Comparative Biochemistry and Physiology Part A 120 (1998) 99 – 105
Review
Evaluation of the swimming activity of Daphnia magna by image
analysis after administration of sublethal Cadmium concentrations
Gretel Wolf
a,
*, Paul Scheunders
b
, Marcel Selens
c
a
Department of Biology, Uni
6ersity of Antwerp, RUCA, Groenenborgerlaan,
171
,
2020
Antwerpen, Belgium
b
Vision Laboratory, Uni
6ersity of Antwerp, RUCA, Groenenborgerlaan,
171
,
2020
Antwerpen, Belgium
c
Department of Physics, Uni
6ersity of Antwerp, RUCA, Groenenborgerlaan
171
,
2020
Antwerpen, Belgium
Received 24 March 1997; received in revised form 6 September 1997; accepted 10 September 1997
Abstract
The swimming activity of Daphnia magna was determined by a real time image analysis, using a video camera and a PC
equipped with a standard low cost frame grabber. For a sequence of 20 images, where 30 organisms are moving simultaneously,
the trajectories have been reconstructed in a binary image. The average velocity was derived from the statistical analysis of these
trajectories, the participation in swimming activity and the distribution (or migration) of the organisms in the measuring cell are
derived from data analysis. The swimming activity of the organisms has been measured in normal conditions and after application
of an acute sublethal cadmium stress (concentrations of 3.5 and 5.0
mg l
− 1
). This metal stress induced an important decrease of
the swimming activity. Application of a rise of temperature (from 20 to 25°C) in combination with the sublethal cadmium stress
amplified this decrease. The alterations of the swimming velocity, participation and distribution have been quantified. The results
were available within 24 h. © 1998 Elsevier Science Inc. All rights reserved.
Keywords
: Biomonitoring; Cadmium; Daphnia; Image analysis; Sublethal stress; Swimming activity; Toxicity
1. Introduction
The swimming activity of Daphnia magna L. is
closely connected to its energy status. Submission of the
organism to changing external conditions induces a
stress and the organism will use a part of its energy in
a different way. As a response to the imposed stress it
will change its swimming activity. Changes in the envi-
ronment may reduce fitness and disturb the normal
functions so that energy destined for normal metabolic
functions such as growth, reproduction and locomo-
tion, must be used to restore the imbalance. Daphnia
magna reacts to stress-situations, which can be induced
by chemical compounds and also by changes in the
natural environmental conditions (temperature, pH,
oxygen availability), by an escape (increased swimming
activity), an adaptation (use of energy for adaptional
mechanisms) or a protection reaction (decreased swim-
ming activity by loss of co-ordination).
In this study a biomonitoring system measuring the
swimming activity of 30 simultaneous swimming daph-
nia’s is evaluated in normal conditions and after appli-
cation of sublethal stress. The swimming behaviour of
Daphnia magna is related to light. Phototactic swim-
ming of this organism is used in this biomonitoring
system. It is a vector consisting of two components: (a)
velocity; and (b) direction [12]. The mechanism can be
expected to consist of a kinetic mechanism regulating
speed and duration and an orientation mechanism set-
ting the direction. A photoreceptor system localised in
the compound eye of the organism takes part in the
* Corresponding author. Tel.: + 32 180347; fax: + 32 180497;
e-mail: gewe@ruca.ua.ac.be
1095-6433/98/$19.00 © 1998 Elsevier Science Inc. All rights reserved.
PII
S1095-6433(98)10016-8
G. Wolf et al.
/
Comparati
6e Biochemistry and Physiology, Part A
120 (1998) 99 – 105
100
regulation of phototaxis [1,3,13,15]. Definition will be
made of the used conditions (light intensity, wave-
length) leading to a constant response of the organisms.
The imposed stress consists of the heavy metal cad-
mium, known for its toxicity. The effect of sublethal
cadmium concentrations (3.5 and 5.0
mg l
− 1
) on the
swimming behaviour is followed. The application of a
second stress (a rise of temperature from 20 to 25°C) in
combination with the cadmium stress is also investi-
gated. An additional effect after application of both
stresses belongs to the possibilities.
2. Materials and methods
2.1.
Animals
Females of a positively phototactic laboratory clone
of Daphnia magna (Charles River, Hampton, VA) are
used. The animals are grown in 1-l jars in culture
medium containing 0.2 mM MgSO
4
7H
2
O; 0.4 mM
NaHCO
3
; 0.095 mM CaCl
2
2H
2
O; 0.04 mM KCl; 0.43
mM CaO at a final pH 8.0 at 20°C and a diurnal light
rhythm of 14 h of light and 10 h of darkness. They are
fed a mixture of Chlamydomonas reinhardtii L. and
Selenastrum capricornutum L. in concentrations of 5 ×
10
5
cells ml
− 1
and 5 × 10
4
cells ml
− 1
, respectively. The
medium is completed with finely ground fish food
(0.100 mg ml
− 1
). Experiments are conducted using 30
daphnia’s (between 7 and 11 days old) and carried out
in a dark room at a temperature of 20°C.
The experimental set-up is presented in Fig. 1. The
measuring cell containing the daphnia’s is a rectangular
flow-through cell (10 × 7.5 × 1.5 cm) in Plexiglas fur-
nished with a thermostatic regulation). The inlet of the
flow through tubing is localised in the lower corner and
the outlet in the opposite upper corner. The swimming
activity of the daphnia’s in the cell is not hampered by
a pumping rate of 7 ml min
− 1
.
The cell is illuminated by two light sources, each with
a specific colour and irradiance. Light source (A) illu-
minates the flow-trough cell on the back side. It con-
sists of the light beam of a projector (Leitz Pradovit
Color, with a long living halogen lamp). The light beam
goes through a red colour filter (567 nm, Negative
Color Key 3m
− 1
) and at one end of the black tunnel
through an opal Plexiglas screen(which gives an opti-
mal light dispersion). The top of the cell is illuminated
by light source (B) consisting of the light-beam of a
halogen lamp conducted through a glass fibre (Schott
KL 1500-electronic) and through a yellow filter (404
nm, Negative Color Key 3m
− 1
). In front of the cell a
video camera (SONY Video Hi 8-Handycam CCD-
V800E/PAL), enables the observation and registration
of the swimming activity of the organisms. The images
are digitised, processed and analysed by means of im-
age processing techniques. The hardware consists of a
frame grabber (DT-2862) plugged into a PC-386. The
frame grabber consists of an A/D-converter that digi-
tises images in real time (25 frames s
− 1
), a frame
memory for four images (512 × 512 × 8 bits), lookup
tables and a small logical unit to perform binary logical
operations on images in real time.
Fig. 1. Experimental set-up: A and B light sources, 1, opal Plexiglas screen; 2, black tunnel; 3, flow-through measuring cell with daphnia’s,
(flow-through circulation and thermostatic regulation is not shown); 4, video camera; 5, frame grabber; 6, PC; 7, monitor showing swimming
activity; 8, monitor showing calculated data; and 9, monitor showing curves (velocity, participation and distribution).
G. Wolf et al.
/
Comparati
6e Biochemistry and Physiology, Part A
120 (1998) 99 – 105
101
Fig. 2. (a) real image of the daphnia’s; (b) example of a processed binary image revealing all the trajectories of daphnia’s during a sequence of
1 s.
3. Results
3.1.
Image processing
In order to analyse the swimming activity properly
the organisms should be observed as single objects by
the camera. The swimming behaviour is determined by:
(a) the velocity of the animals; (b) the participation in
swimming activity; and (c) their distribution (or migra-
tion). A homogenous distribution is obtained when
about 50% of the animals are swimming constantly in
the upper half and the other 50% in the lower half of
the flow-through cell.
The calculation of the velocity, participation and
distribution requires the following image processing
and analysis: the images are digitised in real time by the
frame grabber. The images on a monitor are built up by
very small points, the so called pixels. The images are
thresholded in real time, i.e. all grey pixels of the
images below a certain value, chosen by the user, are
equalised to zero (and transformed too white), while all
remaining grey pixels are equalised to one (and trans-
formed to black). The threshold value is chosen in such
a way that the (black) organisms show a definite con-
trast with the (white) background. This results in a
binary image where all object pixels have a value of one
and the background a value of zero. Then a binary
logical operation is performed in real time, which ‘sums
up’ two binary images in the sense that if a pixel is an
object pixel in one or in both images, the result will also
be an object pixel. An algorithm is developed which
‘sums up’ a sequence of images. Intermediate results are
temporarily stored in the frame memory. The result is a
binary reconstruction of individual trajectories of the
objects during the sequence. In Fig. 2 an example of
such a processed binary image is shown, revealing all
the trajectories of objects during a sequence of 1 s.
The participation and distribution of the organisms
can be quantified by counting them using standard
image analysis techniques [14]. As these operations need
Fig. 3. Image analysis in standard conditions at 20°C and without metal addition provides three indicators enabling the quantification and
evaluation of the swimming activity: (1) the swimming velocity; (2) the swimming participation; and (3) the distribution of the organisms in the
measuring cell. The swimming velocity can immediately be derived from curve A. Velocity is expressed in RvU (relative velocity units;
6 is
expressed as the average distance, in pixels, an organism travels during one frame time). From curve B, showing the total number of swimming
organisms the swimming participation can be deduced and from curves C and D, showing the respective number of animals swimming in the lower
and upper half of the measuring cell, the distribution of the animals can be derived. In each curve the average values within the first 2 h of the
experiment are considered as the 100% normal starting conditions (control). In a control experiment a slight decrease of the velocity and small
changes in the participation and distribution values appear after 24 h.
G. Wolf et al.
/
Comparati
6e Biochemistry and Physiology, Part A
120 (1998) 99 – 105
102
not be performed in real time they are done on the first
image of each sequence. The total area of trajectories
during a sequence is a measure for the velocity of an
organism. However, a geometrical model is needed to
correct the influence of external parameters such as the
irradiance,
the
camera
distance
and
the
chosen
threshold value. The average velocity of Daphnia magna
can be approximated by the equation:
6=
A
A
%
− 1
pa
4n
6 is the velocity; A is the total number of object pixels
in the trajectory image; A
% is the total number of object
pixels in one image of the sequence; a is the average
diameter of the organisms; n is the number of images in
the sequence).
The result
6 can be interpreted as the average dis-
tance (in pixels) an organism travels during one frame
time. If necessary the pixel size can be calibrated yield-
ing an absolute value for
6. Averaging 6 over 500
sequences leads to results with statistical variances of
5% or lower.
3.2.
Biomonitoring of the swimming acti
6ity
In order to carry out properly image analysis of the
swimming activity, conditions have to be created to
achieve a homogeneous distribution of the organisms
throughout the whole flow-through cell. The dimen-
sions of the cell are chosen in such a way that the
camera focuses the entire frontal surface exception
made for the cell walls. The registration of artefacts
that can be generated by loss of swimming velocity
after the organisms hit the cell walls should be avoided.
As the organisms should always be sharply discerned as
single objects by the camera, the depth of the cell
should not exceed 1.5 cm.
The choice of the light intensities (3 × 10
− 2
W m
− 2
for light source A and 2 × 10
− 2
W m
− 2
for light source
B) is closely connected with the sensitivity of the pho-
topigments in the compound eye of the daphnia’s situ-
ated in a range between 3 × 10
− 2
and 5 × 10
1
W m
− 2
[4]. Moreover the applied light intensity enables the
creation of sufficient contrast between the organisms
and their background to carry out the camera observa-
tion and subsequent thresholding operation. To achieve
a constant and homogeneous distribution of the ani-
mals throughout the whole flow-through cell, colour
preference of the clone and its response to different
colours of light is used Red light (567 nm) for light
source A and yellow light (404 nm) for light source B
induces a delicate balance of repellence for red and
attraction to yellow light, which mostly results in a
homogeneous distribution of the organisms. It is also
observed that the daphnia’s need an adaptation period
of about 2 h before showing a constant swimming
behaviour. Thereafter biomonitoring of the swimming
activity can be started. The experiments are followed
during about 24 h.
In Fig. 3 the results of image analysis are shown in
normal conditions at 20°C and without metal addition.
The swimming activity remains fairly constant during
this time. Image analysis provides three valuable indica-
tors enabling the quantification and evaluation of the
swimming activity. These indicators are: (a) 1° the
swimming velocity; (b) 2° the swimming participation;
and (c) 3° the distribution of the organisms in the
measuring cell. In normal conditions the velocity, par-
ticipation and distribution show a very slight decrease
(less than 5%) after 24 h.
3.3.
Effects of sublethal stress on swimming acti
6ity
In Fig. 4 results of image analysis of a type casting
experiment with cadmium (5.0
mg l
− 1
and 20°C) is
shown. The experiment starts with normal medium; 2 h
Fig. 4. Image analysis of an experiment with a cadmium concentration of 5.0
mg l
− 1
at 20°C. The experiment starts in standard conditions; 2 h
later the medium containing cadmium is flown through the cell (arrow) and the experiment is continued for at least 22 h. The level of the curves
A, B, C and D within the 2 first h is considered as the 100% normal starting condition of the swimming activity (control). The average of the
values at the end of the experiment can be calculated in function of the initial 100% value. The velocity (curve A) as well as the total number of
swimming animals (curve B) and the number of animals in the lower (curve C) and the upper part of the measuring cell (curve D) decrease after
24 h.
G. Wolf et al.
/
Comparati
6e Biochemistry and Physiology, Part A
120 (1998) 99 – 105
103
Fig. 5. Calculated indicator values of the swimming activity after application of a cadmium stress at 20°C. (a) The effect of sublethal cadmium
concentrations (3.5 and 5.0
mg l
− 1
) on the swimming velocity at the standard temperature of 20°C is shown; The velocity decreases for increasing
metal concentration. (b) The participation and the distribution of the organisms in the measuring cell at the start of the experiment is shown. (c)
After 24 h an important decrease of the participation is observed. The distribution shows a decrease for the animals in the upper part of the cell
for increasing cadmium concentrations.
after the start medium containing cadmium is flown
through the cell (arrow) and the experiment is continued
for at least 20 – 22 h. The average level of each curve
within the first 2 h is considered as the 100% normal
starting condition of the swimming activity. The effect
of cadmium results after 20 h in lower values and the
average level of these values is calculated in function of
the initial 100% starting conditions. Calculations are
made for all the experiments and in Fig. 5 the effect of
cadmium stress (3.5 and 5.0
mg l
− 1
at 20°C is presented.
The swimming velocity and the participation show im-
portant decreases for increasing metal concentration.
The distribution shows a marked decrease of animals
swimming in the upper part of the cell.
In Fig. 6 the effect of two stresses are shown (cad-
mium concentrations of 3.5 and 5.0
mg l
− 1
and rise of
temperature from 20 to 25°C). The application of only
a rise of temperature induces an important increase of
the velocity. However the rise of temperature together
with cadmium stress results in an important decrease of
the participation and the distribution shows a decrease
for the number of animals in the upper part of the cell.
For Figs. 5 and 6 calculations were made for the
standard deviation (S.D.) of the velocity showed always
to be less than 5% and of the participation and distribu-
tion it was less than 4%.
In order to assess the interest of the distribution, the
decrease of the number of the organisms swimming in
the upper part of the cell is considered. Leaving this part
of the cell means that the organisms loose fitness. In Fig.
7 this decrease of fitness is shown (expressed by the ratio
of the percentage of animals in the upper part of the cell
still swimming after 24 h per percentage of animals
swimming at the start of the experiment). The metal
stress and the combination of temperature and metal
stress results in a decrease of this ratio.
4. Discussion
Many studies have discussed the effects of exposure
of Daphnia magna to stress induced by chemical and/or
G. Wolf et al.
/
Comparati
6e Biochemistry and Physiology, Part A
120 (1998) 99 – 105
104
Fig. 6. Calculated indicators of the swimming activity after submission to a cadmium stress and a temperature stress (from 20 to 25°C). (a) An
increase of the velocity is observed for the temperature stress and no cadmium; increase of temperature and addition of cadmium (3.5 and 5.0
mg
l
− 1
) results in a decrease of velocity. (b) The partition and distribution of the organisms at the start of the experiment is shown. (c) After 24 h
an important decrease of participation and distribution of the upper part of the cell is observed in all the cases.
physical stresses. The respective criteria (mortality,
morbidity,
growth,
reproduction,
physiology,
be-
haviour, biochemistry, respiration, cytogenetics, cytol-
ogy,
embryology,
carcinogenesis,
etc.)
and
the
usefulness of acute, sub-chronic, chronic and reproduc-
tion tests have been extensively discussed [4,6,7,9,11].
To improve the comparability and sensitivity of the
tests, Knie [8] developed an automatic biomonitor to
carry out dynamic Daphnia tests. Coulon et al. [2]
developed an automatic tracking system to evaluate the
swimming movements of single rotifers, using a camera,
PC and software. More recently movement analysis of
Euglena gracilis L. was carried out by real time image
analysis [5,10].
In this study it is shown that the developed biomoni-
toring system, using real time image analysis, is very
suitable to evaluate the swimming activity of Daphnia
magna after submission to sublethal stress. The images
contain a large amount of data that generate after
application of the image processing program three valu-
able indicators enabling the quantification and evalua-
tion of the swimming activity. These indicators are in
declining order of importance: (1) the swimming veloc-
ity; (2) the swimming participation; and (3) the distribu-
tion of the animals in the measuring cell. Submission of
the organisms to acute sublethal stress results in impor-
tant alterations of these three indicators.
The imposed cadmium stress induces an important
decrease of the velocity that seems to us the most
important indicator. This decrease can be explained by
the loss of fitness, due to the effect of the metal on
metabolism inducing a loss in energy production. This
leaves the organism with less energy for muscle activity
or locomotion and overcoming the friction of the
aquatic medium during swimming activity. Obviously
the decrease in velocity leads to less swimming partici-
pation and a decrease in the number of animals swim-
ming in the upper part of the cell. For all the imposed
G. Wolf et al.
/
Comparati
6e Biochemistry and Physiology, Part A
120 (1998) 99 – 105
105
Fig. 7. Decrease in fitness is expressed by the ratio of the percent of
animals in the upper part of the cell still swimming after 24 h per
percent of animals swimming a the start of the experiment in the
upper part of the cell.
velocity, participation and distribution and are avail-
able within 24 h.
References
[1] Buchanan C, Goldberg B. The action spectra of Daphnia magna
(Crustacea) phototaxis in a simulated natural environment.
Photochem Photobiol 1981;34:711 – 7.
[2] Coulon PY, et al. An experimental system for the automatic
tracking
of
rotifer
swimming
behaviour.
Hydrobiologia
1983;104:197 – 202.
[3] De Meester L. Colour in Daphnia photobehaviour. The sensi-
tivity of Daphnia magna for lateral yellow and green stray light.
Biologisch Jaarboek Dodonaea 1992;60:54 – 63.
[4] Dwyer FJ, et al. Toxicity of trace elements and salinity mixtures
to Stripped Bass (Morone saxatilis) and Daphnia magna. Envi-
ron Toxicol Chem 1992;11:513 – 20.
[5] Gualtieri P, Cotelli P. A digital microscope for real time
detection of moving micro-organisms. Micron Microscopica
Acta 1989;20(2):99 – 105.
[6] Hicks WW, Parkhurst BR. Performance characteristics of
effluent toxicity tests: variability and its implications for regula-
tory policy. Environ Toxicol Chem 1992;11:793 – 804.
[7] Klkuchi S. Giant mitochondria with crystalline matrices in the
ion transporting cells of the gill epithelium of Daphnia Magna
acclimated to hypertonic environments. J. Submicroscopical
Cytol 1984;16:503 – 14.
[8] Knie J. Der dynamischer Daphnie¨ntest, ein automatischer
Biomonitor zur U
8 berwachung von Gewa¨ssern und Abwa¨ssern.
Z Wasser Boden 1978;12:310 – 2.
[9] Parkhurst BR, Warren-Hicko W, Noel L. Performance charac-
teristics of effluent toxicity tests summarisation and evaluation
of data. Environ Toxicol Chem 1992;11:771 – 91.
[10] Passarelli V, Barsanti L, Gualtieri P. A simple device for
ultramicroscopic studies of motile micro-organisms. Micron
Microscopica Acta 1990;21(1/2):101 – 4.
[11] Report EPS 1/RM/11. Biological test method: acute lethality
test method using Daphnia spp., 1992.
[12] Ringelberg J. The positively phototactic reaction of Daphnia
magna Strauss, a contribution to the understanding of diurnal
vertical migration. Neth J Sea Res 1964;2:319 – 408.
[13] Ringelberg J. Light induced behaviour in Daphnia. In: Peters R,
De Bernardi R, editors, Daphnia. Memorie dell’Instituto ital-
iano di Idrobiologia 1987;45:285 – 1987.
[14] Russ JC. Computer Assisted Microscopy. New York: Plenum,
1992.
[15] Young S. Behavioural correlates of photoreception in Daphnia.
In: Lavernack MS, Cosens DJ, editors. Sense Organs. Glasgow:
Blackie, 1981:49 – 63.
metal stresses the participation show a decrease. Con-
cerning the distribution a decrease in swimming fitness
is obvious for the animals swimming in the upper part
of the cell for all the imposed stresses.
A temperature stress only results in an increased
velocity, which can be explained thermodynamically.
The accompanying decrease in participation and distri-
bution can be due to a (temporary) exhaustion of the
organism as a consequence of the high swimming rate.
Compared to the results of LC
50
tests the image
analysing technique is highly superior as it is very fast
and gives more information concerning the fitness of
the organism. In LC
50
tests, carried out in the labora-
tory, it was shown that no LC
50
was reached at 20°C
for both cadmium concentrations (3.5 and 5.0
mg l
− 1
)
after 20 days. At 25°C a LC
50
was reached for 5.0
mg
l
− 1
cadmium after 10 days. Image analysis enables the
quantification of the fitness of the organisms after
exposure to sublethal stress. The results of real time
image analysis give quantified information concerning
Wyszukiwarka
Podobne podstrony:
więcej podobnych podstron