Äîêóìåíò âçÿò èç êýøà ïîèñêîâîé ìàøèíû. Àäðåñ îðèãèíàëüíîãî äîêóìåíòà : http://library.biophys.msu.ru/matorin/3386.pdf Äàòà èçìåíåíèÿ: Fri Dec 19 12:31:26 2008 Äàòà èíäåêñèðîâàíèÿ: Mon Oct 1 19:49:07 2012 Êîäèðîâêà:

Arch. Hydrobiol.

153

4

685-70 1

Stuttgart, March 2002

High-temperature thermoluminescence of chlorophyll as a method to study lipid peroxidation in planktonic algae
Dmitrii V. Vavilin*, Dmitrii N. Matorin and Andrei B. Rubin
With 7 figures and 3 tables
1

Abstract: Chlorophyll-containing cells of algae, photosynthetic bacteria, and plants often exhibit strong red luminescence when heated to temperatures above 70 °C. This high-temperature thermoluminescence (HTL) originates mostly from thermally induced degradation of lipid cycloperoxides present in thylakoid membranes. The degrading cycloperoxides can transiently form carbonyl species in excited triplet state, whereas chlorophyll molecules readily accept energy from the triplet carbonyls and emit it as luminescence. The HTL emission was measured in phytoplankton collected from different waters and in laboratory cultures of algae and cyanobacteria. The intensity of HTL generally increased at mid-day in the phytoplankton sampled in oligotrophic waters (Lake Baikal, Kandalakshskaya Bay of the White Sea) but showed little changes throughout the day in waters with high concentration of mineral nutrients (Lake Geneva, Marseille Gulf of the Mediterranean Sea, Kotorska Gulf of the Adriatic Sea). When the culture of Chlorella pyrenoidosa was grown under the nitrogen- or phosphorus-limited conditions at moderate light intensity, an increase in the HTL was observed. However, the HT L emission remained low when C. pyrenoidosa or other species of algae were grown in nutrient-rich media even if the algae were exposed to a strong light. Consequently, nutrient availability may be of importance for cell resistance to the light-induced peroxidation of membrane lipids. The HT L is an easy and convenient method to access lipid peroxidation in natural phytoplankton assemblages.

Key words: Lipid cycloperoxides, nutrient deficiency, Chlorella pyrenoidosa.

1 Authors ' address : Departmen t of Biophysics , Faculty of Biology, Mosco w State University, 119899 Moscow, Russia. * Corresponding author. Present address and address for correspondence: Department of Plant Biology, Arizona State University, Box 871601, Tempe, AZ 85287-1601, USA. E-mail: vavilin@imap4.asu.edu

0003-9136/02/0153-0685 $ 4.25
© 2002 E. Schweizerbart'sche Verlagsbuchhandlung, D-70176 Stuttgart


68 6

Dmitrii V. Vavilin, Dmitrii N. Matorin and Andrei B. Rubin

Introduction
Light plays a key role in the performance of photosynthetic photochemical reactions in plants and algae. However, photosynthesis can be viewed as a very risky process because a large amount of oxygen is formed in thylakoids in the vicinity of powerful oxidation-reduction systems that readily reduce oxygen to dangerous superoxide and produce harmful singlet oxygen (DAL TO N 1995). Excess illumination may damage a variety of cellular constituents, especially thylakoid membranes. In addition to inactivation of Photosystem II and Photosystem I reactions in the thylakoids (AR O et al. 1993, SONOIKE 1995), prolonged exposure to a strong light can cause subsequent degradation of thylakoid membrane lipids, a process associated with lipid peroxidation (HEAT H & PACKER 1968, MISHR A & SINGHA L 1992). Th e lipid peroxidation is a universal mechanism of damage to membranes under various stress conditions (FOYE R et al. 1994). Hence, a continuous emphasis on exploring the susceptibility of photosynthetic organisms to photooxidative stress is needed in order to understand competitive relationships between photosynthetic species in ecosystems and effects of stress conditions on the overall productivity of water ecosystems (ALSCHE R & HES S 1993). Despite a great variety of literature concerning oxidative stress in plants, algae and cyanobacteria, little is known about peroxidation processes taking place in natural phytoplankton assemblages. The lipid peroxidation is most commonly detected and quantified by measuring by-products of the reaction, such as ethane, conjugated diens, malondialdehyde, and other carbonyl compounds (HEAT H & PACKER 1968, PRYOR & CASTL E 1984, CHATTERJE E &

1988, HAVAU X & NIYOG I 1999). Separation of specific products of lipid oxidation by HPLC was also employed in a number of studies (e.g. HAVAU X & NIYOG I 1999). However, these techniques cannot b e routinely used for the assessment of lipid oxidation in phytoplankton because the analysis requires high amounts of biological material. Th e second problem is the difficulty of separating the peroxidative reactions that take place in phytoplankton from the reactions occurring in non-phytoplankton paniculat e matter. When chlorophyll-containing material is gradually heated to temperatures above 70 °C, an enhancement of chlorophyll luminescence emission is often observed. This increase in the luminescence intensity at high temperature was named high-temperature thermoluminescence (HTL) (MATORI N et al. 1989, VENEDIKTO V et al. 1989). HTL peaking at 120-130 °C arises due to a thermally induced breakdown of lipid peroxides (presumably lipid cycloperoxides) that were present in the samples before heating. The high-temperature stimulated breakdown of lipid peroxides leads to the formation of carbonyl species in excited triplet state; chlorophyll molecules readily accept the energy of excited
AGARWA L


High-temperature thermoluminescence

68 7

carbonyls and emit it as luminescence upon transition to the ground state (MA TORIN et al. 1989, VAVILIN & DUCRUE T 1998).

1

Recently, we have shown a good correlation between the formation of malondialdehyde and intensity of the HTL emission at 120-130 °C in alga Chlorella pyrenoidosa exposed to a variety of conditions that provoked oxidative stress (VAVILIN et al. 1998). Measurements of the HT L have been used to detect lipid peroxidation in leaves of higher plants subjected to low-temperature stress (MATORI N et al. 1991), water stress (MERZLYA K et al. 1992), and strong light (HAVAUX & NIYOG I 1999). In the present paper, we demonstrate that lipid oxidation in natural assemblages of phytoplankton, as well as in laboratory planktonic photosynthetic organisms, can be conveniently assessed by measuring the HTL .

Materials and methods
The following strains of micoalgae were used in experiments: the chlorophyceae Chlorella pyrenoidosa CALU-175, Chlorella vulgaris SHL 108b and Selenastrum capricornutum CCAP 278/4; the diatoms Navicula accomoda SHL 107 and Phaeodactylum tricornutum; Synechocystis PCC 6803, a cyanobacterium. C. pyrenoidosa strain originated from the collection of the Biology Institute (St. Petersburg State University, Russia), S. capricornutum was purchased from the Culture Collection of Algae and Protozoa (Ambleside, UK), vulgaris and N. accomoda were isolated from Lake Geneva at the Station d'Hydrobiologie Lacustre (Thonon, France), and P. tricornutum was taken from the culture collection of the Institute of Marine Fisheries and Oceanography (Moscow, Russia). C. pyrenoidosa and Synechocystis sp. were grown at 30 °C, under continuos light of 50uErrT 2 s"1 provided by a set of fluorescent tubes (see Table 1 for the recipes of growth media). Other algae were grown at 22 °C, under indirect daylight (maximum light intensity of 120uEirr 2 s~') . For the nitrogen and phosphorus depletion experiments, cells of Chlorella pyrenoidosa in mid-logarithmic growth phase were harvested by centrifugation and resuspended in the modified BG-11 medium that was lacking NaNO3 or K2HPO4, respectively. In some experiments, 50 uM methylviologen (1,dimethyl-4,4'-bipiridinum dichloride) or 10 uM diuron (3-(3,4-dichlorophenyl)-l,ldimethylurea) was added to the growth media to stimulate lipid peroxidation in algae. Methylviologen is a herbicide that promotes generation of superoxide anion radicals in Photosystem I pigment protein complex, whereas diuron is a known inhibitor of electron transfer reactions in Photosystem II. HTL measurements were done essentially as described by VAVILIN & DUCRUET (1998) using the setup similar to that reported by MIRANDA & DUCRUET (1992). In brief, fixed volumes (up to 50 mL) of cell suspension were filtered through 15 mm polyamide filters. Filters with pore diameter of 1 were used for phytoplankton and cyanobacteria and of 5 )im for unicellular algae. The final pigment density varied from 50 ng to 2 jig of chlorophyll-a per square cm of the filter area. Immediately after the


68 8

Dmitrii V. Vavilin, Dmitrii N. Matorin and Andrei B. Rubin

Table 1. Culture media recipes. Mediu m BG-11
(RIPPK A et al.

Species C. pyrenoidosa, Synechocystis sp.

Composition NaNO 3 (1.5 g/L), MgSO 4 -7H 2 O (75mg/L), CaCl 2 · 2H 2 O (36mg/L), K 2 HPO 4 (31 mg/L), Na 2 CO 3 (20 mg/L), citric acid (6 mg/L), ferric ammonium citrate (6 mg/L), H3BO3 (2.86 mg/L), MnCl 2 · 4H 2 O (1.81 mg/L), ZnSO 4 · 7H 2 O (220 ug/ L), Na 2 MoO 4 · 2H 2 O (390ug/L). CuSO 4 · 5H 2 O (79ug/L), Co(NO 3 ) 2 -6H 2 O (49ug/L), NaEDT A (lOmg/L), 5m M TE S buffer (pH = 8.2) KNO3 (1.25 g/L), KH 2 PO 4 (200 mg/L), MgSO 4 -7H 2 O (1.00g/L), CaCl 2 -2H 2 O(100mg/L) , NaHCO 3 (40mg/L), H 3 BO 3 (114mg/mL), FeSO 4 · 7H 2 0 (50 mg/L), ZnSO 4 · 7H 2 O (88 mg/L), MnCb-4H 2 O (14.5mg/L), Na 2 MoO 4 · 2H 2 O (12mg/ L) , CuSO 4 -5H 2 O (16 mg/L), Co(NO 3 ) 2 · 6H 2 O (100ug/L), NaEDT A (50 mg/L) Medium A supplemented with Na 2 SiO 3 (25 mg/L) Mediu m supplemented with NaCl (27.0 g/L), MgSO 4 -7H 2 O (5.6g/L), MgCl 2 (5.5 g/L), CaCl 2 -2H 2 O(1.2g/L ) BG-11 medium; NaNO 3 was replaced by equimolar amount of NaHSO 4 BG-11 medium; K 2 HPO 4 was replaced b y KNO (36 mg/L)
3

1979)

C. vulgaris, S. capricornutum





N. accomoda P. tricornutum

Nitrogenless Phosphorusless

pyrenoidosa . pyrenoidosa

end of filtration, the filter was fixed in a pre-cooled aluminum sample holder equipped with a miniature electric heater, incubated for 5 min in darkness and then heated to 150 °C at a constant heating rate of 0.1 °C/s (heating at a faster rate shifted thermoluminescence peaks to higher temperatures). Thermoluminescence detected by a photomultiplier tube was recorded during the heating. A red longpass filter (X>650nm ) was placed between the sample and the photomultiplier to increase selectivity of the detection system with respect to chlorophyll luminescence (X MA X - 670-680nm) . Temperature was controlled using a thermocouple mounted between the filter and the heating element. In some cases, to decrease the time -period required for recording the thermoluminescence curves, the dark-adapted samples were first rapidly (5-1 0 s) heated to 75 °C and then the heating was continued at a slower rate (0.1 °C/s). This procedure did not change intensity of the thermoluminescence above 100 °C. It should be emphasized that all the samples were allowed to dry in air during the heating; when evaporation of water was retarded by covering the sample with a glass slide, a strongly modified HTL signal was observed (see DUCRUE T & VAVILIN 1999). Simulation of the measured HTL signal by a set of calculated thermoluminescence components was done by a numerical method described by MIRAND A & DUCRUE T (1992) and by VAVILIN & DUC RUE T (1998).


High-temperature thermoluminescence

68 9

Table 2. General characteristics of the sampling sites. Sampling site Global coordinates Sampling time June, 1987 July, 1988 Trophic status Oligotrophic
(KOZHOVA 1987)

52°N 105°E Lake Baikal (near the Biological Station of Irkutsk University, Bolshie Koty); Eastern Russia 68°N, 37°E Kandalakshskaya Bay (near the Biological Station of Moscow State University), White Sea Kotorska Gulf, Adriatic Sea 42°40'N , 18°40'E Marseille Gulf, North-Western 43° 20'N, Mediterranean Sea 4°50' E Lake Geneva (near Thonon); 46° 24'N, Western France/Switzerland 6°28' E

Oligotrophic to mesotrophic
(MAKSIMOVA 1978)

May, 1988

Highly eutrophic
(VILICIC 1989)

March, 1989 Eutrophic
(YORO et al. 1997)

August, 1996 Mesotrophic to eutrophic
(PELLETIER 1998)

To measure chlorophyll content, cells were harvested by filtration and the air-dried filters were kept frozen until analyzed. Pigments were extracted with small volume of 100 % methanol (cultures of algae) or with 100 % acetone (phytoplankton samples). Chlorophyll was determined using extinction coefficients referred in PORRA (1991). The sampling sites of phytoplankton are listed in Table 2.

Results and discussion
Thermoluminescence emission in different species of algae and cyanobacteria
A typical thermoluminescence curve recorded from the dark-adapted green alga Chlorella pyrenoidosa is characterized by a band peaking around 25 °C and by an unresolved band or a shoulder above 100 °C (Fig. 1 A, thin line). Incubation of Chlorella cells in the presence of methylviologen, a herbicide that stimulates oxidation of the membrane lipids, inhibited the 25 °C band and caused a significant rise in the amplitude of the HT L (Fig. 1 A, solid line). Similar enhancement of the luminescence at high temperatures and suppression of the low temperature luminescence (-2 5 °C) were observed in methylviologen-treated cultures of Synechocystis sp. PCC 6803 , Phaeodactylum tricornutum, and in phytoplankton sampled in a pond near the laboratory at CEN, Saclay (Fig. 1 B-D) . In Synechocystis and Phaeodactylum, the methylviologen treatment also stimulated thermoluminescence at about 60-7 5 °C (Fig. 1B.C) .


69 0

Dmitrii V. Vavilin, Dmitrii N. Matorin and Andrei B. Rubin

\iiIii 20 40 60 80 100 120 140

0

20 40 60

100 120 140

100 120 140

Temperature,

Temperature,

Fig. 1. Thermoluminescence curves measured in Chlorella pyrenoidosa (A), Synechocystis sp. PCC 6803 (B), Phaeodactylum tricornutum (C) cells, and in phytoplankton sampled in a pond on the territory of the Centre d'etude s Nucleaires (Saclay, France) early in the morning on August, 1998 (D). Thermoluminescence was recorded in nontreated cells (thin lines) and in the cells incubated for 5 h in the presence of methylviologen at light intensity of 40uEm~ 2 s~ 1 (thick lines) at 25 °C.

Thermoluminescence emitted by photosynthetic organisms in the temperature range from 0 to 150 °C has a heterogeneous origin. The 25 °C ban d (photosynthetic thermoluminescence) is produced by charge-recombination reactions in the Photosystem II pigment-protein complex (reviewed by VASS & GOVINDJE E 1996). These reactions were inhibited in algae subjected to methylviologen-induced oxidative stress. The mechanism of thermoluminescence emission intensified at 60-75 °C remains unclear. Some evidence suggests involvement of hydrogen peroxide in the generation of this band (DUCRUE T & VAVILIN 1999). Th e HTL emission at 90-14 0 °C (peak maximum around 125 °C), which is the focus of this study, arises due to the thermally induced breakdown of lipid cycloperoxides accompanied by the formation of chloro-


High-temperature thermoluminescence

691

phyll molecules in excited state (VENEDIKTO V et al. 1989, VAVILIN & Duc RUE T 1998).

Similar to the photo synthetic thermoluminescence (MIRAND A & DUCRUE T 1992), the HTL bands can be modeled assuming that formation of a band is driven by two competing factors: (i) rise of temperature causes an increase in the rate of decomposition of the peroxidized specie(s) and therefore an increas e in the luminescence intensity; (ii) diminution in the concentration of these species as decomposition proceeds leads to a decreas e in the luminescence (VAVILIN & DUCRUE T 1998). The first process dominates at relatively low temperatures, whereas the second process is responsible for the decay of thermoluminescence at higher temperatures. According to the Arrhenius-Eyring rate constant theory, the rate of decomposition of lipid peroxides can be written as dP / d t = P S T exp (-EA / k B T) (1),

where P represents the amount of lipid peroxides, which degradation leads to the luminescence emission, S is a frequency factor, EA is the activation energy, is the Bolzmann constant, and t is time . During the time period At, AP lipid peroxide molecules degrade at a given temperature T: = P (t, T) - P (t + At, T) (2).

Consequently, the amount of lipid peroxides in the sample by the end of the At period becomes equal to P (t + At, T) = P (t, T) - P (t, T) S T At exp (-EA / k B T) (3).

Starting at T = To and P - Po (P o equals the initial amount of lipid peroxides in a non-heated sample, which degradation at higher temperatures leads to the thermoluminescence emission) and increasing the temperature by At VT (V T is the heating rate) every time after calculating the corresponding P (t +At, T) point, the values of P (or AP) and T can be obtained at every time point t for the given parameters S and E A . Considering that thermoluminescence intensity is proportional to the rate of degradation of lipid peroxides AP/At, the plot of AP (or AP / At) versus T represents a theoretical thermoluminescence band, i n which P o corresponds t o the area under the thermoluminescence curve. I n reality, the measured thermoluminescence curves represent superposition of several thermoluminescence bands; each of these bands can be characterized by a set of the parameters S, E A , and PQ . By varying these parameters in order to find the best match between the theoretical and experimental thermoluminescence curves, the relative concentration of lipid peroxides in the sample (P o) can be estimated as illustrated in Fig. 2. Fig. 2 compares two HTL curves measured in Chlorella before (control) and after 4 days of incubation in the presence of photosynthesis inhibitor diuron. The inhibition of photosynthesis finally results in the culture death ac-


69 2

Dmitrii V. Vavilin, Dmitrii N. Matorin and Andrei B. Rubin

1.2x104-

8.0x103 -

4.0x103-

100

120

140

Temperature, °C
Fig. 2. Mathematical resolution of high-temperature thermoluminescence glow curves measured in Chlorella pyrenoidosa cells before (control) and after 4 days of incubation in the presence of 10 uM diuron. Shown are experimental data points (both samples had 1 (ig of chlorophyll-a per square cm of the filter area), individual thermoluminescence bands obtained from the mathematical curve resolution into two components (dotted line, control cells; dashed line, diuron-treated cells) and resulting theoretical thermoluminescence curves (solid lines). Decomposition parameters of the individual thermoluminescence bands are: control - Po (1) =7.6 x 104 photon counts, EA(1) = 16.6kcal/mol, So (1) = 2.3x 10V \ Po (2) = 2.7x 107 photon counts, EA(2) = 24.2kcal/ mol, So (2) = 1.4x 105 s"1; diuron-treated cells - Po (1) = 3.5 x 105 photon counts, EA(1) = 16.8 kcal/mol, So (1) = 3.0xl0 4 s- \ Po (2) = 3.0 x 107 photon counts, EA(2) = 24.0kcal/mol, So (2) = 1.2 x 105 s"1.

companied by light-stimulated oxidative degradation of lipid membranes. Both HTL curves can be satisfactory modeled by a sum of two bands, peaking at about 125 °C and above 150 °C. As can be seen from comparison of the integral intensities of the individual bands (see the legend to Fig. 2 for Po values), lipid peroxidation stimulated the 125 °C component, whereas the component peaking at higher temperature exhibited substantially less pronounced changes. According to our previous results (VAVILI N & DUCRUE T 1998, Du CRUET & VAVILI N 1999), the band peaking above 150 °C is not related to the preaccumulated lipid peroxides but likely is caused by thermal oxidation of other cellular components. It should be also noted that in certain cases introduction of a third component peaking at 60-7 5 °C (see for example Fig. 1C) is required to obtain a good agreement between theoretical and experimental thermoluminescence curves in the range of temperatures from 70 to 150 °C.


High-temperature thermoluminescence

69 3

3x10 -

2x10 -

1x10 -

0.0

3.0x10J

6.0x10J

9.0x10J

1.2x104

HTL intensity at 125 °C
Fig.3. Relationship between the integral intensity of the 125 °C HTL band (Po(l)) and the amplitude of the HTL measured at 125 °C. TL curves were recorded from Chlorella pyrenoidosa cells incubated with the diuron for different periods of time (up to 10 days). The above-described procedure of decomposition of the thermoluminescence curves into separate components, which involves adjustment of the three parameters (P o , a n d S ) for every band, i s rather time-consuming. However, the band peaking above 150 °C generally has a minor contribution to the 125 °C emission, therefore measurements of the amplitude of the HTL at 125 °C can be used as a reasonably good indicator of the lipid peroxides content in the samples (Fig. 3). This simplified approach was employed for the detection of lipid peroxides in phytoplankton sampled in different waters. The high-temperature thermoluminescence in phytoplankton Fig. 4 shows typical diurnal changes in the HT L intensity of phytoplankton from Lake Baikal sampled from different depths. The phytoplankton community was dominated by Chroomonas acuta (Cryptophyta), Koliella longiseta (Chlorophyta), and Nitzschia acicularis (Bacillariophyta). The measuring period was preceded by a stormy weather, which caused almost uniform distribution of the phytoplankton biomass throughout the water column with a chlorophyll concentration of about 0.6 ug/L. The HTL signal increased strongly during the first half of the day in the samples collected near the surface of the lake, whereas in deeper water layers the HTL exhibited minor


69 4

Dmitrii V. Vavilin, Dmitrii N. Matorin and Andrei B. Rubin

1.54-

1.04-

/

\

--·-- --·--12 -+- 25

5.0x103-

\*

*--

·

--

·

0.0-

12

16

20

24

Time, h
Fig. 4. Diurnal dynamics of the amplitude of HTL measured at 125 °C in phytoplankton sampled in the Lake Baikal at the depth of 0, 12 and 25 m. Samples contained about 50 ng of chlorophyll-a per cm2 of the filter area. Each data point represents an average of two to three measurements. S.D. was less than 15 % of the average values.

changes. The phytoplankton biomass varied within about 15 % throughout the whole day at different depths. Therefore, the observed changes in the HTL emission are likely to reflect intensification of lipid peroxidation upon exposure of surface phytoplankton to strong daylight. It should be emphasized that strong HT L was also characteristic for the phytoplankton probes collected in mid-day time at other sites of the lake. Similar effect of light on the diurnal dynamics of the HTL emission was observed in phytoplankton sampled in Kandalakshskaya Bay of the White Sea on the North-East of Russia (Fig. 5) . Th e diatom Skeletonema costatum overwhelmingly dominated in the samples; the diatoms Chaetoceros decipiens, Ch. curvisetus, and Ch. diadema were also abundant. In this experiment, the phytoplankton collected at midnight from approximately 0.5 m depth was placed into two 3-L trace-metal clean glass containers in order to prevent effects of vertical phytoplankton mixing throughout the day. One of the containers was screened from sunlight by black cloth and both containers were left floating on the surface of the water. The HTL signal increased sharply in phytoplankton exposed to direct sunlight, whereas no increase in the HTL emission was observed in the darkened container. Contrary to the results obtained in Lake Baikal and in the White Sea, no light-stimulated mid-day rise in the intensity of the HTL was detected in


High-temperature thermoluminescence

69 5

9.0x10

16


6.0x103-

3.0x10-

6

8

10

12

14

16

18

Time, h
Fig. 5. Diurnal dynamics of the amplitude of thermoluminescence signal measured at 125 °C in phytoplankton incubated under full sunlight (·) and in the dark (·). Phytoplankton was sampled at the Kandalakshskaya Bay of the White Sea. Filters prepared for the HTL measurements contained about 200 ng of chlorophyll-a per cm2 of the filter area. Each data point represents an average of two to three measurements. SD was less than 11 % of the average values.

phytoplankton sampled in the Kotorska Gulf of the Adriatic Sea, in the Marseille Gulf of the Mediterranean Sea, and in Lake Geneva. For example, intensity of the HT L in the Marseille Gulf samples varied within the range of ± 14 % of an average value during the 40 h observation period that started at 6:0 0 in the morning (samples from the sea surface were taken every 1.54.0 h) , with the highest values detected before the sunrise on the second day of the experiment. Average intensity of the HT L from phytoplankton sampled in Lak e Geneva between 11:0 0 and 13:0 0 was only 1.05 ± 0.15 times higher compared to the intensity of HT L in the samples collected in the morning at 8 :3 0 or in the evening at 18 :0 0 (the samples were taken for 6 days and the signal intensity measured in mid-day was normalized to the value obtained in the morning or in the evening at the same day). It should be pointed out that all experiments with phytoplankton were done on cloudless days when maximum subsurface light intensities at noon reached values above lOOOuEm^s" 1 . Therefore, low light intensity due to a cloudy sky, for example, cannot account for the absence of diurnal dynamics of the HTL in the above-mentioned regions. Interestingly, when different species of algae and cyanobacteria grown


69 6

Dmitrii V. Vavilin, Dmitrii N. Matorin and Andrei B. Rubin

Table ferent nated ments

3. Effect of strong light on the amplitude of the HTL measured at 125 °C in difcultures of algae and cyanobacteria. The amplitude of the HTL in non-illumicells was taken as 100%. Shown are the average values +S D of 3 to 4 measure. Light intensity, uEirT 2 s"1 1000 1000 1000 1000 1500 Duration of light treatment, h
4.0 4. 0 2.0

Species C. pyrenoidosa C. vulgaris S. capricornutum N. accomoda Synechocystis sp. PCC 6803

HTL intensity 11 + 12 98±1 0 97 ± 6 93±1 0 105+ 8 100±13

3.5 2.0 3.5

in the laboratory under optimum conditions were exposed to strong light for up to 4 hours, the HTL intensity remained low as well (Table 3). Besides the light intensity, numerous factors or a combination of factors can affect the physiological state of phytoplankton in natural environment and thereby modulate the population response to high light stress. Thus, nutrient deficiency can increase production of potentially toxic reactive oxygen species such as hydrogen peroxide, superoxide anion radicals, or hydroxyl radicals in plants (FOYE R et al. 1994). In turn, these highly reactive forms of oxygen can cause peroxidation of the membrane lipids. We propose that combination of strong light and nutrient limitation might be responsible for the mid-day stimulation of lipid peroxidation in phytoplankton. This trend can be clearly seen when trophic status of different phytoplankton sampling sites (see Table 2) is compared to the results of the HT L measurements. Indeed, Lake Baikal is characterized by extremely low concentration of dissolved mineral nutrients (KOZHOV A 1987). Productivity and biomass of phytoplankton in the Whit e Sea are usually low in mid-summer period primarily because of nitrogen limitation (MAKSIMOVA 1978). Accordingly, phytoplankton sampled in these two regions in the middle of the day was characterized by increased HTL (however, we cannot exclude the possibility that bottle artifacts contributed to some extent to the HT L rise observed in the light-exposed bottle in the White Sea). In contrast, waters of Lake Geneva, Kotorska Gulf and the coastal region of the Me diterranean Sea are generally enriched in mineral nutrients and no mid-day enhancement of the HTL emission was detected at these sampling sites.

Effect of macronutrients limitation on the high-temperature thermoluminescence in Chlorella
In order to test the effect of mineral deprivation on lipid peroxidation under controlled conditions, the culture of Chlorella pyrenoidosa was grown for two


High-temperature thermoluminescence

697

20

30

40

50

Time, h
Fig. 6. Amplitude of the HTL signal measured at 125 °C in Chlorella pyrenoidosa cells incubated in the nitrogen-deficient (· ) and in the nitrogen-rich media (V) at 100 uE m ~2 s"1 and in the nitrogen-deficient medium in darkness (·) . Different volumes of the culture were harvested by filtration to achieve equal amounts of chlorophyll-a (1 ng/cm 2) for the HTL measurements. Shown are the results of a single experiment and SD values obtained in three independent experiments in the cultures grown for 24 h under nitrogen-deficient conditions.

days in the absence of mineral nitrogen. In this experiment, initial concentration of NO3~ ions was less than 0.01 mM , compared to 1.8 mM in the full growth medium. Under these conditions, early symptoms of nitrogen deficiency (e.g. decrease in photosynthetic activity) were observed within about 4 h following the nitrogen removal (VAVILI N et al. 1999). Besides the inhibition of photosynthesis, the nitrogen removal also caused a continuous rise in the intensity of HTL in the culture exposed to light, whereas no HTL rise was observed in the dark-incubated culture (Fig. 6). Similar results were obtained with Chlorella cells grown at strongly reduced concentration of mineral phosphorus. However, in the latter case the increase in HT L was observed 5 to 7 days after the phosphorus was removed from the medium (Fig. 7).

Concluding remarks
The HTL measurements hold substantial potential to access the importance of peroxidative reactions in natural phytoplankton assemblages. High sensitivity of photomultiplier tubes used to detect the luminescence allows reliable re-


69 8

Dmitrii V. Vavilin, Dmitrii N. Matorin and Andrei B. Rubin

I
to

6.0x10 -

4.0x103-

2.0x103-

Fig. 7. Amplitude of the thermoluminescence signal measured at 125 °C in Chlorella pyrenoidosa cells incubated in the phosphate-deficient medium at 100 uE irT2 s"' () and in darkness (·). Different volumes of the culture were harvested by filtration to achieve equal amounts of chlorophyll-a (1 |ig/cm ) for the HTL measurements. On the 3 day of experiment, the culture was diluted 50 times with fresh BG-11 medium containing no phosphate. Shown are the results of a single experiment. The rate of HTL increase in the phosphate-deficient Chlorella culture exposed to light varied to some extent from experiment to experiment. However, in all experiments the amplitude of the HTL increased at least 3 times during the 7-day period of light incubation in the absence of phosphate in the medium.

cording of the HTL signals in samples collected from highly oligotrophic waters with chlorophyll concentration of about 0.1ug/l. We believe that HTL peaking at -12 5 °C mostly reflects lipid peroxidation reactions occurring in the phytoplankton material. Really, the HTL is emitted by chlorophyll molecules in excited state, which are formed upon temperaturestimulated degradation of the lipid peroxides (VENEDIKTO V et al. 1989, VAVILIN & DUCRUE T 1998). Consequently, selective detection of thermoluminescence in the red region of the spectra corresponding to the wavelengths of maximum chlorophyll emission (670-68 0 nm) provides a good separation of the luminescence originating from chlorophyll-containing photosynthetic organisms from chemiluminescence emitted by chlorophyll-less paniculat e matter, which is generally blue-shifted compared to the chlorophyll luminescence. Moreover, results of the modeling of the thermoluminescence curves (see Fig. 2) suggest that only reactions with certain thermodynamic characteristics (i.e. S and EA values) may lead to the HT L emission at 90-14 0 °C.


High-temperature thermoluminescence

69 9

Th e intensity of lipid peroxidation can be assessed from measurements of the integral intensity of the 125 °C band (Po ) or from measurements of the amplitude of HTL around 125 °C. The latter approach is the easiest; however, it should be taken into account that under certain conditions the 125 °C emission can be significantly overestimated because of the contribution of thermoluminescence peaking above 150 °C. The results of the experiments presented in this paper demonstrate significant variations in the intensity of lipid peroxidation reactions in phytoplankton. These variations were modulated by diurnal changes in light intensity. However, comparison of the diurnal dynamics of the HT L in the phytoplankton sampled at different places, as well as the experiments done with the cultures of algae, show that light intensity is essential but not the only factor responsible for the increase in the lipid peroxides content in the cells. Preliminary results give evidence for the importance of nutrient availability in cell resistance to the light-induced oxidative damage. Acknowledgements This research was supported in part by a research fellowship from the European Science Foundation (Program "Toxicology and Environmental Toxicology") awarded to D.V.V. We thank Dr. J.-M. DUCRUET for the use of his thermoluminescence setup in some experiments, for stimulating discussions, and for hospitality during DV's stay at the Centre d'Etudes Nucleaires (Saclay, France). We also express our sincere thanks to Dr. A. E L JAY and Dr. J. P. PELLETIER (Station d'Hydrobiologie Lacustre, Thonon, France), Dr. L. R. IZMESTIEVA (Irkutsk State University, Russia) for their valuable help at different stages of this study. References ALSCHER, R. G. & HESS, J. L. (1993): Antioxidants in Higher Plants. - CRC Press Inc., Boca Raton. ARO, E.-M., VIRGIN, I. & ANDERSSON, B. (1993): Photoinhibition of Photosystem II. Inactivation, protein damage and turnover. - Biochim. Biophys. Acta. 1143: 113 -- 134. CHATTERJEE, S. N. & AGARWAL, S. (1988): Liposomes as membrane model for study of lipid peroxidation. - Free Radical Biol. Med. 4: 51-72. DALTON, D. A. (1995): Antioxidant Defenses of Plants and Fungi. - In: AHMAD, S. (ed.): Oxidative Stress and Antioxidant Defences in Biology. - Chapman & Hall, N.Y., pp. 298-355. DUCRUET, J.-M. & VAVILIN, D. V. (1999): Chlorophyll high-temperature thermoluminescence emission as an indicator of oxidative stress: Perturbating effects of oxygen and leaf water content. - Free Radical Res. 31: S187-S192.
FOYER, R., LELANDAIS, M. & KUNERT, K. J. (1994): Photooxidative stress in plants.

- Physiol. Plantarum 92: 696-717.


700

Dmitrii V. Vavilin, Dmitrii N. Matorin and Andrei B. Rubin

HAVAUX, M. & NIYOGI, . . (1999): The violaxanthin cycle protects plants from photooxidative damage by more than one mechanism. - Proc. Natl. Acad. Sci. USA 96: 8762-8767. HEATH, R. L. & PACKER, L. (1968): Photoperoxidation in isolated chloroplasts. I. Kinetics and stoichiometry of fatty acid peroxidation. - Arch. Biochem. Biophys. 125: 189-198. KOZHOVA, . . (1987): Phytoplankton of lake Bakal: Structural and functional characteristics. -Arch. Hydrobiol., Beih. Ergebn. Limnol. 25: 19-37. MAKSIMOVA, M. P. (1978): Concentration of nutrients and balance of nitrogen, phosphorus, and silicon in the White Sea. - Oceanologia (Moscow). 18: 58-63 .
MATORIN, D. N., VAVILIN, D. V., KAFAROV, R. S. & VENEDIKTOV, P. S. (1989): High-

temperature thermoluminescence of chlorophyll as a method for study of lipid peroxidation in plants. - Doklady Akademi Nauk SSSR. 309: 764-768. - - - - (1991): Effect of freezing-thawing on lipid peroxidation in chlorophyllcontaining plant tissues. - Sov. Plant Physiol. 38: 401-406.
MERZLYAK, M. N., PAVLOV, V. K. & ZHIGALOVA, T. V. (1992): Effect of desiccation

on chlorophyll high-temperature chemiluminescence in Acer platanoides L. and Aesculus hippocastanum L. leaves. - J. Plant. Physiol. 139: 629-631. MIRANDA, T. & DUCRUET, J.-M. (1992): Graphical and numerical analysis of the thermoluminescence and fluorescence Fo emission in photosynthetic material. - Photosynth. Res. 24: 15-27. MISHRA, R. K. & SINGHAL, G. S. (1993): Photosynthetic activity and peroxidation of thylakoid lipids during photoinhibition and high-temperature treatment of isolated wheat chloroplasts. - J. Plant Physiol. 141: 286-292. PELLETIER, J. P. (1998): Phytoplankton as the basis of the trophic web in Geneva Lake. - Archives des Sciences 51: 37-44. PORRA, R. J. (1991): Recent advances and re-assessment in chlorophyll extraction and assay procedures for terrestrial, aquatic, and marine organisms, including recalcitrant algae. - In: SCHEER, H. (ed.): Chlorophylls. - CRC Press, Boca Raton, pp. 31-58. PRYOR, W. A. & CASTLE, L. (1984): Chemical methods for the detection of lipid hydroperoxides. - Methods Enzymol. 105: 293-305.
RIPPKA, R., DERUELLES, J., WATERBURY, J. ., HERDMAN, M. & STAINER, R. Y.

(1979): Generic assignments, strain histories and properties of pure cultures of cyanobacteria. - J. Gen. Microbiol. 24: 1--61. SONOIKE, K. (1995): Selective photoinhibition of photosystem I in isolated thylakoid membranes from cucumber and spinach. - Plant Cell Physiol. 36: 825-830. VASS, I. & GOVINDJEE (1996): Thermoluminescence from the photosynthetic apparatus. - Photosynth. Res. 48: 117-126. VAVILIN, D. V. & DUCRUET, J.-M. (1998): The origin of 115-130 °C thermoluminescence bands in chlorophyll-containing material. - Photochem. Photobiol. 68: 191198.
VAVILIN, D. V, DUCRUET, J.-M., MATORIN, D. N., VENEDIKTOV, P. S. & RUBIN, A. B.

(1998): Membrane lipid peroxidation, cell viability and photosystem II activity in the green alga Chlorella pyrenoidosa subjected to various stress conditions. - J. Photochem. Photobiol. B. Biol. 42: 233-239.


High-temperature thermoluminescence

701

VAVILIN , D . V., MATORIN , D . N., VENEDIKTOV , P. S. & RUBIN , A. B . (1999): Two types

of inactive photosystem II centers in green alga Chlorella grown under conditions of nitrogen deficiency. - Russ. J. Plant Physiol. 46 : 592-597 .
VENEDIKTOV , P. S., MATORIN , D. N. & KAFAROV , R. S. (1989): Chemiluminescence of

chlorophyll upon lipid photoperoxidation in thylakoid membranes. - Biofizika (Moscow) 34 : 241-245 . VILICIC , D. (1989): Phytoplankton population density and volume as indicators of eutrophication in the eastern part of the Adriatic Sea. - Hydrobiologia 174: 117132.
YORO, S . C , SEMPERE , R., TURLEY , C , UNANUE , M . A., D E MADRON , X . D . &

BIANCHI , M. (1997): Cross-slope variations of organic carbon and bacteria in the Gulf of Lions in relation to water dynamics (northwestern Mediterranean). - Mar. Ecol. Progr. Ser. 161: 255-264 . Submitted: 15 July 2000; accepted: 26 October 2001.