variance and
consisted of inorganic and organic suspended solids, total
phosphorus, chlorophyll a, and bacterial and cyanobac-
terial numbers. PC 2 explained 17.3% of the observed
variance and contained salinity, conductivity, and DOC.
PC 2 contained also the variable wind with a negative sign,
which can be interpreted that in periods with high eva-
porative concentration the weather situation is mostly
calm and viceversa. PC 3 (temperature and BSP) and PC 4
(alkalinity and pH) were of less importance, explaining
only 10.0 and 7.6% of the variance, respectively. PC 1 can
be described as a turbidity factor as it includes all par-
ticulate variables. However, it should be mentioned that
the variables chlorophyll a and cyanobacterial numbers
represent the manifested result of primary productivity,
which is tightly coupled to the annual light cycle and
which can be seen as an integrated component of PC 1. PC
2, on the other hand, can be described as a concentration
factor as it contains all dissolved variables.
The PCAs of the single pools showed slightly diverging
pictures than the PCA performed with the whole data set
(Table 4). PC 1 and PC 2 could be described as the tur-
bidity factor and concentration factor, together explaining
more than 56% and 73% in OS and US, respectively. At ZL,
PC 1 included both the turbidity and the concentration
factor, explaining 55% of the total variance alone, and PC
2 consisted of total phosphorus and alkalinity. A strong
biotic component (PC 2) explaining 20% of the total var-
Table 2. Microbial characterization of the ve investigated soda pools
OS US ZL LL WL
Bacterial numbers
(10
9
cells L
)1
)
72.2 (11.2268) 55.6 (4.0232) 10.1 (0.866) 41.3 (7.0195) 23.6 (2.3105)
Cyanobacterial numbers
(10
9
cells L
)1
)
70.3 (1.0307) 36.9 (0208) 0.9 (011) 37.4 (0.8288) 6.2 (046)
BSP (lgCL
)1
h
)1
)93(2738) 90 (3434) 67 (5348) 119 (3626) 61 (1376)
Specic growth rate
(l)(h
)1
)
0.060 (0.0040.60) 0.085 (0.0100.55) 0.247 (0.0301.14) 0.149 (0.0211.65) 0.110 (0.0160.34)
Doubling time (h) 54.5 (1.2182) 21.2 (1.367.2) 7.2 (0.623.5) 21.8 (0.4125) 13.1 (1.942.3)
a
Values represent annual average and range (in parentheses) of 16 BSP, Bacterial carbon production; OS, Oberer Stinker; US, Unterer Stinker; ZL,
Zicklacke; LL, Lange Lacke; WL, Wo
¨
rthen Lacke
Fig. 4. Example of a microautoradiogram of autouorescent
picocyanobacteria from OS after the addition of 180 nM
3
H-
leucine (1200· magnication). The overwhelming majority of
picocyanobacterial cells are not labeled. Arrows indicate sporadic
autouorescent cells with
3
H-leucine label incorporation. The
other silver grains in the leucine autoradiogram were associated
with bacteria (at UV excitation 340380 nm, barrier lter 430
nm).
Table 3. Principal component analysis showing the principal
components (PC) with an eigenvalue >1 and their explained
variance
PC 1 PC 2 PC 3 PC 4
Cyanobacterial numbers 0.87 ——
Bacterial secondary production ——0.82
Bacterial numbers 0.83 ——
Dissolved organic carbon 0.82 ——
Chlorophyll a 0.84 ——
Temperature ——0.90
Inorganic suspended solids 0.84 ——
Organic suspended solids 0.90 ——
Total phosphorus 0.72 ——
Alkalinity ———0.73
pH ———0.77
Conductivity 0.77 ——
Salinity 0.74 ——
Wind )0.75 ——
Explained variance (%) 45.2 17.3 10.0 7.6
a
Variables not of importance for the explanation of ve systems were
omitted
Bacterial Communities in Soda Pools 49
iance was found at LL, whereas PC 1 (33% explained
variance) is interpreted as a combination of suspended
solids and phytoplankton. The turbidity and the concen-
tration factor were of inverted importance in WL. PC 1,
explaining 52% of the observed variance, additionally in-
cluded BSP, chlorophyll a, and total phosphorus.
Discussion
Soda Pools as Sites of Extreme Bacterial Abundance and Growth
The soda pools of the national park Neusiedler
SeeSeewinkel appear to be sites of extremely high mi-
crobial activity and biomass. In past studies on hyper-
trophic aquatic environments highest bacterial numbers
up to 356 · 10
9
L
)1
were reported by Kilham [16] for
African alkaline lakes. Grant et al. [12] reported that aer-
obic heterotrophic bacteria can reach numbers of 100 ·
10
9
L
)1
in dilute soda lakes, and Zinabu and Taylor [42]
found maximal bacterial numbers of 117 · 10
9
L
)1
in a
variety of different Ethiopian soda lakes. These values are
thus of the same magnitude as our measurements (Table
5). Ecosystems similar to the Austrian soda pools exist in
Eastern Europe, Russia, and East Africa, but to the best of
our knowledge no information is available on hetero-
trophic bacterial production and growth rates. The highest
BSP rates found in the literature, up to 129 lgCL
)1
h
)1
,
were reported by Boon [5] for the Australian bill-
abongsvalues which were far exceeded by our meas-
urements. White et al. [42] reported maximal l values of
0.36 h
)1
for freshwater ecosystems and 1.28 h
)1
for salt-
water habitats, which also included data from an articial
aquaculture pond. In Table 5 ecosystems are listed which
exhibit similarities to the Austrian soda pools in salinity,
trophy, depth, turbidity, and/or alkalinity. With the ex-
ception of Big Soda Lake, all of the mentioned environ-
ments are shallow and highly productive. The study of Big
Soda Lake was chosen because it provides the only data on
bacterial production in aquatic alkaline environments. The
presented environments in Table 5 exhibit signicantly
lower bacterial production and growth rates than in this
study, except for the Indus River Delta: Bano et al. [3]
reported an unprecedented bacterial growth rate of 1.0 h
)1
for a natural assemblage, coinciding with a dense cyano-
bacterial bloom. For the East African soda lakes it is
conceivable that growth rates of magnitude similar to
those observed for the Austrian soda pools may be en-
countered, but no data are available. We suppose that the
combination of shallowness, the existence of growing algal
blooms, and high temperature is probably the driving
factor enabling growth rates of l 1.
The extremely high specic growth rates of up to 1.65
h
)1
, corresponding to doubling times of 25 min, suggest
that bacterioplankton in the pools has the potential to
grow at close to maximal rates. It was shown for labora-
tory cultures of Escherichia coli and other Enterobacteri-
aceae that the minimal time necessary for cell division
under optimal nutrient and temperature conditions is in
the range of 1520 min [22]. We are aware that uptake
rates of radiotracers do not directly measure growth.
However, we believe that this approach provides a good
estimate of bacterial growth rates, when conversion factors
are chosen carefully and methodological limitations are
taken into account. A recommended isotopic dilution of 2
[34] was applied for converting radioactive leucine uptake
into carbon production. It is generally assumed that even
at substrate concentrations leading to saturation of the
incorporated label, intracellular isotope dilution does still
occur. However, it is conceivable that in these extremely
hypertrophic ecosystems bacteria do not synthesize leu-
Fig. 5. Two-dimensional plot of the principal component
analysis (PCA) (rotation: varimax with Kaiser normalization)
performed for the whole data set of all investigated soda pools.
cond, conductivity; doc, dissolved organic carbon; sal, salinity;
alk, alkalinity; bsp, bacterial secondary production; ptot, total
phosphorus; sso, organic suspended solids; ssi, inorganic sus-
pended solids; bn, bacterial numbers; cyb, cyanobacterial num-
bers; temp, temperature; chla, chlorophyll a.
50 A. Eiler et al.
cine in their cells when external sources are available at
high concentrations. Therefore, when no isotopic dilution
may be taken into account, the maximal l would then
amount to 1.0 h
)1
(LL) and 0.65 h
)1
(ZL), respectively,
corresponding to g = 42 min and g = 64 min. This is still
only approximately two to four times slower than the
minimal doubling time of Escherichia coli. Other sources
of error are the conversion factors used in the calculations
for converting cell volumes into carbon content and leu-
cine incorporation into carbon production, i.e., mol% of
leucine in protein and the cellular protein carbon moiety.
However, when bacterial abundance is determined by the
direct-count method, the calculated growth rates are most
probably underestimated because of the presence of
nonactive cells [17]. As indicated by our microautoradi-
ography results, the percentage of active bacteria in the
pools varied from 40% to 90%, which would increase the
calculated growth rates by a factor up to 2.5, corre-
sponding to l > 1.65 h
)1
. Microautoradiograms, on the
other hand, indicated clearly that cyanobacteria could be
ruled out as contributing signicantly to the leucine up-
take, which would have led to an overestimation of
Table 4. Principal component analysis of the ve investigated soda pools showing the principal components (PC) with an eigenvalue
>1 and their explained variance
PC 1 PC 2 PC 3 PC 4
OS
Variables cyb, bn, ssi, sso, ptot doc, cond, sal, -wind chla, temp bsp, ptot
Expl. variance (%) 34.9 21.8 12.3 11.3
US
Variables cyb, ssi, sso, ptot bn, doc, cond, sal, -wind bsp, temp, pH
Expl. variance (%) 58.6 15.0 9.8
ZL
Variables cyb, bn, bsp, ssi, sso, doc, cond, sal ptot, alk chla, temp, pH, -wind
Expl. variance (%) 55.0 17.8 12.3
LL
Variables chla, ssi, sso, cond cyb, bsp, bn, doc temp, ptot, pH alk, sal
Expl. variance (%) 32.6 19.8 13.8 10.2
WL
Variables bsp, doc, chla, ptot, cond, sal, -wind cyb, bn, ssi, sso temp, pH
Expl. variance (%) 52.2 14.1 11.1
a
Variables of importance for the explanation (r > 0.65) of each system are shown. Abbreviations: cond, conductivity; doc, dissolved organic carbon; sal,
salinity; alk, alkalinity; bsp, bacterial carbon production; ptot, total phosphorus; sso, organic suspended solids; ssi, inorganic suspended solids; bn,
bacterial numbers; cyb, cyanobacterial numbers; temp, temperature; chla, chlorophyll a
Table 5. Bacterial specic growth rates, abundance, chlorophyll a, dissolved organic carbon, pH, and salinity in selected estuarine
areas, river deltas, and soda lakes
Ecosystem Description
Salinity
(g L
)1
)pH
DOC
(mg C L
)1
)
chl a
(lgL
)1
)
Bact.
numbers
(10
9
L
)1
)
BSP
(lgCL
)1
h
)1
Growth
rate (h
)1
)
Indus River delta [3] Mangrove ecosystem 32
38 n.d. n.d. 141 14237.5 0.0421
Urdaibai estuary [32] Shallow, hypertrophic,
turbid
1
35 n.d. <5 1.0126.0 <18 <29 0.0040.12
Humboldt Lake [33] Shallow, hypertrophic <6 n.d. 20
28 <5840 3.5110 0.0615.8 0.00010.0089
Australian billabongs [5] Shallow, hypertrophic,
turbid
<1 6.9
9.7 n.d. 101400 1157 <0.28>129 <0.01>0.15
Big soda lake [43] Meromictic lake 26
88 9.7 2060 <30 <11 <1.44 0.00170.0033
Ethiopian soda lakes [44] Meso-, hypertrophic <0.5
31 ~10.0 n.d. 1.4616 3.2117 n.d. n.d.
East African rift valley
[12, 13, 18
a
]
Meso-, hypertrophic 50
330 811.5 n.d. <200 10100 (356
b
) n.d n.d.
Austrian soda pools
(present study)
Shallow, hypertrophic,
turbid
232 8.811.0 14424 0430 0.8268 1738 0.0041.65
a
Ecosystems are listed which exhibit similarities to the Austrian soda pools in salinity, trophy, depth, turbidity and/or alkalinity. Abbreviations: DOC,
dissolved organic carbon; chla, chlorophyll a; BSP, bacterial carbon production
b
Data refer only to [18]
Bacterial Communities in Soda Pools 51
heterotrophic bacterial growth rates. Only a very small
percentage of the silver grains were found to be associated
with cyanobacterial cells (<0.5%) in all samples (data not
shown). This nding is in contrast to the conclusion by
Kamjunke and Ja
¨
hnichen [15] that measurement of bac-
terial production in highly eutrophic environments via the
leucine method may not be suitable due to signicant
uptake of leucine by cyanobacteria. Also, Nilsson and
Sundback [25] showed that cyanobacteria are able to take
up amino acids. However, these authors investigated
amino acid uptake by Microcystis sp. and Phormidium sp.
two genera which have not been detected so far in saline
soda lakes. Summing up, we believe that the observed
maximal l values 1 in the soda pools are realistic esti-
mates and that even higher growth rates may be encoun-
tered.
The ‘‘Fluid Sediment Concept’’
An important factor inuencing the bacterial communi-
ties in the pools is represented by the high turbidity
dependent on the character of the sediment and on the
high Na
+
values in combination with high pH values. The
ne sandy sediments of OS and US are more easily re-
suspended by the action of wind than the silty (LL, WL)
and clay (ZL) sediments (Table 1). In addition, the clay
sediments in ZL are stabilized by a thick biolm during
the major part of the year. Inorganic suspended solids in
the pools consist mainly of colloidal clay minerals; iron,
manganese, and aluminum oxides; and amorphous sili-
cates and carbonates, which possess a high density of
supercial negatively charged O
)
groups [24]. These
groups together with the Na
+
ions build an electrostatic
double layer responsible for the stability of turbidity [37].
Organic particles adsorb to the surfaces of the inorganic
suspended solids and in turn remain suspended in the
water column [7]. These observations lead us to char-
acterize the turbid water columns of the pools as ‘‘fluid
sediments.’’ Especially in OS, the pelagic heterotrophic
bacterial numbers and production rates were only 10·
and 3· lower than in the sediment (Kirschner et al., in
prep.). Normally, these parameters differ by three orders
of magnitude in shallow aquatic environments [20].
However, the observed doubling times were in the lowest
range observed for water-column bacteria (see above)
and thus signicantly lower than normally observed in
sediments [21]. Several speculations exist for the reasons
of the long doubling times of benthic bacteria. Sander
and Kalff [32] speculated that lower bacterial loss rates
lead to higher cell numbers in sediments, and thus to
longer doubling times. To
¨
rnblom and Bostro
¨
m [36]
proposed that benthic bacterial populations consist of a
high percentage of dormant or empty cells. The results of
this study clearly suggest that after resuspension, bacteria
attached to sediment particles have better access to uti-
lizable organic matter and oxygen, leading to the con-
clusion that benthic bacteria are most probably limited
by the availability of nutrients and oxygen.
Regulation of Bacterial Populations over the Seasons
The annual uctuations of the bacterioplankton abun-
dance in the soda pools were strongly inuenced by the
semiarid climate. High temperatures during summer and
low precipitation rates led to an evaporative concentration
of salinity and DOC. The highest DOC and conductivity
values were recorded in ZL in August, shortly before
complete evaporation, concomitantly with maximal bac-
terial numbers and production rates in this pool. However,
PCA results indicated that turbidity seemed to be more
important for the annual variation of bacterial commu-
nities and the description of the soda pools than the
concentration of dissolved solutes. Nevertheless, most
variables contained in both PC 1 and PC 2 were signi-
cantly positively intercorrelated, showing that evaporation
positively affected the turbidity via a volume decrease of
the water body. By this, the water body becomes shallower
and wind effects gain a stronger importance for sediment
resuspension. In addition, evaporative concentration in-
creases the Na
+
content and supports the electrostatic
adsorption of suspended solids. Temperature was obvi-
ously of minor importance in describing the variance of
the soda pools, but was clustered with BSP rates, and also a
signicantly positive correlation to BSP rates was observed
(r
s
= 0.58, p < 0.001). Against expectation, wind velocity
was not clustered with the variables contained in PC 1, and
no signicant correlation with total suspended solids was
found. This leads us to the conclusion that even light
winds can lead to signicant sediment resuspension be-
cause of the shallowness of the soda pools. The PCA re-
sults of the single pools underline the differences between
the ve investigated soda pools despite of their geo-
graphical proximity. Although the turbidity and the con-
centration factor are essential controlling forces for all
pools, they have a different importance for the description
52 A. Eiler et al.
of each ecosystem. It is obvious that the proposed model
of ‘‘fluid sediment’’ ts better to OS and US because of the
combination of relatively coarse sediments with high Na
+
concentrations and pH values, which is corroborated by
the PCA, where the turbidity factor was of most signi-
cance for the variance of both systems. Such preconditions
were only partly met in ZL because the clay sediment and
the intense biolm stabilizing the sediment prevent wind-
induced sediment resuspension. In LL and WL biotic
factors gain more importance in describing the variance of
these systems, because the ne particle size of the sedi-
ment and the lower salinity values allow a stable formation
of a ‘‘fluid sediment’’ only in periods with strong wind
impact.
Over all investigated pools, bacterial numbers, cyano-
bacterial numbers, and bacterial production rates were
strongly intercorrelated, as well as the concentration of
total suspended solids. These strong correlations empha-
size once more the key role of wind-induced sediment re-
suspension and salinity-enabled formation of a ‘‘fluid
sediment’’ in regulating the bacterial community in the
soda pools. In addition, bacterial production was also
strongly inuenced by temperature, as normally observed
in aquatic ecosystems [1, 19, 32, 38]. Although DOC was
correlated with BSP, we do not assume that organic or
inorganic nutrients were limiting for the bacterial com-
munity, because of the observed high specic growth rates.
Additionally, Shiah and Ducklow [33] pointed out that
nutrient supply is not a limiting factor for bacterial sec-
ondary production when a strong correlation between
temperature and BSP is observed.
A strong top-down control can also be supposed to
exist in the saltwater pools regulating the bacterial and
cyanobacterial numbers. Cladoceran and copepod densi-
ties of more than 2 · 10
3
L
)1
and also ciliate abundances of
>1 · 10
6
L
)1
were frequently observed (personal obser-
vation), representing efcient potential regulators of the
bacterial populations. In addition, viral lysis is a possible
important regulating factor. Especially in eutrophic ma-
rine ecosystems, mortality due to viral lysis was assumed
to account for more than 100% of the bacterial and cy-
anobacterial production [26]. Because of their extremely
high abundances, it is also conceivable that bacteria and
cyanobacteria are regulating their maximal population
density via quorum sensing [2, 4].
Summing up, the investigated soda pools in the na-
tional park Neusiedler SeeSeewinkel in Eastern Austria
harbor extremely abundant populations of heterotrophic
prokaryotes, and their productivity values are among the
highest reported in the literature. We believe that the
combination of dense phytoplankton blooms, high tem-
perature, high turbidity, and evaporative nutrient con-
centration is the prerequisite enabling the development of
such extremely productive microbial populations in a
natural aquatic environment.
Acknowledgments
Special thanks are due to Monika Bright and Christian
Rinke (Institute of Marine Biology, University of Vienna)
for providing all facilities for microautoradiographic
studies and to Mr. Rauchwarter (Biological Research In-
stitute, Illmitz) for the measurement of chlorophyll a,
DOC, and suspended solids. Additional thanks to Lars
Tranvik (University of Uppsala, Department of Limnol-
ogy), Jon Zehr (University of California, Santa Cruz, De-
partment of Ocean Sciences), and three anonymous
reviewers for valuable comments on an earlier version of
the manuscript. The study was nanced by a grant of the
national park NeusiedlerseeSeewinkel (NP-24; Dir.
Kirchberger).
References
1. Autio RM (1992) Temperature regulation of brackish water
bacterioplankton. Ergebn Limnol 37:253263
2. Bachofen R, Schenk A (1998) Quorum sensing autoinducers:
Do they play a role in natural microbial habitats? Microbiol
Res 153:6163
3. Bano N, Nisa M-U, Khan N, Saleem M, Harrison PJ, Ahmed
F, Azam F (1997) Signicance of bacteria in the ux of or-
ganic matter in the tidal creeks of the mangrove ecosystems
of the Indus River delta, Pakistan. Mar Ecol Prog Ser 157:
112
4. Bassler BL (1999) How bacteria talk to each other: regulation
of gene expression by quorum sensing. Curr Opin Micrio-
biol 2:582587
5. Boon PI (1991) Bacterial assemblages in rivers and bill-
abongs of southeastern Australia. Microb Ecol 22:2752
6. Duckworth AW, Grant WD, Jones BE, van Steenbergen R
(1996) Phylogenetic diversity of soda lake alkaliphiles. FEMS
Microbiol Ecol 19:181191
7. Evangelou VP (1998) Soil and Water Chemistry. John Wiley
and Sons, New York
8. Fischer H, Pusch M (1999) Use of the
14
C leucine incorpo-
ration technique to measure bacterial production in river
sediments and the epiphyton. Appl Environ Microbiol
65:44114418
Bacterial Communities in Soda Pools 53
9. Forsberg CG, Ryding S-O (1980) Eutrophication parameters
and trophic state indices in 30 Swedish waste-receiving
lakes. Arch Hydrobiol 89:189207
10. Fuhrman JA, Azam F (1982) Thymidine incorporation as a
measure of heterotrophic bacterioplankton production in
marine surface waters: evaluation and eld results. Mar Biol
66:109120
11. Grant WD, Tindall BJ (1986) The alkaline, saline environ-
ment. In: RA Herbert, GA Codd (eds) Microbes in Extreme
Environments. Academic Press, London, pp 2254
12. Grant WD, Jones BE, Mwatha WE (1990) Alkaliphiles:
ecology, diversity and applications. FEMS Microbiol Rev
75:255270
13. Grant WD (1992) Alkaline environments. In: J Lederberg
(ed) Encyclopedia of Microbiology, vol 1. Academic Press,
San Diego, pp 7380
14. Jones BE, Grant WD, Collins NC, Mwatha WE (1994) Al-
kaliphiles: diversity and identication. In: FG Priest, A Ra-
mos-Cormenzana, BJ Tindall (eds) Bacterial Diversity and
Systematic. Plenum, New York, pp 195229
15. Kamjunke N, Ja
¨
hnichen S (2000) Leucine incorporation by
Microcystis aeruginosa. Limnol Oceanogr 45:741743
16. Kilham P (1981) Pelagic bacteria: extreme abundances in
African saline lakes. Naturwissenschaften 68:380381
17. Kirchman D, Ducklow H, Mitchell R (1982) Estimates of
bacterial growth from changes in uptake rates and biomass.
Appl Environ Mirobiol 44:12961307
18. Kirchman D, KNees E, Hodson R (1985) Leucine incorpo-
ration and its potential as a measure of protein synthesis by
bacteria in natural aquatic systems. Appl Environ Microbiol
49:599607
19. Kirschner AKT, Velimirov B (1997) A seasonal study of
bacterial community succession in a temperate backwater
system, indicated by variation in morphotype numbers,
biomass, and secondary production. Microb Ecol 34:2738
20. Kirschner AKT, Velimirov B (1999) Modication of the
3
H-
leucine centrifugation method for determining bacterial
protein synthesis in freshwater samples. Aquat Microb Ecol
17:201206
21. Kirschner AKT, Velimirov B (1999) Benthic bacterial sec-
ondary production measured via simultaneous
3
H-thymi-
dine and
14
C-leucine incorporation, and its implication for
the carbon cycle of a shallow macrophyte-dominated back-
water system. Limnol Oceanogr 44:18711881
22. Koch AL (1980) The inefciency of ribosomes functioning in
Escherichia coli growing at moderate rates. J Gen Microbiol
116:165171
23. Komarek J, Anagnostidis K (1999) Granoprokaryota. In: H
Ettl, G Ga
¨
rtner, H Heywig, F Mollenhauer (eds) Su
¨
ßwass-
erora von Mitteleuropa. Gustav Fischer, Jena
24. Krachler Ru, Krachler Re, Milleret E, Wesner W (2000)
Limnochemische Untersuchungen zur aktuellen Situation
der Salzlacken im burgenla
¨
ndischen Seewinkel. Bur-
genla
¨
ndische Heimatbla
¨
tter 62:349
25. Nilsson C, Sundback K (1996) Amino acid uptake in natural
microphytobenthic assemblages studied by microautoradi-
ography. Hydrobiologia 332:119129
26. Noble RT, Fuhrman JA (2000) Rapid virus production and
removal as measured with uorescently labeled viruses as
tracers. Appl Environ Microbiol 66:37903797
27. Norland S (1993) The relationship between biomass and
volume of bacteria. In: PF Kemp, EB Sherr, JJ Cole (eds)
Current Methods in Aquatic Microbial Ecology. Lewis
Publishers, Boca Raton, FL, pp 303307
28. Nusch EA (1980) Comparison of different methods for
chlorophyll and phaeopigment determination. Arch
Hydrobiol Beih Ergebn Limnol 14:1436
29. Porter KG, Feig YS (1980) The use of DAPI for identifying
and counting aquatic microora. Limnol Oceanogr 25:943
948
30. Revilla M, Iriarte A, Madariaga I, Orive E (2000) Bacterial
and phytoplankton dynamics along a trophic gradient in a
shallow temperate estuary. Est Coast Shelf Sci 50:297313
31. Robarts RD, Arts MT, Evans MS, Waiser MJ (1994) The
coupling of heterotrophic bacterial and phytoplankton
production in a hypertrophic shallow prairie lake. Can J Fish
Aquat Sci 51:22192226
32. Sander BC, Kalff J (1993) Factors controlling bacterial pro-
duction in marine and freshwater sediments. Microb Ecol
26:7999
33. Shiah F-K, Ducklow HW (1995) Multiscale variability in
bacterioplankton abundance, production, and specic
growth rate in a temperate salt-marsh tidal creek. Limnol
Oceanogr 40:5566
34. Simon M, Azam F (1989) Protein content and protein syn-
thesis rates of planktonic marine bacteria. Mar Ecol Prog Ser
51:201213
35. Strickland JD, Parsons TR (1968) A practical handbook of
seawater analysis. Bull Fish Res Board Can 167:xx
36. To
¨
rnblom E, Bostro
¨
m B (1995) Benthic microbial response
to a sedimentation event at low temperature in sediments of
a eutrophic lake. Mar Freshwater Res 46:3343
37. Voigt HJ (1990) Hydrogeochemie. Springer Verlag, Berlin
38. White PA, Kalff J, Rasmussen JB, Gasol JM (1991) The effect
of temperature and algal biomass on bacterial production
and specic growth rate in freshwater and marine habitats.
Microb Ecol 21:99118
39. Zar JH (1974) Biostatistical Analysis. Prentice Hall, Engle-
wood Cliffs, NJ
40. Zavarzin GA (1993) Epicontinental soda lakes as probable
relict biotopes of terrestial biota formation. Microbiology
62:473479
41. Zehr JP, Harvey RW, Oremland RS, Cloern JE, George LH,
Lane JL (1987) Big Soda Lake (Nevada) 1. Pelagic bacterial
heterotrophy and biomass. Limnol Oceanogr 32:781793
42. Zinabu GM, Taylor WD (1997) Bacteriachlorophyll rela-
tionships in Ethiopian lakes of varying salinity: are soda
lakes different? J Plankton Res 19:647654
54 A. Eiler et al.
... The hypertrophic shallow soda lake (Oberer Stinker, OS) is characterized by high total salt concentrations and turbidity (Eiler et al., 2003;Kirschner et al., 2004). It was formed by mineral solutes ascending with the groundwater flux (Krachler et al., 2000) and is characterized by pH values ranging from 9.4 to 10 (Eiler et al., 2003). ...
... The hypertrophic shallow soda lake (Oberer Stinker, OS) is characterized by high total salt concentrations and turbidity (Eiler et al., 2003;Kirschner et al., 2004). It was formed by mineral solutes ascending with the groundwater flux (Krachler et al., 2000) and is characterized by pH values ranging from 9.4 to 10 (Eiler et al., 2003). Na + is the dominating cation, and HCO 3 − , CO 3 2− , Cl − , and SO 4 2− represent the major anions. ...
... Na + is the dominating cation, and HCO 3 − , CO 3 2− , Cl − , and SO 4 2− represent the major anions. Salinity of the soda lake varies strongly with seasons (Eiler et al., 2003). ...
Article
Full-text available
Molecular diagnostic methods are increasingly applied for food and environmental analysis. Since several steps are involved in sample processing which can affect the outcome (e.g., adhesion of DNA to the sample matrix, inefficient precipitation of DNA, pipetting errors and (partial) loss of the DNA pellet during DNA isolation), quality control is essential at all processing levels. In soil microbiology, particular attention has been paid to the inorganic component of the sample matrix affecting DNA extractability. In water quality testing, however, this aspect has mostly been neglected so far, although it is conceivable that these mechanisms have a similar impact. The present study was therefore dedicated to investigate possible matrix effects on results of water quality analysis. Field testing in an aquatic environment with pronounced chemo-physical gradients [total suspended solids (TSS), inorganic turbidity, total organic carbon (TOC), and conductivity] indicated a negative association between DNA extractability (using a standard phenol/chloroform extraction procedure) and turbidity (spearman ρ = −0.72, p < 0.001, n = 21). Further detailed laboratory experiments on sediment suspensions confirmed the hypothesis of inorganic turbidity being the main driver for reduced DNA extractability. The observed effects, as known from soil samples, were also indicated to result from competitive effects for free charges on clay minerals, leading to adsorption of DNA to these inorganic particles. A protocol modification by supplementing the extraction buffer with salmon sperm DNA, to coat charged surfaces prior to cell lysis, was then applied on environmental water samples and compared to the standard protocol. At sites characterized by high inorganic turbidity, DNA extractability was significantly improved or made possible in the first place by applying the adapted protocol. This became apparent from intestinal enterococci and microbial source tracking (MST)-marker levels measured by quantitative polymerase chain reaction (qPCR) (100 to 10,000-fold median increase in target concentrations). The present study emphasizes the need to consider inorganic turbidity as a potential loss factor in DNA extraction from water-matrices. Negligence of these effects can lead to a massive bias, by up to several orders of magnitude, in the results of molecular MST and fecal pollution diagnostics.
... To determine the influence of the elements' composition on the distribution of SRB strains, cluster analysis by the PCA method was performed. This method is widely used for the analysis of multidimensional systems in biology, physics, medicine, etc. [44,45,48,49]. ...
... In the work by Eiler (2003) [48], five diluted soda lakes in eastern Austria and factors of abundance of heterotrophic bacteria, their production, and their controlling factors were investigated. In these ecosystems, the environmental factors, which are responsible for the control of the microbial community in the shallow soda pools, were investigated during an annual cycle. ...
... In the work by Eiler (2003) [48], five diluted soda lakes in eastern Austria and factors of abundance of heterotrophic bacteria, their production, and their controlling factors were investigated. In these ecosystems, the environmental factors, which are responsible for the control of the microbial community in the shallow soda pools, were investigated during an annual cycle. ...
Article
Full-text available
The monitoring of trace metals in microbial cells is relevant for diagnosis of inflammatory bowel disease (IBD). Sulfate-reducing bacteria (SRB) represent an important factor in the IBD development. The content of trace metals in bacterial cells may reflect the functioning of the enzyme systems and the environmental impact on the occurrence of SRB. The aim of our research was to compare the content of trace elements in the cells of SRB cultures isolated from fecal samples of patients with IBD and healthy people. The contents of 11 chemical elements in the bacterial cells of SRB were analyzed by the inductively coupled plasma-mass-spectrometry (ICP-MS) method. Significant changes in the content of calcium, zinc, magnesium, potassium, and iron were observed in patients with IBD compared to healthy individuals. Through a principal component analysis (PCA), a total variability of 67.3% in the difference between the samples was explained. The main factors influencing the total variability in the bacterial cells of SRB isolated from patients suffering from IBD were the content of the micro- and trace elements, such as manganese (with power 0.87), magnesium and cobalt (0.86), calcium (0.84), molybdenum (0.81), and iron (0.78). Such changes in the elemental composition of SRB under different conditions of existence in the host may indicate adaptive responses of the microorganisms, including the inclusion of oxidative stress systems, which can lead to changes in SRB metabolism and the manifestation of parameters of IBD in humans. The use of PCA might make it possible in the future to predict the development and ratio of SRB in patients with various diseases.
... The environmental variables measured reflected multiple extreme conditions and were similar to those that had previously been reported in soda pans of the Carpathian Basin 2,3 ; i.e. the turbid Zab-szék pan had a higher average concentration of ISS (1436 mg l −1 ) than the colored Sós-ér pan (86 mg l −1 ), while DOC values where higher in the latter (75 mg l −1 vs. 490 mg l −1 , respectively), pH was alkaline (~ 9-10) and the salinity ranged between 1.5 to 18.9 g l −1 (subsaline to hyposaline, according to Hammer 17 ). Prokaryotic cell numbers were in the same magnitude as had been measured previously in the soda pans of the region studied (10 6 -10 8 cells ml −1 ) 22 , but higher than in five soda pans from the Seewinkel region (Austria) (10 6 cells ml −1 ) 23 . Interestingly, the number of cells did not increase linearly with the depth of the pans, indicating the dynamic nature of the prokaryotic community and the ecosystem. ...
... Members of the acI clade, another small cellsized planktonic Actinobacteria, are capable of utilizing N-acetylglucosamine (monomeric unit of chitin and component of the bacterial murein cell-wall) 10,46,47 , which can also contribute to the increase of their relative abundance with zooplankton numbers. In previous works on soda pans of the Seewinkel region 23,48 , the authors stated that the observed high abundance of cladocerans (2 × 10 3 specimen l −1 ) and protists must be an important regulatory factor on the bacterial population. In conclusion, intensive grazing by zooplankton could favor planktonic Actinobacteria by the elimination of their larger, fast-growing bacterial competitors and also by their ability to degrade chitin, which could present in high amounts in the water by the end of the zooplankton 'bloom' . ...
Article
Full-text available
Astatic soda pans of the Pannonian Steppe are unique environments with respect to their multiple extreme physical and chemical characteristics (high daily water temperature fluctuation, high turbidity, alkaline pH, salinity, polyhumic organic carbon concentration, hypertrophic state and special ionic composition). However, little is known about the seasonal dynamics of the bacterial communities inhabiting these lakes and the role of environmental factors that have the main impact on their structure. Therefore, two soda pans were sampled monthly between April 2013 and July 2014 to reveal changes in the planktonic community. By late spring in both years, a sudden shift in the community structure was observed, the previous algae-associated bacterial communities had collapsed, resulting the highest ratio of Actinobacteria within the bacterioplankton (89%, with the dominance of acIII-A1 lineage) ever reported in the literature. Before these peaks, an extremely high abundance (> 10,000 individuum l⁻¹) of microcrustaceans (Moina brachiata and Arctodiaptomus spinosus) was observed. OTU-based statistical approaches showed that in addition to algal blooms and water-level fluctuations, zooplankton densities had the strongest effect on the composition of bacterial communities. In these extreme environments, this implies a surprisingly strong, community-shaping top-down role of microcrustacean grazers.
... The abundance of planktonic bacteria is around 10 6 -10 7 cells/mL and 10 7 -10 8 cells/mL in the soda lakes and pans, respectively (Eiler et al. 2003;Kirschner et al. 2002;Szabó et al. 2020;Szabó-Tugyi et al. 2019;Vörös et al. 2008), and they belong mainly to the phyla Actinobacteria, Bacteroidetes and Proteobacteria (Bell et al. 2018;Bullerjahn et al. 2020;Szabó et al. 2017Szabó et al. , 2020Fig. 3). ...
... Remarkable diel changes in the planktonic microbial activity and abundance were also observed in the shallow soda pans, and external abiotic factors (diluting the effect of rainfall which decreases salinity, but also the grazing pressure and viral lysis, the wind-induced sediment resuspension, and daily fluctuations of temperature and irradiation) seem to have a regulating effect on them (Kirschner et al. 2002;Krammer et al. 2008). The generation time of bacteria in these environments can be as little as a few hours (Eiler et al. 2003;Krammer et al. 2008). Growth studies on planktonic bacterial isolates have shown that the type of dominant anion determines the adaptation of prokaryotes, since strains isolated from soda pans grew better in media containing (hydrogen) carbonate than in media containing the same amount of chloride (Bedics et al. 2019). ...
Article
Full-text available
In this review, I would like to summarize the current knowledge on the microbiology of soda lakes and pans of the Carpathian Basin. First, the characteristic physical and chemical features of these sites are described. Most of the microbiological information presented deals with prokaryotes and algae, but protists and viruses are also mentioned. Planktonic bacterial communities are dominated by members of the phyla Actinobacteria, Bacteroidetes and Proteobacteria; small-sized trebouxiophycean green algae and Synechococcus/Cyanobium picocyanobacteria are the most important components of phytoplankton. Based on the current knowledge, it seems that mainly temperature, salinity, turbidity and grazing pressure regulate community composition and the abundance of individual microbial groups, but the external nutrient load from birds also has a significant impact on the ecological processes.
... Physical and chemical parameters, total suspended solids, water level and discharge data At the samplings, the water temperature, pH, electrical conductivity and oxygen concentration values were measured on site with the handheld multi-parameter meter Multi 3430 (WTW, Xylem Analytics, Germany). Total suspended solids (TSS) and inorganic particulate matter were determined during the annual cycle according to Eiler et al. (2003). Water discharge and level data were obtained from national monitoring systems for the sampling site (https:// www. ...
Article
Full-text available
Epilithic biofilms are ubiquitous in large river environments and are crucial for biogeochemical processes, but their community structures and functions remain poorly understood. In this paper, the seasonal succession in the morphological structure and the taxonomic composition of an epilithic bacterial biofilm community at a polluted site of the Danube River were followed using electron microscopy, high-throughput 16S rRNA gene amplicon sequencing and multiplex/taxon-specific PCRs. The biofilm samples were collected from the same submerged stone and carried out bimonthly in the littoral zone of the Danube River, downstream of a large urban area. Scanning electron microscopy showed that the biofilm was composed of diatoms and a variety of bacteria with different morphologies. Based on amplicon sequencing, the bacterial communities were dominated by the phyla Pseudomonadota and Bacteroidota, while the most abundant archaea belonged to the phyla Nitrososphaerota and Nanoarchaeota. The changing environmental factors had an effect on the composition of the epilithic microbial community. Critical levels of faecal pollution in the water were associated with increased relative abundance of Sphaerotilus , a typical indicator of “sewage fungus”, but the composition and diversity of the epilithic biofilms were also influenced by several other environmental factors such as temperature, water discharge and total suspended solids (TSS). The specific PCRs showed opportunistic pathogenic bacteria (e.g. Pseudomonas spp., Legionella spp., P. aeruginosa , L. pneumophila , Stenotrophomonas maltophilia ) in some biofilm samples, but extended spectrum β -lactamase (ESBL) genes and macrolide resistance genes could not be detected.
... Total suspended solids and inorganic particulate matter were determined during the annual cycle according to Eiler et al. (2003). ...
... One specific environmental factor often involved in bacterial competition is the concentration and quality of dissolved organic matter (Eiler et al., 2003;Fierer et al., 2007). In this regard, the RCC proposes that labile allochthonous organic compounds are rapidly used at upstream sites where the stream has its maximum interface with the landscape. ...
Preprint
Full-text available
The bacterioplankton diversity in large rivers has thus far been undersampled, despite the importance of streams and rivers as components of continental landscapes. Here, we present a comprehensive dataset detailing the bacterioplankton diversity along the midstream of the Danube River and its tributaries. Using 16S rRNA-gene amplicon sequencing, our analysis revealed that bacterial richness and evenness gradually declined downriver in both the free-living and particle-associated bacterial communities. These shifts were also supported by beta diversity analysis, where the effects of tributaries were negligible in regards to the overall variation. In addition, the river was largely dominated by bacteria that are commonly observed in freshwaters. Dominated by the acI lineage, the freshwater SAR11 (LD12) and the Polynucleobacter group, typical freshwater taxa increased in proportion downriver and were accompanied by a decrease in soil and groundwater bacteria. Based on the River Continuum Concept, we explain these taxonomic patterns and the accompanying changes in alpha and beta diversity by the physical structure and chemical conditions coupled with the hydrologic cycle along the length of the river.(version 2.1 resubmitted to Environmental Microbiology 2014-Nov-17)
... As suspended inorganic particles (e.g. clay, silt; Eiler et al., 2003) interfere with the filter-feeding process (Kirk & Gilbert, 1990), their high concentration in the water is expected to have a negative effect on zooplankters (Dejen, Vijverberg, Nagelkerke, & Sibbing, 2004;Teffera et al., 2018;Zhou, Qin, & Han, 2018). There is, however, a difference among suspension feeders in their ability of coping with (often extreme) turbid conditions. ...
Article
Full-text available
1. Omnivory is widespread in food webs, with an important stabilising effect. The strength of omnivorous trophic interactions may change considerably with changes in the local environment. 2. Shallow temporary waters are often characterized by high levels of inorganic turbidity that may directly limit the food uptake of filter-feeding organisms, but there is little evidence on how it might affect omnivorous species. Anostracans are key species of temporary waters and recent evidence suggests that these organisms are omnivorous consumers of both phyto-and zooplankton. 3. Using Branchinecta orientalis as a model species, our aim was to test how turbidity affects the feeding of an omnivorous anostracan. To do this, we used short-term feeding experiments and stable isotope analyses, with animals collected from soda pans in eastern Austria. In the feeding experiments, algae and zooplankton were offered as food either separately or in combination. The prey type treatments were crossed with turbidity levels in a factorial design. 4. There was a pronounced decrease in the ingested algal biomass with increasing turbidity. Conversely, ingestion rates on zooplankton were less affected by turbidity. Stable isotope analyses from field material supported our experimental results by showing a positive relationship of the trophic position of anostracans and the trophic niche of the communities with turbidity. 5. Our results show that turbidity modulates the intraguild trophic relationship between anostracans and their prey by shifting the diet of anostracans from more herbivorous in transparent to more carnivorous in turbid waters. Thus, inorganic turbidity might also have a community shaping role in plankton communities of temporary waters through altering trophic relationships.
Article
Full-text available
Seasonal environmental variation is a leading driver of microbial planktonic community assembly and interactions. However, departures from usual seasonal trends are often reported. To understand the role of local stressors in modifying seasonal succession, we sampled fortnightly throughout three seasons five nearby shallow soda lakes exposed to identical seasonal and meteorological changes. We characterised their microeukaryotic and bacterial communities by amplicon sequencing of the 16S and 18S rRNA gene, respectively. Biological interactions were inferred by analyses of synchronous and time-shifted interaction networks, and the keystone taxa of the communities were topologically identified. The lakes showed similar succession patterns during the study period with spring being characterised by relevance of trophic interactions and certain level of community stability followed by a more dynamic and variable summer-autumn period. Adaptation to general seasonal changes happened through shared core microbiome of the pans. Stochastic events such as desiccation disrupted common network attributes and introduced shifts from the prevalent seasonal trajectory. Our results demonstrated that despite being extreme and highly variable habitats, shallow soda lakes exhibit certain similarities in seasonality of their planktonic communities, yet local stressors such as droughts instigate deviations from prevalent trends to a greater extent for microeukaryotic than for bacterial communities.
Preprint
Full-text available
Seasonal environmental variation is a leading driver of microbial planktonic community assembly and interactions. Yet, unexpected departures from general seasonal successional trends are often reported. To understand the role of local stochastic events in modifying seasonal succession, we sampled fortnightly throughout three seasons (spring, summer, and autumn) five nearby shallow soda lakes exposed to the same seasonal meteorological changes. We characterised their microeukaryotic and bacterial communities by 18S and 16S rRNA gene sequencing, respectively. Biological interactions were inferred by the analyses of synchronous and time-shifted interaction networks, and the keystone taxa were topologically identified. The pans showed similar succession patterns during the study period with spring being characterised by high relevance of trophic interactions and certain level of community stability followed by a more dynamic and variable summer-autumn period both in respect of community composition and microbial interactions. Adaptation to general seasonal changes happened through the abundant shared core microbiome of the pans. However, stochastic events such as desiccation and cyanobacterial blooms disrupted common network attributes and introduced shifts from the prevalent seasonal trajectory. These were more pronounced for microeukaryotes than for bacteria which was reflected in increased turnover and contribution of non-core microeukaryotes. Our results demonstrated that despite being extreme and highly variable habitats, shallow soda lakes exhibit certain similarities in the seasonality of their planktonic communities, yet random stochastic events such as droughts can instigate substantial deviations from prevalent trends for the microeukaryotic but not bacterial communities.
Article
Full-text available
A laboratory study was performed in order to investigate the short-term response of a benthic microbial community to a spring bloom sedimentation event at low temperature. In the laboratory, organic material collected with phytoplankton nets was added to undisturbed sediment cores receiving the following four treatments: (1) addition of organic material and aeration of the water overlying the sediment, (2) only aeration, (3) only organic material addition, and (4) no organic material addition or aeration. Changes in microbial activity and biomass in the sediments were followed during six days at an experimental temperature of 4ºC by means of measurements of heat production, bacterial production ([³H]thymidine incorporation), oxygen consumption, bacterial biomass, ATP concentration and chlorophyll a concentration. In treatments with organic material additions, a response in terms of simultaneously increasing heat production, bacterial production and oxygen consumption rates was observed. Microbial activity increased slowly during the first four days in cores with organic matter addition, and between Days 4 and 6 microbial activity approximately doubled in aerated cores and decreased in non-aerated cores. Mineralization rates calculated from heat production showed that a small proportion of the added organic matter was mineralized in the surface sediment during the experiment. The results of this study show that the benthic microbial community could respond to an input of organic material at low temperature within a few days. A well oxygenated overlying water phase was necessary for the response to proceed.
Chapter
Micro-organisms that grow in hostile or extreme environments are currently a popular subject for study. The fascination with so-called ‘extremophiles’ reflects a perceived microbial biotechnological importance and a desire to investigate possible early conditions on Earth and the origins of life. Extremes of pH represent a particularly hostile regime for microbial life, especially when combined with extremes of salinity or temperature.This review deals with some of the aspects of micro-organisms that inhabit environments in the alkaline range of the pH spectrum.