chlorophyll fluorescence as a bioindicator of effects on growth in

Transkript

Environmental Toxicology and Chemistry, Vol. 20, No. 4, pp. 890–898, 2001
q 2001 SETAC
Printed in the USA
0730-7268/01 $9.00 1 .00
CHLOROPHYLL FLUORESCENCE AS A BIOINDICATOR OF EFFECTS ON GROWTH IN
AQUATIC MACROPHYTES FROM MIXTURES OF POLYCYCLIC
AROMATIC HYDROCARBONS
CHRISTOPHER A. MARWOOD,†‡ KEITH R. SOLOMON,‡ and BRUCE M. GREENBERG*†
†Department of Biology, University of Waterloo, Waterloo, Ontario N2L 3G1, Canada
‡Centre for Toxicology, University of Guelph, Guelph, Ontario N1G 2W1, Canada
( Received 20 April 2000; Accepted 28 August 2000)
Abstract—Chlorophyll-a fluorescence induction is a rapid technique for measuring photosynthetic electron transport in plants. To
assess chlorophyll-a fluorescence as a bioindicator of effects of polycyclic aromatic hydrocarbon mixtures, chlorophyll-a fluorescence
parameters and plant growth responses to exposure to the wood preservative creosote were examined in the aquatic plants Lemna
gibba and Myriophyllum spicatum. Exposure to creosote inhibited growth of L. gibba (EC50 5 7.2 mg/L total polycyclic aromatic
hydrocarbons) and M. spicatum (EC50 5 2.6 mg/L) despite differences in physiology. Creosote also diminished maximum PSII
efficiency (Fv/Fm) (EC50 5 36 and 13 mg/L for L. gibba and M. spicatum) and the effective yield of photosystem II photochemistry
(DF/Fm9 ) (EC50 5 13 and 15 mg/L for L. gibba and M. spicatum). The similarity between growth and chlorophyll-a fluorescence
EC50s and slopes of the response curves suggests a close mechanistic link between these end points. The predictive power of
chlorophyll-a fluorescence as a bioindicator of whole-organism effects applied to complex contaminant mixtures is discussed.
Keywords—Aquatic plants
Chlorophyll fluorescence
Polycyclic aromatic hydrocarbons
Bioindicator
Photosynthesis
els in the light-adapted plant are used to calculate DF/F9m , the
quantum yield of photochemistry in functional PSII reaction
centers during steady-state photosynthesis. This parameter represents the efficiency with which excitation energy captured
by the chlorophyll antenna is used in electron transport. It is
closely related to the quantum yield of carbon assimilation [9]
and to the photochemical quenching parameter describing photochemical quenching (qP). The qP term describes the fraction
of fluorescence quenched in a light-adapted plant relative to
a dark-adapted plant due to induction of photochemistry and
estimates the extent to which electron transport is restricted at
PSII [6]. Nonphotochemical quenching of fluorescence, described by qN, is due to a number of biophysical and biochemical mechanisms that dissipate excess excitation energy
in the pigment antenna to protect PSII from overexcitation
[7,10]. Chlorophyll-a fluorescence can potentially detect impacts of contaminants not acting directly on electron transport
since chemicals that inhibit any cellular process downstream
of PSII, such as carbon assimilation or respiration, or damage
membranes or proteins associated with photosynthetic electron
transport will lead to excitation pressure on PSII [11,12].
We have previously shown that chlorophyll-a fluorescence
can be used to detect inhibition of photosynthetic electron
transport in plants exposed to polycyclic aromatic hydrocarbons (PAHs) and photomodified PAHs [5,13,14]. The PAHs
consist of aromatic ring compounds and are common contaminants in aqueous environments [15,16]. The toxicity of many
PAHs is greatly enhanced in the presence of ultraviolet radiation [15,17]. The observed photoinduced toxicity is generally
attributed to generation of singlet oxygen and hydroxyl radicals in the tissue of organisms exposed to PAHs, leading to
oxidation of membrane lipids and other biomolecules [18,19].
However, the light-mediated production of oxyPAHs, which
often have greater aqueous solubility and toxicity than the
INTRODUCTION
Risk assessment requires rapid, sensitive techniques to assess exposure and effects of environmental contaminants on
organisms. Bioindicators are rapid physiological or biochemical assays that can often detect effects of contaminants in
organisms even when the toxicant is unknown. While an abundance of potentially useful assays seems to be available, the
primary impediment to their widespread use is the lack of a
demonstrated link between effects at the suborganism level
with effects on growth or reproduction at the individual level
and with higher levels of biological organization, such as populations or communities [1]. A bioindicator for which there is
a defined link to growth of the organism can be used in a
predictive manner to assess the potential impacts of a contaminant before detrimental effects to populations occur.
Photosynthetic electron transport is a universal feature of
higher plants, algae, and cyanobacteria. This essential process
is sensitive to many classes of environmental contaminants,
including herbicides [2,3], metals [4], and organic contaminants [5]. Pulse amplitude modulated (PAM) chlorophyll-a
fluorescence is a rapid method for measuring photosynthetic
electron transport from plants in vivo, and it requires very
little sample preparation [6]. Stimulation of fluorescence from
chlorophyll-a in photosystem II (PSII) reaction centers under
different light conditions produces several parameters, each
describing the efficiency of a photochemical reaction or process within the photosynthetic apparatus. The variable fluorescence ratio Fv /Fm, which describes the maximum efficiency
of PSII photochemistry [7,8], is calculated from the minimum
fluorescence (Fo) in the dark and the maximum fluorescence
(Fm) on application of a saturating light to the dark-adapted
plant. The minimum (Fs) and maximum (F9m) fluorescence lev* To whom correspondence may be addressed
([email protected]).
890
Environ. Toxicol. Chem. 20, 2001
Chlorophyll fluorescence as bioindicator of creosote toxicity
parent PAH, is also an important mechanism of toxicity
[12,20].
Organisms are rarely exposed to single PAHs in the aqueous
environment. Samples from contaminated sites contain complex mixtures of PAHs [16] as well as PAH photoproducts
(X.-D. Huang, B.J. McConkey, and B. Greenberg, unpublished
observations). The PAHs in mixtures often are in greater aqueous concentrations than single PAHs because of cosolubility
effects [21]. Therefore, to assess the applicability of chlorophyll-a fluorescence as a bioindicator, we examined the response of various fluorescence parameters in plants exposed
to the wood preservative creosote, a mixture composed primarily of PAHs, with minor quantities of oxyPAHs and substituted PAHs [22]. Two aquatic plant species, Myriophyllum
spicatum (Eurasian Milfoil) and Lemna gibba (duckweed),
were exposed to a range of creosote concentrations in defined
growth medium and lighting conditions. The concentration
response of chlorophyll-a fluorescence parameters was compared to conventional growth-based end points.
MATERIALS AND METHODS
Plant culture
Cultures of Lemna gibba L. (G-3) were propagated axenically on half-strength Hutner’s medium fortified with 30 g/
L sucrose in large glass flasks under 50 mmol/m2/s continuous
photosynthetically active radiation (PAR, 400–700 nm) supplied by cool-white fluorescent lamps.
Whole Myriophyllum spicatum L. (Haloragaceae) plants
were collected locally from a shallow pond. The meristematic
tips were excised and washed in 2% hypochlorite solution to
sterilize the plant surface. Plants were cultured axenically in
capped 25 3 150-mm quartz tubes containing Andrew’s medium [3] fortified with 15 g/L sucrose. The plants were maintained in a growth chamber at 258C under 100 mmol/m2/s PAR
supplied by cool-white fluorescent lamps on a 16:8-h light:
dark cycle.
891
bers at 258C. The medium containing dimethylsulfoxide with
or without creosote was replaced every 2 d to maintain concentrations of nutrients and creosote. The entire experiment
was repeated three times.
Myriophyllum spicatum were exposed to creosote in 12-d
static renewal toxicity tests [26]. Solutions of 1,0003 creosote
in dimethylsulfoxide were stirred into test tubes containing 40
ml sterile growth medium to achieve nominal creosote concentrations from 0.01 to 100 mg/L. Dimethylsulfoxide at 0.1%
has no measurable effect on Myriophyllum sibiricum [26] and
had no effect on growth or photosynthesis of M. spicatum in
our experiments. The top 3 cm of shoots were removed and
transferred aseptically to the tubes containing creosote solutions. Three replicate tubes at each concentration were incubated in growth chambers on a 16:8-h light:dark cycle at 258C.
The entire experiment was repeated three times.
During creosote exposures, plants received 110 mmol/m2/s
continuous simulated solar radiation, which contains PAR
(400–700 nm), ultraviolet-A (UVA, 320–400 nm) and ultraviolet-B (UVB, 290–320 nm) radiation. Photosynthetically active radiation was generated by cool-white fluorescent lamps.
Ultraviolet radiation was generated by photoreactor lamps
(Southern New England Ultraviolet, Branford, CT, USA) with
peak irradiance at 350 and 300 nm for UVA and UVB, respectively. Light from the 300-nm photoreactor lamps was
filtered through cheesecloth to attenuate UVB to the desired
fluence rate. The light from all lamps was filtered through a
single layer of cellulose acetate (0.08 mm, Johnston Plastics,
Mississauga, ON, Canada) to remove wavelengths lower than
290 nm. The spectral output of the lighting was measured using
a calibrated spectroradiometer (Oriel, Stratford, CT, USA) with
a diode array detector and an integrating sphere. The simulated
solar radiation light source contained 99.5 mmol/m2/s PAR,
11.3 mmol/m2/s UVA, and 0.83 mmol/m2/s UVB. Although
these fluence rates are much lower than those found in sunlight,
the ratio of PAR to UVA to UVB is approximately equal to
the ratio in sunlight in southern Ontario during the summer,
except for a slightly greater UVB component [13].
Creosote exposure
Lemma gibba were exposed to marine-grade liquid creosote
(Stella Jones, New Westminster, BC, Canada) in 8-d static
renewal toxicity tests [23]. The PAH composition of the creosote had been partially characterized previously by high-performance liquid chromatography. The 15 most abundant PAHs
detected were found to constitute 38.4% (w/v) of the creosote
[24]. Aqueous solutions of creosote were made up at nine
nominal concentrations of total creosote in a logarithmic series
from 0.01 to 100 mg/L. All creosote concentrations are reported in the figures and tables as a sum of the 15 most abundant PAHs (SPAH in mg/L). To convert between SPAH and
total creosote, multiply SPAH by 2.6. Ten microliters of
1,0003 stock solutions of creosote dissolved in dimethylsulfoxide (Aldrich Chemicals, Milwaukee, WI, USA) were diluted
into 10 ml of plant growth medium in 5-cm-diameter polystyrene petri dishes with constant stirring. The final concentration of dimethylsulfoxide in the growth medium in all treatments was 0.1% (v/v), which does not diminish growth or
influence the toxicity of PAHs to L. gibba [25]. The PAHs
introduced to aqueous medium in this manner are rapidly assimilated by L. gibba and are quantitatively extractable with
organic solvents. Two L. gibba colonies (eight fronds) were
transferred to petri dishes containing creosote. Three replicate
samples at each concentration were incubated in growth cham-
Leaf chlorophyll concentration
Chlorophyll was extracted from whole L. gibba plants and
from the leaves of M. spicatum. Weighed plants or leaves were
extracted in N,N9-dimethylformamide (Aldrich Chemicals) for
24 h at 48C in the dark. Chlorophyll-a and chlorophyll-b concentrations were measured simultaneously by spectrophotometry [27]. Leaf chlorophyll concentrations were calculated on
a fresh-weight basis.
Growth end points
The number of L. gibba fronds was counted every 2 d, and
a growth rate was calculated on the basis of the increase in
fronds over 8 d according to an exponential growth model.
Growth rate was calculated as
t21·log(Ft /F0)·(log 2)21
(1)
where t is the duration of the test in days, F0 is the initial
number of fronds, and Ft is the number of fronds at the end
of the test.
Growth measurements were taken from M. spicatum at the
end of the 12-d exposure. The shoot length and number of leaf
nodes of the main shoot as well as side shoots were recorded.
The number and length of all roots were recorded. Fresh weight
was measured from whole plants after excess water was re-
892
C.A. Marwood et al.
moved by blotting. Since the aseptic technique precluded
weighing the plants at the beginning of the test, the increase
in fresh weight over the initial weight was calculated by subtracting the average fresh weight of 10 unexposed shoots 3
cm in length.
Chlorophyll-a fluorescence
Chlorophyll-a fluorescence induction was measured from
L. gibba after 4 and 8 d of exposure to creosote and from M.
spicatum after 4, 8, and 12 d of exposure. Plants were removed
from the growth chamber and dark adapted for 15 min prior
to measurement. Chlorophyll-a fluorescence was measured
from whole plants using a PAM fluorometer (PAM-2000, Walz,
Effeltrich, Germany). All light sources and fluorescence measurements were directed through a fiberoptic light guide. For
L. gibba, the end of the light guide was positioned 1 mm
above plants floating in petri dishes; for M. spicatum, measurements were taken directly through the wall of the test tube
without removing the plants from the growth medium.
Since chlorophyll-a fluorescence parameters vary considerably under different measurement conditions, fluorescence
induction curves were optimized for each plant by incrementally adjusting the light levels on the PAM fluorometer. A weak
measuring light (,1 mmol/m2/s) was used to determine Fo, the
minimum fluorescence obtained in the dark-adapted state.
Maximum fluorescence (Fm) was determined for the darkadapted state by applying a saturating pulse of white light
(4,000 mmol/m2/s, 600-ms duration) from a halogen lamp.
Pulses of white light at greater fluence rates did not change
Fm, indicating that the fluence rate used was saturating to the
dark-adapted plants. Actinic light from red diodes (655 nm)
was used to induce electron transport. An actinic fluence rate
of 30 mmol/m2/s was found to be optimal for both L. gibba
and M. spicatum. Steady-state fluorescence (Fs) was achieved
after exposure to actinic light for 10 min, as evidenced by
unchanging fluorescence levels. Maximum fluorescence under
steady-state conditions (Fm9 ) was determined by applying pulses of the saturating white light every 60 s when the actinic
light was on. Far-red light was applied for 5 s immediately
after the actinic light was turned off for the determination of
Fo9, the minimum fluorescence in the light-adapted state.
Maximum quantum efficiency of PSII photochemistry was
calculated as
Fv /Fm 5 (Fm 2 Fo)/Fm
(2)
The effective quantum yield of PSII photochemistry [9] was
calculated as
DF/Fm9 5 (Fm9 2 Fs)/Fm9
(3)
Photochemical and nonphotochemical quenching of fluorescence parameters were calculated as
qP 5 (Fm9 2 Fs )/(Fm9 2 Fo9 )
and
(4)
qN 5 1 2 (Fm9 2 F9o)/(Fm 2 Fo ).
(5)
Because nonphotochemical quenching generally increases in
plants under stress, the parameter 1 2 qN was used as an index
for nonphotochemical quenching.
Statistical analysis
Chlorophyll-a fluorescence data often do not exhibit a
Gaussian (normal) distribution [28]. Common descriptors of
variance, such as standard deviation, which assume normal
Fig. 1. Inhibition of growth of Lemna gibba exposed to creosote. The
number of fronds produced after 8 d of exposure to creosote was used
to calculate a growth rate. Data from three separate experiments are
shown as a box and whisker plot. The median, 25th, and 75th percentiles of four replicate measurements are represented by the center
line and the upper and lower edges of the box. The 5th and 95th
percentiles are shown as whiskers below and above the box. A curve
was fit to the mean values at each concentration, as a percentage of
control, using nonlinear regression.
distribution, are inappropriate for this type of data. Thus, we
have presented concentration–response curves of chlorophylla fluorescence data using box plots with median, 25th, and
75th percentiles to indicate the distribution of the data. For
comparison, plots of leaf chlorophyll concentration and growth
end points are also presented in this manner. The response data
were plotted on the basis of SPAH, the sum concentration of
the 15 predominant PAHs at each creosote concentration [24],
which was calculated by multiplying the nominal creosote concentration by the fraction of PAHs in the creosote we used
(38.4%). Chlorophyll-a fluorescence measurements and the
shape of the concentration–response curves were similar at
each time point; therefore, only curves from the final measurement (8 d for L. gibba and 12 d for M. spicatum) are
shown.
The concentration response for each end point was described using a trimmed logit model suitable for continuous
response data [29]. Iterative regression minimizing x2 adjusted
for degrees of freedom was used to fit the function
% Inhibition 5 100(1 1 eb(x2m))21
(6)
to the measured data, where x is the logarithm of the SPAH
concentration, m is the logarithm of the EC50, and b represents
the slope of the concentration–response curve [20,29]. The
model used only data from the steepest part of the response
curve by trimming data for which the mean response fell outside 1% of the predicted response (1 , % Inhibition , 99).
In this way, the regressions are driven by the data in the center
of the distributions. The regression provided estimates of variance for m and b that were used to calculate 95% confidence
intervals for the EC50 and slope. Only EC50s are presented
in this study, but any level of inhibition (e.g., 20%) could be
determined by substituting that value into the equation and
solving for the concentration x.
RESULTS
Creosote effects on plant growth
The growth rate of L. gibba was inhibited by creosote in
a concentration-dependent manner (Fig. 1). Plants in highercreosote treatments were smaller than controls, and colonies
consisted of one or two fronds rather than three or four fronds,
as in control plants. There was almost no new frond production
893
points exhibited hormesis; that is, these responses were stimulated rather than inhibited in plants exposed to low creosote
concentrations. Hormesis in these end points resulted in reduced fit of the regression curve (lower R2) but had little effect
on the EC50 because the curve-fitting procedure is driven
primarily by data close to the EC50 and is not susceptible to
outliers in the tails of the data set [29]. Estimated EC50s for
inhibition of growth were higher for L. gibba (7.2 mg/L SPAH;
Table 1) than for M. spicatum (0.14–2.6 mg/L SPAH, depending on the end point; Table 2). Although the growth end
points in M. spicatum were more sensitive than growth in L.
gibba, higher variability existed in the former, especially shoot
length and root length, resulting in poorer regression coefficients.
Creosote effects on leaf chlorophyll concentrations
Fig. 2. Inhibition of growth of Myriophyllum spicatum exposed to
creosote for 12 d. (A) Increase in fresh weight. (B) Increase in total
shoot length. (C) Total length of all roots produced.
after 8 d at the highest concentration tested, and remaining
plants were very small and chlorotic.
All growth-based end points in M. spicatum were also diminished by exposure to creosote (Fig. 2). Growth in the highest-creosote treatment was completely inhibited, resulting in
nearly zero accumulation of biomass (fresh weight) (Fig. 2A).
Increase in shoot length was inhibited by creosote (Fig. 2B)
because of diminished shoot elongation, fewer nodes, and fewer new shoots (data not shown). However, low creosote concentrations stimulated shoot elongation up to 125% of control
values. Root production and elongation were inhibited by very
low creosote concentrations and were completely suppressed
at concentrations greater than 1 mg/L SPAH (Fig. 2C).
In general, growth end points had high experimental variability. In addition, the shoot length and fresh-weight end
Plants exposed to the highest creosote concentration were
bleached after only 2 d, indicating very low chlorophyll concentrations. Total chlorophyll extracted from L. gibba at the
end of the 8-d exposure confirmed that chlorophyll was diminished in a concentration-dependent manner (Fig. 3A) from
approximately 0.7 mg/mg fresh weight in controls to less than
0.1 mg/mg in the highest-creosote treatment. Concentrations
of both chlorophyll-a and chlorophyll-b were diminished (data
not shown).
Leaf chlorophyll concentrations in unexposed M. spicatum
were approximately 1 mg/mg fresh weight; this value is somewhat higher than the concentration in unexposed L. gibba.
Exposure of M. spicatum to creosote resulted in diminished
chlorophyll at concentrations greater than 1 mg/L SPAH (Fig.
3B). Although moderate variability in this end point was seen,
regressions produced good fits to the data set. The EC50s for
chlorophyll concentration for L. gibba and M. spicatum were
similar: 3.4 and 5.6 mg/L SPAH, respectively.
Creosote effects on photosynthesis
Exposure to creosote had strong, distinct effects on chlorophyll-a fluorescence scans in L. gibba. Variable fluorescence
in the dark-adapted (Fv 5 Fm/Fo) and light-adapted (F9v 5 Fm9/
F9o) states was diminished substantially compared to control
plants because of diminishment of both Fm and Fo (Fig. 4). At
the highest concentration (38 mg/L SPAH), the increase in
fluorescence induced by the saturating light under steady-state
photosynthesis (Fm9) was nearly indistinguishable from back-
Table 1. Regression terms describing concentration–response curves for growth, leaf chlorophyll (chl) concentration, maximum efficiency of
photochemistry (Fv /Fm), quantum yield of photochemistry (DF/F9m ), photochemical quenching (qP), and nonphotochemical quenching (1 2 qN)
from Lemna gibba exposed to creosote. The median response concentration (EC50) and slope of the curve (b) were estimated by fitting the
mean response from three experiments to a nonlinear function. The deviation of the data from the prediction (x2) and the correlation factor of
the predicted curve (R2) are shown
Measurement end point
Growth
Leaf chl concn.
Fv /F m
DF/Fm′
qP
1 2 qN
a
Exposure
(d)
EC50 (95% CI)
(mg/L SPAH)a
8
8
4
8
4
8
4
8
4
8
7.2 (3.8, 13.8)
3.4 (2.0, 6.0)
.38
36 (4.8, 270)
22 (9.4, 53)
13 (7.3, 22)
.38
.38
.38
.38
b (95% CI)
22.2
22.2
21.8
22.0
22.1
22.8
21.4
20.62
20.69
21.0
(23.3,
(23.1,
(23.1,
(24.2,
(23.2,
(24.2,
(26.4,
(22.0,
(21.3,
(22.5,
21.1)
21.3)
20.49)
0.31)
20.94)
21.5)
3.6 )
0.75)
20.06)
20.37)
x2
R2
0.259
0.020
0.037
0.192
0.092
0.597
0.079
0.386
0.144
0.380
0.966
0.997
0.989
0.989
0.988
0.963
0.931
0.297
0.774
0.329
Estimates are based on the sum concentration of the 15 most abundant polycyclic aromatic hydrocarbons in creosote.
894
C.A. Marwood et al.
Table 2. Regression terms describing concentration–response curves for growth, leaf chlorophyll (chl) concentration, maximum efficiency of
photochemistry (Fv /Fm ), quantum yield of photochemistry (DF/F9m ), and photochemical quenching (qP ), from Myriophyllum spicatum exposed
to creosote. The median response concentration (EC50) and slope of the curve (b) were estimated by fitting the mean response from three
experiments to a nonlinear function. The deviation of the data from the prediction (x2) and the correlation factor of the predicted curve (R2) are
shown. Curves could not be fit to the nonphotochemical quenching (1 2 qN ) parameter
Measurement
end point
Exposure
(d)
Fresh weight
Shoot length
Root length
Chl content
Fv /F m
12
12
12
12
4
8
12
4
8
12
4
8
12
DF/Fm′
qP
a
b
EC50 (95% CI)
(mg/L SPAH)a
2.6
1.6
0.14
5.6
(1.4, 4.9)
(1.0, 2.7)
(0.06, 0.30)
(2.9, 10.7)
.38
36 (12, 113)
13 (6.0, 28)
.38.4
34 (8.5, 140)
15 (5.6, 40)
.38.4
.38.4
.38.4
b (95% CI)
22.3
21.8
23.0
21.8
21.5
21.8
22.8
21.4
21.6
22.2
(23.5, 21.1)
(22.3, 21.3)
(25.4, 20.68)
(22.6, 20.94)
(22.4, 20.47)
(22.9, 20.76)
(24.6, 21.0)
(22.2, 20.47)
(22.7, 20.57)
(23.7, 20.70)
NAb
NA
21.4 (217, 14)
x2
R2
1.53
3.54
0.849
0.160
0.202
0.218
0.151
0.216
0.302
0.174
NA
NA
0.494
0.953
0.814
0.894
0.986
0.968
0.953
0.987
0.943
0.961
0.964
NA
NA
0.772
Estimates are based on the sum concentration of the 15 most abundant polycyclic aromatic hydrocarbons in creosote.
NA 5 not applicable; the data could not be fit to the model because of a lack of response.
ground fluorescence, and little recovery of Fm9 occurred on the
removal of the actinic light (Fig. 4D). Chlorophyll fluorescence
scans from plants treated with lower creosote concentrations
appeared similar to controls.
Photosynthetic parameters were diminished in L. gibba
only at the upper range of creosote concentrations tested (.10
mg/L SPAH) (Fig. 5). Greater concentrations were not tested
because the highest concentration of creosote in this experiment was above the aqueous solubility limits of most PAHs,
and attempts to suspend greater concentrations of creosote in
the media were unsuccessful. Thus, it was not possible to
obtain complete concentration–response curves for all fluorescence parameters, but accurate predictions of the EC50 were
still possible. In general, toxicity was greater after longer exposures to creosote. Exposure to creosote diminished some
chlorophyll fluorescence parameters in a concentration-dependent manner but not others. The Fv/Fm and DF/Fm9 values were
diminished at higher creosote concentrations (Fig. 5A and B),
Fig. 3. Effect of creosote on leaf chlorophyll concentration in (A)
Lemna gibba and (B) Myriophyllum spicatum. Chlorophyll was extracted from plant leaves after 8 d for L. gibba or 12 d for M. spicatum.
but a clear concentration response was not observed with qP
or 1 2 qN (Fig. 5C and D). At the highest creosote concentration, the quenching parameters qP and 1 2 qN could not
be measured accurately from the fluorescence induction scan
because of a lack of signal from low chlorophyll concentrations. However, it cannot be assumed that these parameters
were zero because both the numerator and the denominator
approach zero when the fluorescence signal is low.
Effects of creosote on chlorophyll-a fluorescence in M.
spicatum were similar to effects on L. gibba. Responses of
Fv/Fm and DF/Fm9 to creosote exposure were clearly concentration dependent (Fig. 6A and B). While no inhibition of these
parameters was seen at concentrations below 1 mg/L SPAH,
the highest concentration diminished Fv/Fm to 20% of control
values after 12 d. This parameter in control plants and plants
exposed to low creosote concentrations had little variability,
but at higher PAH concentrations moderate variability was
seen. Similar to L. gibba, the qP and 1 2 qN parameters were
not diminished by creosote, even after 12 d of exposure (Fig.
6C and D). At high creosote concentrations, qN was diminished rather than increased, resulting in 1 2 qN values greater
than one. This response was greatest in plants exposed for 12 d.
The logit function fit to the response data adequately described the concentration responses of Fv/Fm and DF/Fm9 , with
R2 . 0.94. The estimated EC50s for these end points were
approximately an order of magnitude greater than the growthbased end points. The model estimated 8-d EC50s for Fv/Fm
and DF/Fm9 of 36 and 13 mg/L SPAH, respectively, in L. gibba
(Table 1). The EC50s in M. spicatum were similar to L. gibba
after 8 d but decreased after 12 d of exposure to 13 and 15
mg/L SPAH for Fv/Fm and DF/Fm9 , respectively (Table 2).
Meaningful curves could not be fit for the quenching parameters qP and 1 2 qN because of the lack of a positive (diminished) response.
The slopes of the response curves (estimated by the b term
in the model) were similar for Fv/Fm and DF/Fm9 at the end of
the exposure in both L. gibba (Table 1) and M. spicatum (Table
2). In addition, the slopes for Fv/Fm and DF/Fm9 were similar
to the slopes estimated for inhibition of chlorophyll concentration and growth end points. In L. gibba, the slopes for Fv/
Fm, DF/Fm9 , growth, and chlorophyll concentration were all
Fig. 4. Effect of creosote on chlorophyll fluorescence in Lemna gibba.
chlorophyll-a fluorescence scans were measured from plants exposed
for 8 d to creosote at A) 0, (B) 10, (C) 30, or (D) 100 mg/L creosote.
Minimum (Fo) and maximum (Fm) fluorescence levels were determined from dark-adapted plants at the beginning of the scan. Plants
were brought to steady-state photosynthesis using a red actinic light
as indicated. Steady-state (Fs) and light-adapted maximum (Fm9) fluorescence was measured after approx. 8 to 10 min. Minimum fluorescence under far-red light (F9o) was measured immediately after the
actinic light was removed. Signals were normalized to Fo for comparison.
within the range 22.8 , b , 22.0 (Table 1). The slopes of
responses measured in M. spicatum after 12 d were somewhat
more variable, but all fell within the range 23.0 , b , 21.8
and were not statistically different (Table 2). The response for
root production had the steepest slope but also had higher
variability in the data, resulting in a large confidence interval.
DISCUSSION
A major impediment to ecotoxicological assessments is a
lack of validated bioindicators that can be used to predict
contaminant effects in natural ecosystems. There are numerous
physiological or biochemical assays that detect effects of toxicants on various cellular processes, but the applicability of
these assays as predictive bioindicators is often compromised
by a lack of any quantitative relationship with higher-level end
points [1]. The ability of a bioindicator to predict effects at
higher biological levels can be determined by comparing sub-
895
Fig. 5. Chlorophyll-a fluorescence parameters from Lemna gibba exposed to creosote. Maximum efficiency of photochemistry (Fv/Fm),
quantum yield of photochemistry (DF/Fm9), photochemical quenching
(qP), and nonphotochemical quenching (1 2 qN) were measured from
plants after 8 d of exposure. The missing data at high concentrations
represent severely chlorotic plants from which measurements could
not be obtained.
organism and whole-organism responses. If the response is
tested over the entire toxicant concentration range rather than
simply contrasting point estimates, such as EC50s, mechanistic
relationships between different biological levels can be better
postulated [14]. Bioindicators for which correspondence exists
between both the slope of the response curve and the median
toxicity end point (e.g., EC50) have a close mechanistic relationship with effects at higher levels of organization and are
more valuable as predictive tools.
The results of this study show that chlorophyll-a fluorescence in aquatic plants is a bioindicator that is somewhat predictive of whole-plant toxicity. Inhibition of photosynthesis in
L. gibba and M. spicatum exposed to PAHs can be linked
with inhibition of plant growth. The chlorophyll-a fluorescence
parameters describing the quantum efficiency of PSII (Fv/Fm)
and the quantum yield of PSII photochemistry (DF/Fm9 ) were
diminished at PAH concentrations that also diminished leaf
chlorophyll concentration and inhibited growth of L. gibba
(frond production) and M. spicatum (shoot growth, root
growth, and fresh weight). Although growth end points were
896
Fig. 6. Chlorophyll-a fluorescence parameters from Myriophyllum
spicatum exposed to creosote. Maximum efficiency of photochemistry
(Fv/Fm), quantum yield of photochemistry (DF/Fm9), photochemical
quenching (qP) and nonphotochemical quenching (1 2 qN) were measured from plants after 12 d of exposure.
generally more sensitive, EC50s for inhibition of chlorophylla fluorescence and EC50s based on growth end points were
within an order of magnitude, and the DF/Fm9 parameter in L.
gibba was almost as sensitive as frond production. In addition,
the slopes of the concentration–response curves were similar
for the growth and chlorophyll-a fluorescence end points, implying that toxicity from PAHs can be associated primarily
with effects on the photosynthetic apparatus in these plants.
Some components of the mixture apparently affected nonphotosynthetic processes (e.g., respiration and cell division), resulting in diminished growth at low concentrations of creosote.
Chlorophyll-a fluorescence has the ability to be a powerful
bioindicator of PAH effects despite the lower sensitivity of
these end points compared to growth end points. Even a moderate diminishment of chlorophyll-a fluorescence will translate
to deleterious effects on photosynthetic energy production because of the tight relationship between these two processes.
Thus, a plant in which chlorophyll-a fluorescence is diminished by even 50% will eventually suffer severely diminished
plant growth. The measurement of plant growth as a toxicity
end point is not practical in the field since time-zero measurements usually cannot be taken. Chlorophyll-a fluores-
C.A. Marwood et al.
cence, alternatively, is ideal for rapidly assessing instantaneous
and/or continuous effects of contaminants in the field. Thus,
although chlorophyll-a fluorescence parameters were not diminished at exactly the same concentrations as growth for the
contaminants used in this study, this assay has the ability to
predict effects at the whole-organism level. Our results reveal
that when impacts on chlorophyll-a fluorescence are observed,
effects such as diminished growth are most likely occurring
as well. We have found in previous studies with phytoplankton
[30] and in unpublished work with aquatic plants that this
technique is an elegant method for detecting contaminant impacts in the field.
Of the chlorophyll-a fluorescence parameters examined in
this study, DF/Fm9 was the most sensitive end point to PAH
toxicity. The DF/Fm9 parameter is strongly correlated with photosynthetic C uptake and O2 evolution [9,31]. Mechanistically,
DF/Fm9 is a function of the fraction of open reaction centers
and the quantum efficiency of electron transport through PSII
[9,32]. Thus, this parameter integrates the effects of many
processes that contribute to inhibition of electron transport
either by decreasing the quantum yield of photochemistry or
by increasing the yield of nonphotochemical (thermal) excitation decay processes. The sensitivity to PAH toxicity and
relevance to energy production make DF/Fm9 the most suitable
chlorophyll-a fluorescence parameter to use as a bioindicator
of toxicity in plants.
Diminished Fv /Fm and DF/Fm9 can indicate a large proportion
of inactive PSII reaction centers due to oxidation or degradation of D1 proteins [8]. Severely reduced pigment concentrations in reaction centers would also diminish these parameters, which would be consistent with the diminished chlorophyll content of plants exposed to creosote. Diminished PSII
efficiency can also indicate inhibition of electron transport at
other sites in the photosynthetic apparatus downstream of PSII.
Because PSII is the first step in photosynthetic electron transport, inhibition of electron transport anywhere within the electron transport chain exerts excitation pressure on PSII, which
can be detected as diminished Fv/Fm or DF/Fm9 [11]. Photoproducts of anthracene are suspected to target photosynthetic
components downstream of PSII, most likely cytochrome b6/
f [33,34]. In the respiratory electron transfer pathway, oxyPAHs specifically inhibit cytochrome bc1, the component analogous to cytochrome b6/f in the photosynthetic pathway [35].
The lack of response of the chlorophyll-a fluorescence parameter describing photosynthetic energy quenching (qP) suggests that PAHs had no direct impact on the primary photochemical reactions of PSII. Some quinones are known to inhibit PSII directly because of structural similarity with the QBbinding niche in PSII [36]. It has been demonstrated that
quinones are the primary photoproducts produced on exposure
of many PAHs to light [20]. In a mixture of PAHs exposed
to UV radiation in an aquatic matrix, quinones capable of
binding the QB site were almost certainly present. However,
the lack of effect on qP suggests that quinones capable of
binding PSII were not present in significant concentrations
within the thylakoid membranes.
The variable and inconsistent response of the 1 2 qN parameter also suggests PAHs had no direct effect on electron
transport. The parameter qN represents the loss of fluorescence
yield due to nonphotochemical quenching of chlorophyll excitation, which dissipates excess energy as heat. These mechanisms are induced by excessive excitation pressure on PSII
from strong light [8], low temperature [11], or blocked electron
transport [37]. While 1 2 qN tended to decrease slightly at
higher concentrations of creosote in L. gibba, the lack of a
strong response implies that electron transport at PSII was not
blocked to a great extent in plants exposed to creosote. The
diminished qN (elevated 1 2 qN) observed in M. spicatum
may have been caused by chlorosis or by disruption of the
thylakoid membrane proton gradient required for nonphotochemical quenching processes [10].
The different responses of the various chlorophyll-a fluorescence parameters, when taken together, support a mechanism of PAH toxicity in which photosynthetic efficiency was
impaired as a result of general damage to essential cellular
components, possibly from reactive oxygen species. For instance, the presence of excited chlorophyll pigments in thylakoid membranes in which electron transport has been
blocked leads to production of singlet oxygen [7]. Alternatively, free radicals may be produced in the chloroplast via
photosensitization reactions by PAHs or oxyPAHs on absorbance of UV light [18,19]. The most abundant photoproducts
of PAHs are anthraquinones, which are very efficient redox
cyclers, producing appreciable quantities of superoxides in the
light when a reducing agent is present [38]. Regardless of the
source, the presence of radical species in the chloroplast leads
to oxidation of membranes, proteins (e.g., the D1 protein of
PSII), and pigments. Based on the loss of leaf chlorophyll in
plants exposed to low PAH concentrations, destruction of chlorophyll pigments is apparently an important mechanism of
PAH toxicity in plants and may precede damage to the proteins
of the photosynthetic apparatus. However, the different PAH
components in creosote likely have multiple effects at the biochemical level, such as disruption of electron transport by
oxyPAHs that act as electron acceptors [36] or inhibit electron
transport [34], or loss of the thylakoid proton gradient due to
membrane oxidation.
The results of this study using the PAM fluorescence technique support previous studies of PAH effects on electron
transport. During the early events of the chlorophyll-a fluorescence induction curve, the fluorescence ratio Fq/Fm is a
sensitive indicator of electron transport in L. gibba exposed
to photomodified anthracene, which inhibits electron transport
downstream of PSII [33]. The half-time required to reduce the
plastoquinone pool (t½) in L. gibba exposed to creosote has
also been shown to be a sensitive end point to PAH toxicity
(EC50 5 11.6 ppm creosote) [14]. Gensemer et al. found inhibition of frond production (EC50 5 10.7 ppm creosote) and
Fv/Fm (EC50 5 16.6 ppm creosote) were also sensitive end
points [14]. The EC50s for growth and Fv/Fm were somewhat
lower than those found in this study, but the creosote used in
that study was weathered (stirred in air) prior to use and likely
contained a different assortment of PAHs.
The hormesis response of the M. spicatum growth end point
to values greater than those measured in unexposed plants is
a potentially confounding factor in bioindicator applications.
When hormesis responses occur, toxicity estimates such as
EC50s can be based on the response relative to the controls,
the greatest value observed, or the low concentrations pooled
with the real control response [39]. With respect to calculation
of EC50s in this study, hormesis effects were ignored because
of the small relative stimulation response observed. The
trimmed logit model described the data reasonably well, with
only a minor deviation produced by the stimulation response.
This statistical model resists shifts in the logit function due to
stimulation by trimming responses from the data set that lie
897
outside specified limits during the iterative regression. Thus,
only those points close to the midpoint in the concentration
response curve were used in the EC50 estimate. Other statistical approaches exist for describing larger stimulation responses, such as nonlinear functions that can accommodate
hormesis responses.
The two plant species used in this study responded similarly
to creosote exposure. For responses after 4 or 8 d, Fv/Fm and
DF/Fm9 were inhibited at nearly the same SPAH concentrations
in both plant species. Growth of L. gibba and M. spicatum
was inhibited at similar concentrations, even though different
measurement end points were examined. The PAHs apparently
had similar effects on electron transport in both plant species.
The similar responses of Fv/Fm and DF/Fm9 to PAHs in these
two plant species and in phytoplankton [30] is encouraging.
These parameters may be appropriate bioindicators that can
be measured in different plants in the field exposed to PAHs.
An important aspect of this assay is that unlike growth-based
endpoints, chlorophyll-a fluorescence can be used to rapidly
assess contaminant effects in the field at any time, independent
of the duration of the exposure.
Acknowledgement—This study was supported by equipment and
grants from Environment Canada through the Canadian Network of
Toxicology Centres and by the Natural Sciences and Engineering
Research Council grants to B.M. Greenburg.
REFERENCES
1. Huggett RJ, Kimerle RA, Mehrle PM, Bergman HL. 1992. Biomarkers: Biochemical, Physiological, and Histological Markers
of Anthropogenic Stress. Lewis, Boca Raton, FL, USA.
2. Ralph PJ. 2000. Herbicide toxicity of Halophila ovalis assessed
by chlorophyll a fluorescence. Aquat Bot 66:141–152.
3. Christopher SV, Bird KT. 1992. The effects of herbicides on development of Myriophyllum spicatum L. cultured in vitro. J Environ Qual 21:203–207.
4. Laberge D, Chartrand J, Rouillon R, Carpentier R. 1999. In vitro
phytotoxicity testing using immobilized spinach thylakoids. Environ Toxicol Chem 18:2851–2858.
5. Huang X-D, Babu TS, Marwood CA, Gensemer RW, Solomon
KR, Greenberg BM. 1997. Inhibition of photosynthesis as an
endpoint for the photoinduced toxicity of intact and photomodified PAHs. In Dwyer F, Doane T, Hinman M, eds, Environmental Toxicology and Risk Assessment: Modeling and Risk Assessment, Vol 6. STP 1317. American Society for Testing and
Materials, Philadelphia, PA, pp 443–455.
6. Schreiber U, Schliwa U, Bilber B. 1986. Continuous recording
of photochemical and non-photochemical chlorophyll fluorescence quenching with a new type of modulation fluorometer. Photosynth Res 10:51–62.
7. Long SP, Humphries S, Falkowski PG. 1994. Photoinhibition of
photosynthesis in nature. Ann Rev Plant Physiol Plant Mol Biol
45:633–662.
8. Anderson JM, Park Y-I, Chow WS. 1997. Photoinactivation and
photoprotection of photosystem II in nature. Physiol Plant 100:
214–223.
9. Genty B, Briantais J-M, Baker NR. 1989. The relationship between the quantum yield of photosynthetic electron transport and
quenching of chlorophyll fluorescence. Biochim Biophys Acta
990:87–92.
10. Oxborough K, Horton P. 1988. A study of the regulation and
function of energy-dependent quenching in pea chloroplasts.
Biochim Biophys Acta 934:135–143.
11. Huner NPA, Öquist G, Sarhan F. 1998. Energy balance and acclimation to light and cold. Trends Plant Sci 3:224–230.
12. Huang X-D, Krylov SN, Ren L, McConkey BJ, Dixon DG, Greenberg BM. 1997. Mechanistic quantitative structure–activity relationship model for the photoinduced toxicity of polycyclic aromatic hydrocarbons: II. An empirical model for the toxicity of
16 polycyclic aromatic hydrocarbons to the duckweed Lemna
gibba, L. G-3. Environ Toxicol Chem 16:2296–2303.
898
13. Marwood CA, Smith REH, Furgal JA, Charlton MN, Solomon
KR, Greenberg BM. 2000. Photoinhibition of natural phytoplankton assemblages in Lake Erie exposed to solar ultraviolet radiation. Can J Fish Aquat Sci 57:371–379.
14. Gensemer RW, Ren L, Day KE, Solomon KR, Greenberg BM.
1996. Fluorescence induction as a biomarker of creosote phototoxicity to the aquatic macrophyte Lemna gibba. In Bengtson D,
Henschel DS, eds, Environmental Toxicology and Risk Assessment: Biomarkers and Risk Assessment, Vol 5. STP1306. American Society for Testing and Materials, Philadelphia, PA, pp 163–
176.
15. Oris JT, Giesy JP. 1986. Photoinduced toxicity of anthracene to
juvenile bluegill sunfish (Lepomis macrochirus Rafinesque): Photoperiod effects and predictive hazard evaluation. Environ Toxicol
Chem 5:761–768.
16. Naes K, Oug E. 1998. The distribution and environmental relationships of polycyclic aromatic hydrocarbons (PAHs) in sediments from Norwegian smelter-affected fjords. Chemosphere 36:
561–576.
17. Huang X-D, Dixon DG, Greenberg BM. 1993. Impacts of UV
radiation and photomodification on the toxicity of PAHs to the
higher plant Lemna gibba (Duckweed). Environ Toxicol Chem
12:1067–1077.
18. Newsted JL, Giesy JP. 1987. Predictive models for photoinduced
acute toxicity of polycyclic aromatic hydrocarbons to Daphnia
magna Strauss (Cladocera, Crustacea). Environ Toxicol Chem 6:
445–461.
19. Oris JT, Giesy JP. 1987. The photoinduced toxicity of polycyclic
aromatic hydrocarbons to the larvae of the fathead minnow (Pimephales promelas). Chemosphere 16:1395–1404.
20. McConkey BJ, Duxbury CL, Dixon DG, Greenberg BM. 1997.
Toxicity of a PAH photo-oxidation product to the bacteria Photobacterium phosphoreum and the duckweed Lemna gibba: Effects of phenanthrene and its primary photoproduct, phenanthrenequinone. Environ Toxicol Chem 16:892–899.
21. Shiu WY, Maijanen A, Ng ALY, Mackay D. 1988. Preparation
of aqueous solutions of sparingly soluble organic substances: II.
Multicomponent systems—Hydrocarbon mixtures and petroleum
products. Environ Toxicol Chem 7:125–137.
22. Mueller JG, Chapman PJ, Pritchard PH. 1989. Creosote-contaminated sites: Their potential for bioremediation. Environ Sci Technol 23:1197–1201.
23. American Society for Testing and Materials. 1998. Standard guide
for conducting static toxicity tests with Lemna gibba G3. E 141591. In Annual Book of ASTM Standards, Vol 11.05. Philadelphia,
PA, pp 805–814.
24. Bestari KTJ, Robinson RD, Solomon KR, Steele TS, Day KE,
Sibley PK. 1998. Distribution and composition of polycyclic aromatic hydrocarbons within experimental microcosms treated
with liquid creosote. Environ Toxicol Chem 17:2359–2368.
25. Greenberg BM, Huang X-D, Dixon DG. 1992. Applications of
the aquatic higher plant Lemna gibba for ecotoxicological assessment. J Aquat Ecosys Health 1:147–155.
26. American Society for Testing and Materials. 1998. Standard guide
for conducting static, axenic, 14-day phytotoxicity tests in test
tubes with the submerged aquatic macrophyte, Myriophyllum si-
C.A. Marwood et al.
27.
28.
29.
30.
31.
32.
33.
34.
35.
36.
37.
38.
39.
biricum Komarov. E 1913-97. In Annual Book of ASTM Standards, Vol 11.05. Philadelphia, PA, pp 1428–1442.
Porra RJ, Thompson WA, Kriedemann PE. 1989. Determination
of accurate extinction coefficients and simultaneous equations for
assaying chlorophylls a and b extracted with four different solvents: Verification of the concentration of chlorophyll standards
by atomic absorption spectroscopy. Biochim Biophys Acta 975:
384–394.
Lazár D, Naus J. 1998. Statistical properties of chlorophyll fluorescence induction parameters. Photosynthetica 35:121–127.
Sanathanan L, Gade E, Shipkowitz N. 1987. Trimmed logit method for estimating the ED50 in quantal bioassay. Biometrics 43:
825–832.
Marwood CA, Smith REH, Solomon KR, Charlton MN, Greenberg BM. 1999. Intact and photomodified polycyclic aromatic
hydrocarbons inhibit photosynthesis in natural assemblages of
Lake Erie phytoplankton exposed to solar radiation. Ecotoxicol
Environ Saf 44:322–327.
Maxwell K, Badger MR, Osmond CB. 1998. A comparison of
CO2 and O2 exchange patterns and the relationship with chlorophyll fluorescence during photosynthesis in C3 and CAM plants.
Aust J Plant Physiol 25:45–52.
Oxborough K, Baker NR. 1997. Resolving chlorophyll a fluorescence images of photosynthetic efficiency into photochemical
and non-photochemical components—Calculation of qP and F9v/
Fm9 without measuring F9o. Photosyn Res 54:135–142.
Huang X-D, McConkey BJ, Babu TS, Greenberg BM. 1997.
Mechanisms of photoinduced toxicity of photomodified anthracene to plants: Inhibition of photosynthesis in the aquatic higher
plant Lemna gibba (Duckweed). Environ Toxicol Chem 16:1707–
1715.
Babu TS, Tripuranthakan S, Greenberg BM. 2000. Synergistic
toxicity of a mixture of 1,2-dihydroxyanthraquinone and Cu on
the aquatic plant L. gibba. In Greenberg BM, Hull R, Roberts M,
Gensemer R, eds, Environmental Toxicology and Risk Assessment: Science, Policy and Standardization—Implications for Environmental Decisions, Vol 10. STP 1403. American Society for
Testing and Materials, Philadelphia, PA, USA (in press).
Tripuranthakam S, Duxbury CL, Babu TS, Greenberg BM. 1999.
Development of a mitochondrial respiratory electron transport
bioindicator for assessment of aromatic hydrocarbon toxicity. In
Henshel D, Black M, Harrass M, eds, Environmental Toxicology
and Risk Assessment: Standardization of Biomarkers for Endocrine Disruption and Environmental Assessment, Vol 8. STP
1364. American Society for Testing and Materials, Philadelphia,
PA, pp 350–361.
Oettmeier W, Masson K, Donner A. 1988. Anthraquinone inhibitors of photosystem II electron transport. FEBS Lett 321:259–
262.
Behra R, Genoni GP, Joseph AL. 1999. Effect of atrazine on
growth, photosynthesis, and between-strain variability in Scenedesmus subspicatus (Chlorophyceae). Arch Environ Contam
Toxicol 37:36–41.
Hartman PE, Goldstein MA. 1989. Superoxide generation by photomediated redox cycling of anthraquinones. Environ Mol Mutagen 14:42–47.
Bailer AJ, Oris JT. 1998. Incorporating hormesis in the routine
testing of hazards. BELLE Newsletter 6:2–5.

chlorophyll fluorescence as a bioindicator of effects on growth in

Transkript

Podobné dokumenty

earproblems

Corina`s thesis: “The Carpet of the Sun”