APPLIED AND ENVIRONMENTAL MICROBIOLOGY, Nov. 1997, p. 41784184 0099-2240/97/$04.

000 Copyright 1997, American Society for Microbiology

Vol. 63, No. 11

Inhibitory Effect of Solar Radiation on Thymidine and Leucine Incorporation by Freshwater and Marine Bacterioplankton
RUBEN SOMMARUGA,1* INGRID OBERNOSTERER,2 GERHARD J. HERNDL,2 1 AND ROLAND PSENNER Institute of Zoology and Limnology, University of Innsbruck, 6020 Innsbruck,1 and Department of Marine Biology, Institute of Zoology, University of Vienna, 1090 Vienna,2 Austria
Received 8 May 1997/Accepted 18 August 1997

We studied the effect of solar radiation on the incorporation of [3H]thymidine ([3H]TdR) and [14C]leucine ([14C]Leu) by bacterioplankton in a high mountain lake and the northern Adriatic Sea. After short-term exposure (3 to 4 h) of natural bacterial assemblages to sunlight just beneath the surface, the rates of incorporation of [3H]TdR and [14C]Leu were reduced at both sites by up to 70% compared to those for the dark control. Within the solar UV radiation (290 to 400 nm), the inhibition was caused exclusively by UV-A radiation (320 to 400 nm). However, photosynthetically active radiation (PAR) (400 to 700 nm) contributed almost equally to this effect. Experiments with samples from the high mountain lake showed that at a depth of 2.5 m, the inhibition was caused almost exclusively by UV-A radiation. At a depth of 8.5 m, where chlorophyll a concentrations were higher than those in the upper water column, the rates of incorporation of [3H]TdR were higher in those samples exposed to full sunlight or to UV-A plus PAR than in the dark control. In laboratory experiments with articial UV light, the incorporation of [3H]TdR and [14C]Leu by mixed bacterial lake cultures was also inhibited mainly by UV-A. In contrast, in the presence of the green alga Chlamydomonas geitleri at a chlorophyll a concentration of 2.5 g liter1, inhibition by UV radiation was signicantly reduced. These results suggest that there may be complex interactions among UV radiation, heterotrophic bacteria, and phytoplankton and their release of extracellular organic carbon. Our ndings indicate that the wavelengths which caused the strongest inhibition of TdR and Leu incorporation by bacterioplankton in the water column were in the UV-A range. However, it may be premature to extrapolate this effect to estimates of bacterial production before more precise information on how solar radiation affects the transport of TdR and Leu into the cell is obtained. Within aquatic ecosystems, pelagic bacteria constitute an important resource in the food web available to protists and metazoans (7). They are also responsible for cycling up to 50% or more of the primary production (3), thus playing a pivotal role in the global cycling of carbon. In contrast to the case for phytoplankton, on which most of the research effort has been concentrated, our understanding of the effects of solar UV radiation (290 to 400 nm) on natural bacterial assemblages and their activities in aquatic systems is limited. Within the plankton, however, bacteria are likely to be among the most UVsensitive organisms, because due to their small size, they may be inefcient in internal self-shading by absorbing cell matter (11). For example, damage to DNA due to solar radiation was found to be two times higher in bacterioplankton than in eukaryotic plankton (20). Also, UV-absorbing compounds, such as mycosporine-like amino acids that confer certain protection to eukaryotic organisms and cyanobacteria, are not known to exist in heterotrophic bacteria (22). However, the role played by carotenoid pigments present in certain aquatic bacterial species has not been investigated in detail, although they are known to protect bacteria against oxidative damage by solar radiation (28). Most literature about the damaging effects of UV radiation on bacteria is based on studies with bacterial isolates and articial radiation, particularly UV-C (200 to 280 nm), which
* Corresponding author. Mailing address: University of Innsbruck, Institute of Zoology and Limnology, Technikerstr. 25, 6020 Innsbruck, Austria. E-mail: Ruben.Sommaruga@uibk.ac.at. Present address: Netherlands Institute for Sea Research, Department of Biological Oceanography, Texel, The Netherlands. 4178

does not reach the Earths surface. Solar radiation is known to affect several bacterially mediated processes in aquatic systems, such as ammonia oxidation (17) or nitrication (18). However, with some exceptions (12), in most studies the contributions to photoinactivation by different wavelengths or regions of the sunlight spectrum were not considered. Knowledge of whether ambient levels of solar UV-B radiation (290 to 320 nm) have a greater impact than UV-A radiation (320 to 400 nm) on aquatic bacteria and other organisms is crucial to understanding and evaluating possible consequences of stratospheric ozone depletion, as only the UV-B range of the solar spectrum is signicantly affected, while changes in the UV-A range are negligible. Although on a photon basis, UV-A radiation contains less energy than UV-B radiation, the greater fraction of solar radiation in the UV-A spectrum than in the UV-B spectrum may result in a signicant source of biological damage. Recent studies have shown that solar UV-A radiation can damage aquatic viruses (41) and heterotrophic agellates (39) and cause growth inhibition in benthic diatoms (5). Furthermore, it is a major cause of inhibition of photosynthesis rates in phytoplankton (6, 8, 24), and model calculations suggest that it dominates total UV photoinhibition regardless of ozone concentrations (2). However, information on the effects of different regions of the solar spectrum on bacterioplankton is not conclusive. Sieracki and Sieburth (36) found that growth delay in marine bacterioplankton was caused mainly by UV-A radiation. The survival of bacterioplankton from Antarctic waters, estimated by changes in viable counts, was more affected by UV-A than by UV-B radiation (15). Photosynthetically active radiation (PAR) (400 to 700 nm) had little effect on sur-

VOL. 63, 1997

EFFECT OF SOLAR RADIATION ON BACTERIOPLANKTON

4179

vival of natural bacterial assemblages but had a strong effect on two bacterial isolates from the same waters. In a tropical lake, bacterial production was reduced even at depths well below the 1% level of UV-B radiation. Although the inhibition by different wavelengths was not assessed, UV-A and PAR were assumed to be responsible for this effect (27). On the other hand, Aas et al. (1) in a study on the estuarine waters of the Santa Rosa Sound, Fla., found that UV-B and UV-A inhibited the incorporation of radiolabeled thymidine (TdR) by similar percentages and that PAR also had a signicant inhibitory effect. In contrast to the results for TdR, incorporation of leucine (Leu) was inhibited mainly by UV-B, and PAR resulted in a stimulation compared to results for the dark control. In the same study, experiments performed at the Gulf of Mexico failed to demonstrate a consistent pattern in differences due to different wavelengths. For the Adriatic Sea, Herndl et al. (16) indicated that solar UV-B radiation reduced bacterial growth and enzymatic activities, although the relative contributions of UV-A and PAR were not studied explicitly. Here we present results from experiments done at one alpine lake and the Adriatic Sea which were designed to evaluate the effects of solar UV-B, UV-A, and PAR on the incorporation by bacterioplankton of radiolabeled TdR and Leu. Furthermore, in laboratory experiments with articial UV radiation and PAR, we tested whether the responses of bacteria to the different range of wavelengths may be different in the absence of phytoplankton. Measurements of TdR and Leu incorporation provide a rapid and easy assessment of the bacterial response to environmental factors and are now widely used to estimate bacterial production in ecological studies.
MATERIALS AND METHODS Study sites and sampling. Experiments were conducted during summer 1996 at Gossenko llesee, situated above the treeline (2,417 m above sea level) in the Central Alps, Tyrol, Austria (4713N, 1101E) and in the northern Adriatic Sea (4505N, 1330E). Ice cover in Gossenko llesee (area, 1.7 ha; maximum depth, 9.9 m) commonly lasts from November through mid-July. Due to the altitude effect (4), high mountain lakes receive considerably greater instantaneous uxes of UV-B than UV-A radiation compared to aquatic systems situated at lower elevations. Surface water samples were collected from both sites with an opaque sampler, dispensed into carboys under dim light conditions, and processed immediately. Additionally, we took samples in the lake at 2.5 and 8.5 m, corresponding to approximately 50 and 10% of the surface UV-B irradiance (305 nm) (see below). Samples for the determination of chlorophyll a concentrations were also collected at the same depths. A station on the shore of the lake and the Center for Marine Research, Ruder Boskovic, Rovinj, Croatia, provided facilities for the experimental work. Field experiments. In all experiments quartz asks (250 ml) with Teon caps were used for solar radiation exposures. Experiments were done on cloudless days, and exposure lasted for 3 h (Gossenko llesee, 22 July) or 4 h (all other experiments) over the local midday. The quartz asks were lled with lake water or seawater and incubated in situ with the asks held horizontally at the same depth of sampling. The following treatments were used: (i) UV-B plus UV-A plus PAR (i.e., full sunlight); (ii) UV-A plus PAR, using Mylar D foil (50% transmittance at 320 nm) to exclude the UV-B; (iii) PAR, using a vinyl chloride foil (CI Kasei Co., Tokyo, Japan; 50% transmittance at 405 nm) to exclude all UV radiation; and (iv) dark (i.e., asks wrapped with three layers of aluminum foil). The dark asks were exposed apart from the others to avoid possible reection. Changes in the transmittances of the different foils were tested repeatedly in a double-beam spectrophotometer, and foils were replaced if necessary. Laboratory experiments. Experiments with articial UV radiation were done at Innsbruck, Austria, inside a walk-in chamber (15 1C) with the same treatments as described above. In this case, open crystallization dishes (9.5-cm diameter, 5-cm height) were used as containers. These experiments were designed to test the effect of UV radiation on bacteria exposed in the absence or presence of microalgae. For this purpose we exposed either a mixed bacterial culture or the same bacterial culture together with the green alga Chlamydomonas geitleri for 4 h. The mixed bacterial culture was obtained by inoculating 100 ml of ltered (0.8-m-pore-size lter) lake water from Gossenko llesee into 1 liter of sterile lake water (ltered through a 0.2-m-pore-size lter and autoclaved) containing two barley seeds to provide a source of nutrients. Bacteria were grown in the dark at 15 1C to a density of 1.5 106 cells ml1 as

determined by epiuorescence microscopy counts with DAPI (4,6-diamidino-2phenylindole) to stain bacterial cells (33). C. geitleri was obtained from the algal collection of the Botanical Institute, University of Innsbruck, and grown in a modied (distilled water instead of seawater) replete f/2 medium (13) under continuous articial PAR at the same intensity (0.01 microeinstein cm2 s1) used during the experiments with articial UV radiation. PAR was provided by two white uorescent tubes of 36 W. In the experiments with C. geitleri, the culture was concentrated by ltration at low pressure onto a 3-m-pore-size lter, washed several times with sterile f/2 medium, mixed with the bacterial culture, and then dispensed into containers for the different treatments. No cell rupture was observed with this procedure before the exposure. The concentration of chlorophyll a was adjusted to 2.5 g liter1. The transmittance of the medium was measured at 310 nm in a double-beam spectrophotometer with a quartz cuvette with a 2-cm path length against a reference cell lled with ltered (0.2-m-pore-size lter) medium. Articial UV-B sources. Articial UV-B irradiation was provided by a set of four tubes of UV-A-340 (Q-Panel Co., Cleveland, Ohio). These lamps have a maximum emission at 340 nm and produce no radiation below 280 nm. The integrated irradiance values between 280 and 320 nm measured at 25 cm from the lamps was 1.4 W m2, corresponding to a biological dose rate (weighted by Setlow DNA action spectrum normalized to 1 at 300 nm) of 0.091 W m2. The spectrum emitted by the lamps had a lower UV-A/UV-B ratio (14) than the natural solar spectrum, which, for example, was 19 at Innsbruck in June. In all experiments, background visible light was additionally supplied by white uorescent tubes as described above. The UV lamps were preburned for 600 h before the spectrum was measured, and experiments were conducted within the next 100 h. Before each experiment, the lamps were switched on for 2 h to reach a constant emission spectrum. Because of the heat produced by the lamps, a water bath was used to keep the temperature inside the containers within 1C. Measurements of Leu and TdR incorporation. We used the dual-labeling method (37) with some modications described below. [3H]TdR (84 Ci mmol1) (Amersham, Amersham, England) and [14C]Leu (310 mCi mmol1) (Amersham) were added at saturating concentrations (10 and 20 nM, respectively) to triplicate subsamples from the different treatments and incubated at in situ temperatures and in darkness for 1 h. Preliminary experiments showed that the uptake of TdR and Leu was linear over this period. Incubations were nished by the addition of formaldehyde (3% nal concentration). Bacteria were then collected by ltering samples through 0.2-m-pore-size cellulose nitrate lters (Sartorius, Goettingen, Germany). Filters were prewashed 10 times with 1 ml of ice-cold 5% (wt/vol) trichloroacetic acid (TCA), extracted for 10 min with 5 ml of ice-cold TCA, and nally washed 10 times with 1 ml of TCA and two times more with 5 ml of ice-cold 80% (vol/vol) ethanol. The metallic ltration funnel was removed, and the borders of the lter were washed with TCA. The dry lters were placed in scintillation vials with 10 ml of scintillation cocktail (Ready Safe; Beckman, Fullerton, Calif.). Disintegrations per minute were determined in a Beckman LS 5000TD scintillation counter after the lters had dissolved completely in the scintillant. All samples were corrected for abiotic incorporation by subtracting the radioactivity in two formaldehyde-killed (3% nal concentration) controls. UV measurements. The spectrum of the lamps was measured at high resolution (0.5 nm) with a Bentham DM150 double-monochromator spectroradiometer by M. Blumthaler (Institute of Medical Physics, University of Innsbruck). During the eld experiments, underwater UV radiation was measured with a multichannel radiometer (PUV-500A; Biospherical Instruments Inc., San Diego, Calif.). This instrument measures the downwelling UV radiation at four nominal wavelengths (305, 320, 340, and 380 nm) and PAR. The diffuse attenuation coefcient for downward radiation (Kd) was calculated from the slope of the regression between depth and the natural logarithm of the irradiance. The depth at which the irradiance is reduced to 10% of the value at the surface (Z10%) was calculated as 2.3/Kd. A correction factor of 2.6 was applied to the absolute irradiance of the 305-nm channel to compensate for the underestimation caused by the lamp calibration method (25). During the experimental work in Gossenko llesee, surface UV-B and UV-A irradiances were measured every 10 min with broad-band sensors (Delta-T Devices Ltd., Cambridge, England). The sensors have a maximum responsiveness at 312 (UV-B) and 370 (UV-A) nm. Chlorophyll a and temperature. For chlorophyll a estimation, 2 liters was ltered onto Whatman GF/F lters, and the lters were kept frozen until analysis. The samples were extracted with 90% acetone and sonicated for 2 min with a tip sonicator (Bandelin Instruments, Berlin, Germany) on an ice bath. The extracts were then ltered through an Anodisc lter of 0.1-m pore size (Whatman, Maidstone, England), and the absorbance was measured at different wavelengths against an acetone blank. The formulas of Jeffrey and Humphrey (19) were used to calculate pigment concentration.

RESULTS Underwater UV penetration. The attenuation of UV radiation at 305 and 340 nm was almost three times lower in Gossenko llesee than in the Adriatic Sea (Fig. 1). Accordingly, 10% of the surface UV-B irradiance at 305 nm almost reached the

4180

SOMMARUGA ET AL.

APPL. ENVIRON. MICROBIOL.

higher than those in the upper water column (Table 1), samples exposed to full sunlight or to UV-A-plus-PAR showed higher incorporation of TdR than samples kept in the dark (Table 1). However, in the same samples the incorporation of Leu was inhibited, although to a lower percentage than at 2.5 m (Table 1). Laboratory experiments. In the experiments in which bacteria were exposed without C. geitleri, the major inhibition of TdR and Leu incorporation (60%) was caused by UV-A radiation (Fig. 3A). The effects of the treatments were significant for TdR and Leu incorporation (Kruskal-Wallis one-way ANOVA on ranks, P 0.05), and there was a signicant difference between the UV-B-plus-UV-A-plus-PAR and UVA-plus-PAR treatments for TdR and Leu (SNK test, P 0.05), indicating an effect of UV-B. TdR and Leu incorporations

FIG. 1. Typical depth proles of underwater UV penetration at 305 and 340 nm measured with a PUV 500A radiometer in Gossenko llesee (solid lines) and the northern Adriatic Sea (dashed lines). The diffuse vertical attenuation coefcients (Kd) are also shown.

bottom at 9.9 m in the high mountain lake, but the same irradiance reached only 4 m in the Adriatic Sea. Field experiments. The experiments at Gossenko llesee and the northern Adriatic Sea consistently showed that the inhibition of TdR and Leu incorporation after short-term exposure of natural bacterial assemblages to solar radiation was caused almost equally by UV-A radiation and PAR (Fig. 2). At both sites, there was no signicant difference in the rate of incorporation of TdR and Leu between the full-sunlight and UV-Aplus-PAR treatments (Kruskal-Wallis one-way analysis of variance [ANOVA] on ranks and post-hoc comparisons with the Student-Newman-Keuls [SNK] test, P 0.05). However, there was a signicant difference between the UV-A-plus-PAR and PAR treatments for TdR and Leu in Gossenko llesee, although there was a signicant difference only for TdR in the Adriatic Sea samples (SNK test, P 0.05). There was also a signicant difference between the PAR and dark treatments in Gossenko llesee (SNK test, P 0.05). In the Adriatic Sea the effect of PAR was signicant for Leu but not for TdR (SNK test, P 0.05). In the experiments done at a depth of 2.5 m in Gossenko llesee, where approximately 50% of the surface UV-B radiation (305 nm) and 75% of the UV-A radiation (340 nm) were found (Fig. 1), the inhibition of TdR and Leu incorporation was caused mainly by UV-A radiation; PAR had only a very small effect (5%) (Table 1). A large difference in the percent inhibition by UV-A was observed in experiments when samples were exposed for 3 h (July 22) or 4 h (August 1). The integral values of surface UV-A irradiance calculated between 1000 and 1400 h (local time) were 885 and 869 W m2, respectively. At a depth of 8.5 m, where chlorophyll a concentrations were

FIG. 2. Inhibition of [3H]TdR (white bars) and [14C]Leu (black bars) incorporation after 4 h of exposure of natural bacterial assemblages to different spectral components of near-surface irradiance in Gossenko llesee (A) and the Adriatic Sea (B). Data are means standard errors for four (A) and three (B) independent experiments done on cloudless days.

VOL. 63, 1997

EFFECT OF SOLAR RADIATION ON BACTERIOPLANKTON

4181

TABLE 1. Inhibition of [3H]TdR and [14C]Leu incorporation with different treatments in the water column of Gossenko llesee
[3H]TdR Date and depth (m)a Temp (C) Chlorophyll a (g liter1) Inhibition of incorporation after exposure to: UV-B UV-A PAR UV-A PAR
b

[14C]Leu Incorporation (mol liter1 h1) (1012) in dark Inhibition of incorporation after exposure to: UV-B UV-A PAR UV-A PAR PAR Incorporation (mol liter1 h1) (1010) in dark

PAR

22 July 2.5 8.5 1 August 2.5 8.5

a b

11.1 7.7 12.4 7.4

1.4 2.6 0.4 2.9

37 (6) 49 (5) 78 (8) 2 (12)

43 (8) 47 (10) 75 (6) 50 (15)

3 (9) 2 (8) 3 (5) 1 (12)

1.35 1.39 1.3 0.96

NDc ND 82 (19) 15 (18)

ND ND 79 (9) 9 (13)

ND ND 5 (5) 5 (12)

ND ND 1.8 1.7

The exposure time was 3 h (22 July) or 4 h (1 August). Expressed as a percentage of the value for the dark control (no inhibition). Numbers in parentheses are the coefcients of variation for three replicates, and negative values indicate values higher than those for the dark control. c ND, not determined.

were not signicantly different for the PAR and dark treatments (SNK test, P 0.05). The transmittance of the medium at 310 nm decreased from 75 to 56% after the bacterial culture was mixed with C. geitleri. In the presence of algae, the incorporation of TdR and Leu by bacteria was less inhibited (Fig. 3B). For example, for the treatment with UV-B-plus-UV-A-plus-PAR, the reductions for TdR in the absence and presence of C. geitleri were 74 and 42%, respectively. The effects of the treatments on TdR and Leu incorporation were signicant (Kruskal-Wallis one-way ANOVA on ranks, P 0.05), and except for the post hoc comparison between the PAR and dark treatments, the rest of the post hoc comparisons indicated a signicant difference (SNK test, P 0.05). DISCUSSION Inhibition of TdR and Leu incorporation by UV solar radiation. We expected to nd a clear inhibitory effect of UV-B radiation on TdR and Leu incorporation, particularly at the high mountain lake, where, due to the altitude effect (4), absolute values of UV-B irradiance and the UV-B/UV-A ratio at the surface were much higher than those at sea level. In contrast, one of the main results of our study was that at both sites, within the solar UV radiation (290 to 400 nm), UV-A caused the inhibition of TdR and Leu incorporation after short-term exposure of natural bacterial assemblages to sunlight (Fig. 2). Only the study by Aas et al. (1) has assessed in a similar way to ours the contributions by UV-B and UV-A (obtained by difference) to the inhibition of TdR and Leu incorporation. In contrast to our ndings, they observed that in estuarine waters of the Santa Rosa Sound, UV-B and UV-A radiation contributed almost equally (39 and 37%, respectively) to the inhibition of TdR incorporation but that inhibition of Leu incorporation was caused mainly by UV-B. Conversely, in experiments done in the Gulf of Mexico, they did not observe any consistent pattern in the effect caused by different wavelengths. They speculated that different bacterial communities at these two sites may have responded differently to UV radiation. Two main differences exist between the methodology used in this study and that used by Aas et al. (1). First, they used polyethylene bags in most of their experiments, whereas we used only quartz asks. Although there have been major concerns about the use of polyethylene because of its potential toxicity when exposed to UV-B radiation (34), Aas et al. (1) indicated that preliminary experiments with polyethylene bags, quartz tubes, and asks gave similar results. The second dif-

ference relates to the exposure of the water sample together with the radiolabeled TdR and Leu (for up to 11.3 h) in their experiments. We exposed water samples for up to 4 h and measured incorporation afterwards by using dark incubations of 1 h. In the latter case, rates of incorporation of TdR and Leu should represent the result of the effect of radiation after exposure, providing that no substantial recovery takes place in the dark during 1 h. This is supported by results from experiments in which the incorporation of TdR and Leu by bacteria (0.8-m-pore-size ltrates from the Adriatic Sea) was monitored after exposure to articial UV-B or solar radiation, showing that recovery under dark conditions was negligible (21). During summer 1995, we conducted a series of experiments in Gossenko llesee in which radiolabeled TdR and Leu were added to whole lake water samples and then exposed to sunlight for 3 to 8 h, i.e., a method comparable to that of Aas et al. (1). Exposure of bacterioplankton together with TdR and Leu gives a more realistic estimation of the in situ incorporation rate because irradiation and uptake processes are not separated in space and time. Results from these experiments, however, were highly variable, and no consistent pattern was found for the effect on samples exposed either to the full sunlight spectrum or to only UV-A-plus-PAR (data not shown). A similar high variability in the responses and percentages of inhibition among the different treatments can also be observed in the time course experiments presented by Aas et al. (1). We have no clear explanation for this high variability, but we do not discard the possibility that UV radiation may affect the radiolabeled TdR and Leu before they are incorporated by bacteria, as photodegradation and photoalteration of even recalcitrant organic molecules have been demonstrated (9). Alternatively, it might be that different bacterial populations have distinct susceptibilities to UV radiation. In contrast to UV-B radiation, which produces direct damage to DNA, UV-A radiation is poorly absorbed by nucleic acids and proteins (30). Therefore, damage to DNA and other cellular components might occur through a photodynamic process requiring exposure to solar radiation, a chromophore, and oxygen (14). Pioneering work by Mathews and Sistrom (28) showed that survival under natural sunlight of the carotenefree (mutant) bacterium Sarcina lutea was affected only with exposure to air and not with exposure to a nitrogen atmosphere. One UV-A radiation effect in bacteria is at the membrane level, where nutrient transport is inhibited (10, 26, 29). This is relevant to the estimation of bacterial production by TdR and Leu incorporation, because transport inhibition at

4182

SOMMARUGA ET AL.

APPL. ENVIRON. MICROBIOL.

FIG. 3. Inhibition of [3H]TdR (white bars) and [14C]Leu (black bars) incorporation after 4 h of exposure in the laboratory of a mixed bacterial culture to an articial source of solar radiation in the absence (A) or presence (B) of C. geitleri with different treatments. Data are means standard errors for triplicate experiments. When error bars are not shown, the standard error was too small to be illustrated.

the membrane level may preclude the uptake of these molecules. In other words, a decrease in TdR and Leu incorporation may not necessarily be coupled with lower bacterial cell (TdR) and protein or carbon (Leu) production if UV-A inhibits the transport of these tracers into the cell. This may be particularly important if damage at the membrane is not accompanied by cell death, since postexposure recovery may take place (20, 23). The effect of articial UV-A radiation on Escherichia coli has been found to be different depending on the dose received (31). Sublethal doses produced a transient growth inhibition due to cell membrane damage, as evidenced by inhibition in the transport of [14C]alanine and slowing of protein synthesis caused by a photomodied tRNA (31, 32). Higher doses resulted in cell death caused by reactive oxygen species. Experiments with Pseudomonas aeruginosa have shown that exposure to articial UV-A radiation at the same

sublethal dose received by E. coli inhibited the transport of [14C]Leu as well as that of other amino acids and was accompanied by a substantial decrease in cell viability (10). To our knowledge, there exists no information on how TdR transport may be affected by UV-A radiation and on the importance of this effect on natural bacterial assemblages exposed to solar radiation. However, these factors need to be evaluated before we can properly assess the impact of solar radiation on bacterial production by these methods, which are used in most marine and freshwater ecological studies. If transport inhibition by solar UV-A radiation is a general effect, this may explain why no UV-B effect was detected by our approach. In addition, damage to bacteria by reactive oxygen species may be exacerbated by enclosed experimental systems that preclude the diffusion of inhibitory photoproducts. Inhibition of TdR and Leu incorporation by PAR. In the eld experiments, we also found a strong inhibitory effect of PAR (up to 30%) on TdR and Leu incorporation in samples exposed just beneath the surface (Fig. 2). This effect was absent in experiments made with articial visible light (Fig. 3) of 20-times-lower irradiance (0.01 microeinstein cm2 s1) than natural sunlight. Aas et al. (1) indicated an important inhibitory effect by PAR in their experiments. As discussed above for UV-A, the effects of visible light (PAR) are attributed to photodynamic processes. However, results from experiments on bacterioplankton viability (viable counts) in Antarctic waters have shown that PAR had only a minor effect on the survival of natural bacterial assemblages after exposures for 5 to 7 h near the surface (15). In contrast, bacterial isolates from the same waters were strongly affected, suggesting that culture procedures may select for UV-sensitive bacteria. For example, threshold values for inhibition of viability in an Acinetobacter sp. were between 0.01 and 0.02 microeinstein cm2 s1. Effect of solar radiation on TdR and Leu incorporation in the water column of Gossenko llesee. The diffuse vertical attenuation coefcients for UV found in Gossenko llesee are among the lowest reported in the literature for freshwater lakes (38). The experiments made on 22 July and 1 August showed that at a depth of 2.5 m, where approximately 75% of the surface UV-A radiation at 340 nm was present (Fig. 1), inhibition of TdR and Leu incorporation was caused almost totally by UV-A radiation (Table 1). Differences in the degree of inhibition by UV-A between these two dates can be explained by the different exposure times used in these experiments, as values of UV-A irradiance at the surface did not change signicantly between these two dates. This agrees well with the results of Aas et al. (1), who found that in most experiments an increase in the inhibition of TdR and Leu incorporation with exposure time occurred. At 8.5 m, chlorophyll a values were higher than those at 2.5 m or at the surface, a pattern that is characteristic during summer for most high mountain lakes. In those samples we observed a low inhibition of Leu incorporation but a strong stimulatory effect of TdR incorporation in samples receiving full sunlight or UV-A-plus-PAR. At present we cannot explain the different responses indicated by TdR and Leu incorporation or the difference in the stimulatory effects of TdR incorporation under full sunlight in the experiments done on 22 July and 1 August. However, the low inhibition or stimulation is a consequence of the lower UV radiation at this depth and probably also of the effect of phytoplankton (see below). Inhibition of TdR and Leu incorporation by UV radiation in the absence or presence of C. geitleri. Even if results from the laboratory experiments with lake cultures must be interpreted with caution (15), our results indicate that inhibition of TdR and Leu incorporation was higher when bacteria were exposed

VOL. 63, 1997

EFFECT OF SOLAR RADIATION ON BACTERIOPLANKTON

REFERENCES

4183

in the absence of C. geitleri. In the absence of algae, inhibition even by UV-B was evident (Fig. 3A). However, although the radiation from our lamps resembles fairly well the solar spectrum in the UV-B range, the lamps emit wavelengths of 295 nm and in this sense simulate the effect of ozone depletion. Nevertheless, most inhibition was caused by UV-A radiation, probably by wavelengths of 340 nm, as radiation from the lamps decreases signicantly at wavelengths above 340 nm (39). Our results agree with those from Aas et al. (1), who found in most experiments that the inhibition of TdR and Leu incorporation by solar radiation was higher in the absence of phytoplankton (size fraction, 0.8 m) than in whole-water samples. As indicated by those authors, the overall response of bacteria to solar radiation in whole-water samples includes the combination of effects of bacterial inhibition and effects on stressed phytoplankton. The exact mechanism of the ameliorating effect remains unclear, but we suggest that it may be a combined effect of shading by C. geitleri and a release of dissolved organic matter by phytoplankton under UV stress and thus a secondary stimulation of TdR and Leu incorporation. Release of different monomeric and polymeric organic compounds during growth of healthy algae and their stimulative effect on bacterial production are well known (40). The concentration of dissolved organic carbon released by stressed algae is larger than that released by healthy, nonstressed algae (35). However, how the release of extracellular dissolved organic carbon is affected by UV radiation is not known. Our results suggest that the interaction of solar UV radiation with bacteria and phytoplankton is complex and needs to be studied more intensively. Implications. Although much attention has been focused on the effects of UV-B radiation, there is clear evidence that UV-A, particularly wavelengths in the 320- to 365-nm range, can produce several types of damage to microorganisms, including formation of pyrimidine dimers in the DNA (42). Short-term experiments involving TdR and Leu incorporation are used nowadays as a standard method to estimate bacterial production in ecological studies. However, most experiments are performed under dark conditions. The results presented here indicate that present levels of solar UV-A radiation and PAR are responsible for the dramatic inhibition of TdR and Leu incorporation by freshwater and marine bacterioplankton. However, we caution that it may be premature to extrapolate this effect to estimates of bacterial production, i.e., cell and carbon production, before more precise information is obtained on how solar radiation affects the stability of TdR and Leu and, particularly, their transport into the cell. It may turn out that short-term experiments involving TdR and Leu incorporation are not the best method to assess the impact of different wavelengths of the solar spectrum on production of new bacterial cells. Furthermore, the effect of solar radiation on bacteria should also be evaluated in the context of the complex interactions with phytoplankton and dissolved organic matter.
ACKNOWLEDGMENTS We thank Ferran Garcia-Pichel, Richard Robarts, Albin Alfreider, and two anonymous reviewers for improvements to the manuscript, Mario Blumthaler for the spectral UV measurements, Birgit Sattler for the experiments performed during summer 1995, and Yasunori Watanabe for providing the vinyl chloride foil. This study was supported by the EU Environmental Program, Project EV5V-CT940512, and partially by the Austrian Science Foundation, Project P11856-BIO (to R.P.).

1. Aas, P., M. Maillie Lyons, R. Pledger, D. L. Mitchell, and W. H. Jeffrey. 1996. Inhibition of bacterial activities by solar radiation in nearshore waters and the Gulf of Mexico. Aquat. Microb. Ecol. 11:229238. 2. Arrigo, K. 1994. Impact of ozone depletion on phytoplankton growth in the Southern ocean: large-scale spatial and temporal variability. Mar. Ecol. Prog. Ser. 114:112. 3. Azam, F., T. Fenchel, J. G. Field, J. S. Gray, L. A. Meyer-Reil, and F. Thingstad. 1983. The ecological role of water-column microbes in the sea. Mar. Ecol. Prog. Ser. 10:257263. 4. Blumthaler, M., W. Ambach, and W. Rehwald. 1992. Solar UV-A and UV-B radiation uxes at two alpine stations at different altitudes. Theor. Appl. Clim. 46:3944. 5. Bothwell, M. L., D. M. J. Sherbot, and C. M. Pollock. 1994. Ecosystem response to solar ultraviolet-B radiation: inuence of trophic-level interactions. Science 265:97100. 6. Bu hlman, B., P. Bossard, and U. Uehlinger. 1987. The inuence of longwave ultraviolet radiation (u.v.-A) on the photosynthetic activity (14C-assimilation) of phytoplankton. J. Plankton Res. 9:935943. 7. Cho, B. B., and F. Azam. 1988. Major role of bacteria in biogeochemical uxes in the oceans interior. Nature 332:441443. 8. Cullen, J. J., P. J. Neale, and M. P. Lesser. 1992. Biological weighting function for the inhibition of phytoplankton photosynthesis by ultraviolet radiation. Science 258:646649. 9. De Haan, H. 1993. Solar UV-light penetration and photodegradation of humic substances in peaty lake water. Limnol. Oceanogr. 38:10721076. 10. Ferna ndez, R. O., and R. A. Pizarro. 1996. Lethal effect induced in Pseudomonas aeruginosa exposed to ultraviolet-A radiation. Photochem. Photobiol. 64:334339. 11. Garcia-Pichel, F. 1994. A model for internal self-shading in planktonic organisms and its implications for the usefulness of ultraviolet sunscreens. Limnol. Oceanogr. 39:17041717. 12. Guerrero, M. A., and R. D. Jones. 1996. Photoinhibition of marine nitrifying bacteria. I. Wavelength-dependent response. Mar. Ecol. Prog. Ser. 141:183 192. 13. Guillard, R. R. L. 1975. Culture of phytoplankton for feeding marine invertebrates, p. 2960. In W. L. Smith, and M. H. Chanley (ed.), Culture of marine invertebrate animals. Plenum Book Publishing Corp., New York, N.Y. 14. Harrison, A. P. 1967. Harmful effects of light, with some comparisons with other adverse physical agents. Annu. Rev. Microbiol. 21:143156. 15. Helbling, E. W., E. R. Marguet, V. E. Villafan e, and O. Holm-Hansen. 1995. Bacterioplankton viability in Antarctic waters as affected by solar ultraviolet radiation. Mar. Ecol. Prog. Ser. 126:293298. 16. Herndl, G. J., G. Mu ller-Niklas, and J. Frick. 1993. Major role of ultraviolet-B in controlling bacterioplankton growth in the surface layer of the ocean. Nature 361:717719. 17. Hooper, A. B., and K. R. Terry. 1974. Photoinactivation of ammonia oxidation in Nitrosomonas. J. Bacteriol. 119:899906. 18. Horrigan, S. G., A. F. Carlucci, and P. M. Williams. 1981. Light inhibition of nitrication in sea-surface lms. J. Mar. Res. 39:557565. 19. Jeffrey, S. W., and G. F. Humphrey. 1975. New spectrophotometric equations for determining chlorophylls a, b and c and c1 in higher plants, algae and natural phytoplankton. Biochem. Physiol. Panz. 167:191194. 20. Jeffrey, W. H., R. J. Pledger, P. Aas, S. Hager, R. B. Cofn, R. Von Haven, and D. L. Mitchell. 1996. Diel and depth proles of DNA photodamage in bacterioplankton exposed to ambient solar ultraviolet radiation. Mar. Ecol. Prog. Ser. 137:283291. 21. Kaiser, E., and G. J. Herndl. 1997. Rapid recovery of marine bacterioplankton activity after inhibition by UV radiation in coastal waters. Appl. Environ. Microbiol. 63:40264031. 22. Karentz, D., F. S. MacEuen, M. C. Land, and W. C. Dunlap. 1991. Survey of mycosporine-like amino acid compounds in Antarctic marine organisms: potential protection from ultraviolet exposure. Mar. Biol. 108:157166. 23. Kelland, L. R., S. H. Moss, and D. J. G. Davies. 1983. Recovery of Escherichia coli K-12 from near-ultraviolet radiation-induced membrane damage. Photochem. Photobiol. 37:617622. 24. Kim, D. S., and Y. Watanabe. 1993. The effect of long wave ultraviolet radiation (UV-A) on the photosynthetic activity of natural populations of freshwater phytoplankton. Ecol. Res. 8:225234. 25. Kirk, J. T. O., B. R. Hargreaves, D. P. Morris, R. B. Cofn, B. David, D. Frederickson, D. Karentz, D. R. S. Lean, M. P. Lesser, S. Madronich, J. H. Morrow, N. B. Nelson, and N. M. Scully. 1994. Measurements of UV-B radiation in two freshwater lakes: an instrument intercomparison. Arch. Hydrobiol. Beih. 43:7199. 26. Kubitschek, H. E., and R. J. Doyle. 1981. Growth delay induced in Escherichia coli by near-ultraviolet radiation: relationship to membrane transport functions. Photochem. Photobiol. 33:695702. 27. Lindell, M., and H. Edling. 1996. Inuence of light on bacterioplankton in a tropical lake. Hydrobiologia 323:6773. 28. Mathews, M. M., and W. R. Sistrom. 1959. Function of carotenoid pigments in nonphotosynthetic bacteria. Nature 184:18921893.

4184

SOMMARUGA ET AL.

APPL. ENVIRON. MICROBIOL.

planktonic marine bacteria in ltered seawater. Mar. Ecol. Prog. Ser. 33:19 27. Simon, M. 1990. Improved assessment of bacterial production: combined measurements of protein synthesis via leucine and cell multiplication via thymidine incorporation. Ergeb. Limnol. 34:151155. Sommaruga, R., and R. Psenner. 1997. Ultraviolet radiation in a high mountain lake of the Austrian Alps: air and underwater measurements. Photochem. Photobiol. 65:957963. Sommaruga, R., A. Oberleiter, and R. Psenner. 1996. Effect of UV radiation on the bacterivory of a heterotrophic nanoagellate. Appl. Environ. Microbiol. 62:43954400. Sondergaard, M., B. Riemann, and N. O. G. Jorgensen. 1985. Extracellular organic carbon (EOC) released by phytoplankton and bacterial production. Oikos 45:323332. Suttle, C. A., and F. Chen. 1992. Mechanisms and rates of decay of marine viruses in seawater. Appl. Environ. Microbiol. 58:37213729. Tyrrell, R. M. 1992. Inducible responses to UV-A exposure, p. 5964. In F. Urbach (ed.), Biological responses to ultraviolet-A radiation. Valdenmar Publishing Co., Overland Park, Kans.

29. Moss, S. H., and K. C. Smith. 1981. Membrane damage can be a signicant factor of inactivation of Escherichia coli by near-ultraviolet radiation. Photochem. Photobiol. 33:203210. 30. Peak, M. J., and V. Peak. 1989. Solar-ultraviolet induced damage to DNA. Photodermatology 6:115. 31. Pizarro, R. A. 1995. UV-A oxidative damage modied by environmental conditions in Escherichia coli. Int. J. Radiat. Biol. 68:293299. 32. Pizarro, R. A., and L. V. Orce. 1988. Membrane damage and recovery associated with growth delay induced by near UV radiation in Escherichia coli K-12. Photochem. Photobiol. 47:391397. 33. Porter, K. G., and Y. S. Feig. 1980. The use of DAPI for identifying and counting aquatic microora. Limnol. Oceanogr. 25:943948. 34. Pre zelin, B. B., and R. C. Smith. 1993. Polyethylene bags and solar ultraviolet-radiation-response. Science 259:534535. 35. Ridal, J. J., and R. M. Moore. 1993. Resistance to UV and persulphate oxidation of dissolved organic carbon produced by selected marine phytoplankton. Mar. Chem. 42:167188. 36. Sieracki, M. E., and J. M. Sieburth. 1986. Sunlight-induced growth delay of

37. 38. 39. 40. 41. 42.