Is UV-B radiation affecting charophycean algae in shallow freshwater systems?


  • Nancy V. J. De Bakker,

    Corresponding author
    1. Institute of Ecological Science, Department of Systems Ecology, Vrije Universiteit, de Boelelaan 1085, 1081 HV Amsterdam, the Netherlands;
      Author for correspondence: N.V.J. de Bakker Tel: +31 20 5986964 Fax: +31 20 5897123 Email:
    Search for more papers by this author
  • P. M. Van Bodegom,

    1. Institute of Ecological Science, Department of Systems Ecology, Vrije Universiteit, de Boelelaan 1085, 1081 HV Amsterdam, the Netherlands;
    Search for more papers by this author
  • W. H. Van De Poll,

    1. University of Groningen, Department of Marine Biology, PO Box 14, 9750 AA Haren, the Netherlands;
    Search for more papers by this author
  • P. Boelen,

    1. Institute of Ecological Science, Department of Systems Ecology, Vrije Universiteit, de Boelelaan 1085, 1081 HV Amsterdam, the Netherlands;
    Search for more papers by this author
  • E. Nat,

    1. National Information Centre on Charophytes, National Herbarium Netherlands Branch Leiden, Department of Algology, PO Box 9514, 2300 RA Leiden, the Netherlands
    Search for more papers by this author
  • J. Rozema,

    1. Institute of Ecological Science, Department of Systems Ecology, Vrije Universiteit, de Boelelaan 1085, 1081 HV Amsterdam, the Netherlands;
    Search for more papers by this author
  • R. Aerts

    1. Institute of Ecological Science, Department of Systems Ecology, Vrije Universiteit, de Boelelaan 1085, 1081 HV Amsterdam, the Netherlands;
    Search for more papers by this author

Author for correspondence: N.V.J. de Bakker Tel: +31 20 5986964 Fax: +31 20 5897123 Email:


  • • The objective of this study was to determine the effects of UV-B radiation on charophycean algae under natural conditions, since charophytes enhance water transparency in freshwater systems and levels of UV-B radiation have increased by ozone depletion.
  • • Potential and actual UV-B effects were studied by combining a glasshouse experiment in which plants were exposed to various levels of UV-B radiation and field measurements in two freshwater systems dominated by charophytes in the Netherlands.
  • • The glasshouse experiment showed that charophytes were sensitive to UV-B radiation. UV-B radiation negatively affected growth, while it increased levels of DNA damage in Chara aspera. Moreover, the charophytes did not seem to develop UV-B screens to protect against UV-B radiation since no increase in UV-B absorbing compounds was found.
  • • At field conditions, both spectroradiometrical measurements and DNA dosimeters showed that UV-B radiation was attenuated quickly in both freshwater systems, indicating that UV-B does not reach the submerged charophyte vegetation. However, specific conditions, like fluctuating water tables, may result in UV-B exposure to charophytes for certain periods annually.

photosynthetic active radiation (400–700 nm)


ultraviolet-A radiation (315–400 nm)


ultraviolet-B radiation (280–315 nm)


biologically effective dose, with action spectrum weighted UV dose


cyclobutane-pyrimidine dimer.


Due to ozone depletion, the levels of harmful ultraviolet-B (280–315 nm; UV-B) radiation reaching the Earth's surface have increased at mid and high latitudes in the Northern hemisphere during the past decades (Madronich et al., 1998; SORG, 1999). Studies on the effects of UV-B radiation showed that UV-B radiation may a.o. cause damage to DNA, affect growth and induce production of protective UV-B absorbing compounds (Rozema et al., 1997 and references therein). The effects of UV-B radiation on different organisms in terrestrial and marine environments have extensively being reviewed (Franklin & Forster, 1997; Caldwell et al., 1998; Häder et al., 1998; Searles et al., 2001; Rozema et al., 2002). Several studies have emphasised the need and importance of studying the ecological implications of penetrating UV-B radiation in freshwater (Scully & Lean, 1994; Williamson, 1995; Arts et al., 2000). However, data are limited for submerged vegetation in freshwater ecosystems.

In aquatic environments attenuation of UV-B radiation in the water column and thus its potential damage is largely affected by absorption and scatter due to, for example dissolved organic carbon (DOC), chlorophyll (phytoplankton) and suspended particles (Kirk, 1994). Although the concentration of DOC is generally higher in freshwater systems than in open oceans, resulting in higher UV-B attenuation, UV-B radiation may still penetrate to considerable depths (Scully & Lean, 1994; Morris et al., 1995; Huovinen et al., 2003). Many studies in freshwater systems primarily focussed on chemical and physical parameters contributing to the attenuation of UV-B radiation in the water column (e.g. Morris et al., 1995; Hodoki & Watanabe, 1998; Arts et al., 2000; Huovinen et al., 2003) or on the effects of UV-B radiation on free-floating plants and planktonic organisms (e.g. De Lange & Van Reeuwijk, 2003; Alonso et al., 2004). In contrast, the effects of UV-B radiation on submerged vegetation have received little attention.

Enhanced levels of UV-B radiation might be of importance in clear (shallow) freshwater lakes. Measurements on UV-B penetration in Dutch freshwater systems by De Lange (2000) showed that UV-B penetration varies from a few centimetres up to several decimetres (1% depth). However, all these systems were dominated by phytoplankton (H. J. De Lange, personal comment). Submerged vegetation and especially the presence of charophycean algae play an important role in increasing water transparency by nutrient accumulation, preventing resuspension and by the production of allelopathic substances (Scheffer et al., 1993; Scheffer, 1998; Blindow et al., 2002).

Charophycean algae are found in freshwater systems from shallow (0.3 m) to deeper sites (> 20 m) (Moore, 1986; Van den Berg, 1999). That charophytes contribute to increased transparency suggests that UV-B radiation might penetrate deeper in these systems. Moreover, enhanced levels of UV-B radiation due to ozone depletion increase the UV-B doses in depth as well. The objective of this study therefore was to determine whether charophycean algae are affected by solar UV-B radiation under natural conditions.

We used a combination of a glasshouse experiment and a field study as an integrated approach to study this question. The glasshouse experiment was designed to answer the following questions: first is there an effect of UV-B dose on plant length, DNA damage and UV-B absorbing capacity within one ozone reduction scenario (dose-effect); and second do increasing levels of UV-B radiation lead to a linear change in those parameters (dose–response relation). In the field the actual effects of solar UV-B radiation on charophycean algae were studied by measurements on attenuation of UV- B radiation and the resulting DNA damage in the charophycean algae.

Materials and Methods

Glasshouse experiment

Experimental design  In a temperature-controlled glasshouse, charophycean algae were exposed to seven different treatments: four different levels of UV-B radiation and three additional UV-A control treatments.

Charophycean algae were grown from sediment, containing oospores and bulbils, collected on 3 May 2002 in the ‘Buiten Muy’ on the island Texel (53°07′ N; 4°46′ E), the Netherlands, and stored at 4°C in dark until the start of the experiment. In aquaria (∅ 21.5 cm; height 23.5 cm) sediment was put, in which eight small perforated pots (∅ 4.4 cm; height 4.8 cm) were placed such that the sediment nearly covered the pots. On top of the sediment, an 18-cm water layer was brought, consisting of tap water inoculated with 2 l water from the Buiten Muy. UV-transmitting Perspex largely closed the top of the aquaria to reduce evaporation. Addition of demineralised water compensated evaporation during the experiment. Air temperature in the glasshouse was between 24°C (day) and 16°C (night). The aquaria were wrapped in black plastic to prevent algal growth on the sides. Photosynthetically active radiation (PAR) of 150 µmol m−2 d−1 was added to the natural background radiation to all aquaria by Philips HPI/T (400 W; Philips, Amsterdam, the Netherlands) for 14 h d−1.

In the middle of this light period charophycean algae were exposed to UV radiation for 7 h d−1. The UV-B treatments covered no UV-B present (‘PAR’), an UV-B dose at ambient ozone conditions (‘Ambient UVAB’) and two different UV-B doses corresponding to 20 and 40% ozone reduction (‘20%’– and ‘40% ozone reduction UVAB’, respectively), based on the average summer dose during the period of the experiment in the Netherlands. Doses were calculated using the model of Björn & Murphy (1985), assuming cloudless sky conditions. The simulated above water UV-B treatments corresponded to average biologically effective UV doses (UVbe) of 0, 3.8, 5.4, 9.3 KJ m−2 d−1, respectively, according to the Generalised Plant Action Spectrum of Caldwell (1971, normalised to 1 at 300 nm). Actual outdoor UVbe irradiance levels measured by the Royal Dutch Meteorological Institute (KNMI) showed similar average levels of ambient UV-B radiation (2.8 ± 0.9 kJ m−2 d−1). By using a portable broadband UV-X radiometer with a UV-X 31 sensor (San Gabriel, CA, USA), which was calibrated with a double-monochromator spectroradiometer (Optronic Model OL 752, Orlando, FL, USA) the different light conditions were set. At the end of the experiment light conditions were measured again at three or four randomly chosen positions with the spectroradiometer to check applied doses.

In the UV-B treatments the Philips TL 40 UV tubes were covered with 0.1 mm thick cellulose acetate foil (Tamboer and Co., Haarlem, the Netherlands), which absorbs radiation with wavelengths < 290 nm. Three UV-A control treatments were adjusted to the corresponding UV-B level. After adjustment, the UV tubes of these treatments were covered with 0.13 mm thick polyester foil (Mylar, Dupont Industries, Wilmington, DE, USA), which absorbs radiation < 313 nm. Mylar and cellulose acetate foils were changed once and twice a week, respectively, to avoid changes in radiation regimes due to photodegradation of the foils. In the ‘PAR’ treatment the UV tubes were off. Six replicate aquaria were used in each treatment.

Unfortunately, plants in all aquaria of the treatment simulating ambient UV-B conditions (‘Ambient UV-B’) were largely covered with periphyton that could not be removed without damaging the plants; this was not observed in any other treatment. Since the periphyton cover may have influenced results, the data from this treatment were not further analysed and presented.

Harvest  After 5 wk, plants from all aquaria were harvested and analysed for length, chemistry and DNA damage. The vegetation consisted of C. aspera Deth. Ex Wild., with only a few plants of C. contraria A. Braun ex Kützing in each aquarium. Therefore only results on C. aspera are presented.

Samples for DNA damage were taken directly after the UV treatment and the next day before the subsequent UV treatment to test whether DNA damage had been repaired in between the UV treatments. Sampling was done by cutting the top two cm from all plants of one small pot, which were put into liquid nitrogen immediately and stored at −80°C until further analyses. Subsequently, two small pots were collected for measurements on plant length. After species identity was verified for each individual plant, total plant length was measured. Samples for analysis of UV-B absorbing compounds were collected from the upper part of the plants, freeze-dried and stored under vacuum until analysis.

Field measurements

Site description  Two freshwater systems in the Netherlands characterised by charophycean algae were analysed in summer 2002. The ‘Buiten Muy’, a natural dune slack on the island of Texel (53°07′ N; 4°46′ E) was analysed on 24 June 2002. Two subsites were selected, one 0.4 m deep previously sod-cutted with a charophycean algal vegetation and one shallower site (0.2 m) that was codominated by Chara aspera and Littorella uniflora. L. uniflora is a submerged macrophyte that also enhances water transparency (Scheffer, 1998). The other site was a shallow freshwater lake, Lake Veluwe (52°23′ N; 5°42′ E; average depth 1.4 m), dominated by charophycean algae (Van den Berg, 1999). Measurements were performed on 16 July 2002. Here three subsites were selected with different depths (0.6, 0.8, 1.1 m). For characteristics on species composition and water depth levels see Table 1.

Table 1.  Characteristics of the field sites and UV-B penetration represented as attenuation coefficients (Kd ± SE) and corresponding 1% depths (Z1%) of Setlow weighted spectroradiometrical measurements and DNA dosimeters
Location Vegetation composition (%)Water depth above plants (m)Spectroradiometer1DNA dosimeter2
Kd-Setlow± SE (m−1)Z1% (m)Kd-DNA ± SE (m−1)Z1% (m)
All dataIncubation lines
  • 1

    Spectroradiometrical attenuation coefficients, weighted with Setlow action spectrum normalised at 300 nm, are based on all measurements at different depths. The r2-values ranged from 0.97 to 0.99.

  • 2

    Attenuation coefficients of DNA dosimeters are based on the individual incubation lines of quartz tubes in depth, which showed significant regressions with r2-values between 0.77 and 0.99.

Buiten Muy – TexelBM IC. aspera (100)0.2440.6 ± 2.4 0.1134.4 ± 2.8 0.13
  (n = 18) (n = 9) 
BM IIC. aspera (80)0.1172.3 ± 2.6 0.0639.8 ± 6.6 0.12
Littorella uniflora (20) (n = 4) (n = 3) 
Veluwe MeerVM IC. aspera (15)0.4530.5 ± 1.9 0.1522.3 ± 0.7 0.21
C. contraria (85) (n = 16) (n = 3) 
VM IIC. aspera (15)0.6031.7 ± 1.7 0.1423.1 ± 3.9 0.20
C. contraria (85) (n = 13) (n = 3) 
VM IIIC. aspera (35)0.9529.4 ± 1.6 0.1625.9 ± 0.7 0.18
C. contraria (65) (n = 14) (n = 3) 

Light measurements  Natural light was measured by a broadband portable UV-X meter with a UV-X 31 sensor (San Gabriel, CA, USA), coupled to a datalogger (CR10X, Campbell Scientific, UK) that recorded the average irradiance per minute in W m−2. From significant regression of these data combined with above water spectroradiometrical measurements (see ‘UV penetration measurements’) the actual Caldwell weighted biologically effective UV doses were estimated.

UV penetration measurements  Attenuation of UV-B radiation in both freshwater systems was measured with first a spectroradiometer, and second by exposing DNA dosimeters at several depths. The spectroradiometer measurements were used to directly estimate the attenuation coefficient for UV-B radiation. A MACAM SR9910 double monochromator scanning spectroradiometer (Macam Photometrics, Livingstone, UK) with a 4.2-m quartz cable connected to a cosine collector was used to measure spectra above water and at different depths UV-B radiation (230–315 nm), UV-A radiation (315–400 nm) and PAR (400–700 nm) at 1-, 5- and 20-nm intervals, respectively. UVbe doses were calculated above water following Generalised Plant Action Spectrum (Caldwell, 1971) for comparison with the glasshouse experiment. For comparing UV-B levels to DNA damage Setlow (1974) UVbe doses were calculated underwater. Both action spectra were normalised to 300 nm.

Complementary, DNA dosimeters (Boelen et al., 1999) gain information about the potential UV-B induced DNA damage at different depths and was used to calculate an attenuation coefficient based on a daily UV-B dose. The most frequent occurring UV-B induced DNA lesions are cyclobutane pyrimidine dimers (CPDs) (Britt, 1999; Dany et al., 2001). Therefore CPDs were taken as a measure for DNA damage in this study. Quartz tubes (outer ∅ 0.7 cm; inner ∅ 0.5 cm; length 7.5 cm) filled with a solution of 12 µg ml−1 calf thymus DNA (Sigma-Aldrich, St Louis, MO, USA) in TE (10 mm Tris-HCl pH 8.0, 1 mm EDTA) and sealed with parafilm were exposed during a 1-d period. The deepest quartz tubes were incubated within the plant layer. At least three replicates were incubated at each depth and at all sites, except at Lake Veluwe II (0.8 cm), where only two replicates were used. After exposure, quartz tubes were transported at 4°C and in the dark, after which the DNA solution was stored at –20°C until analyses for CPDs.

Attenuation coefficients for biologically-weighted UV-B radiation (Kd-Setlow) and for DNA damage (Kd-DNA) were calculated by linear regression of natural log-transformed data in depth using the log-linear part of the curve. In addition 1% depths (Z1%) for both biologically effective irradiances and CPD levels were calculated using the attenuation coefficients, assuming constant Kd values with depth.

Charophyte sampling  To test the occurrence (and accumulation) of CPDs in the charophytes during the day, charophycean algal material was collected before, around midday and after incubation of the dosimeters at the end of the day. Samples were immediately put into liquid nitrogen and stored at −80°C until analysis.

Chemical analysis

Plant DNA extraction and CPD quantification  DNA from the charophycean algae was extracted by grinding the material in liquid nitrogen and suspending it in 0.7 ml lysis buffer (20% SDS, 50 mm EDTA pH 8.0, 20% sarkosyl, 3.2 ml Tris saturated phenol, 21 g urea, 66.4 mg o-ethyl xanthic acid- potasium salt) at 65°C. After 20 min, 0.7 ml chloroform: isoamylalcohol: phenol (ratio 24 : 1 : 25) was added, mixed for 20 min at room temperature and centrifuged for 5 min at 13 000 rpm. 50 µl of sodium acetate (3 m) and 500 µl isopropanol was added to the supernatant and centrifuged for 5 min at 13 000 rpm. The precipitated DNA was washed with 70% alcohol, air dried and dissolved in 0.5 ml ddH2O. Samples were incubated with RNase for 10 min at 65°C. Extra cleaning of the DNA was done by repeating the protocol from the chloroform: isoamylalcohol: phenol (ratio 24 : 1 : 25)-step onwards. DNA was finally dissolved in 0.15 ml ddH2O and stored at −20°C until further analysis.

DNA concentrations were quantified fluorometrically with Picogreen (Molecular Probes, Eugene, OR, USA) using a 1420 Victor multilabel counter (Wallac, Inc. Gaitherburg, MD, USA). Subsequently, 100 ng DNA per sample was used to detect CPDs following Van de Poll et al. (2002).

UV absorbing capacity  UV absorbing compounds in charophycean algae were extracted (Caldwell, 1968) in acidified methanol (CH3OH: demineralised water: HCl in ratio 79 : 20 : 1) for 1 h at 90°C. After centrifugation at 2500 rpm the absorbance of the supernatant was measured at 280–315 nm on a Shimadzu UV-1601PC spectrophotometer. The total absorbances at 280–315 nm and at 315–400 nm were used as a measure for UV-B and UV-A absorbing compounds, respectively.

Statistical analysis

The dose-effects of UV-B and UV-A radiation on plant length, CPDs in DNA, UV-B and UV-A absorbing capacity were tested by applying a one-way anova with a priori comparisons. In the comparisons the UV-A control treatments were tested together against the PAR treatment to investigate the effect of UV-A radiation emitted by the TL-tubes. Within each ozone reduction scenario the UV-B treatment were tested against the UV-A treatment for the dose-effect of UV-B radiation. For plant length the individual measurements were analysed by using a anova nested by aquarium with the described a priori comparisons.

To answer the second, dose–response, question, UV-B induced DNA damage and UV-B absorbing compounds were analysed by a covariance analysis. The unweighted UV-B and unweighted UV-A doses were used as covariates in the ancova and treatment, either UV-A or UV-B, as fixed factor. The data on plant length were analysed by using an ancova nested by aquarium, while the data on CPDs in C. aspera measured on subsequent days were analysed by a repeated measures ancova.

Before performing statistical analyses, normality and homogeneity of variances were tested by using Kolmogorov-Smirnov and Levene's test, respectively. Only CPD data had to be 10log transformed to meet the assumptions for the ancova. Subsequently, for the ancova; presence of interaction between the covariates and treatments were tested. As such interactions were absent in all analyses, an ancova without interaction terms was performed (Neter et al., 1996).

All statistical tests were performed using SPSS version 10.0; significance level was 0.05.


Glasshouse experiment on UV-B effects

Increasing UV-B doses significantly reduced average plant length of C. aspera (ancovaP < 0.001; Fig. 1). No significant dose–response effect was found for the different levels of UV-A radiation (ancovaP = 0.948). The negative effect of UV-B on the average plant length had already become significant at the above water UV-B dose of 5.4 kJ m−2 d−1, the dose corresponding to the 20% ozone depletion scenario. Here, plants were on average 14% shorter than in the UV-A control treatments that received no UV-B radiation (anova contrast P = 0.001). At 40% ozone depletion the UV-B effect was significant as well (anova contrast P = 0.027).

Figure 1.

The average plant length (+ SE, n = 6) at the end of the experiment for the different treatments. For each ozone scenario the grey bars represent UV-B treated plants; dashed white bars represent the UV-A control treatments. The PAR treatment serves as control for both UV-A and UV-B treatments (grey and dashed bar). The small accents (*) mark the significant difference (P < 0.05) between the UV-A and UV-B treatment from the a priori comparisons in the one-way anova. The ancova analysis showed a significant negative dose–response relation for UV-B radiation (P < 0.001).

In addition to the reduced average plant length, a significant increase in the amount of CPDs in the DNA was found as a result of increasing UV-B radiation (ancovaP = 0.008; Fig. 2). The UV-B doses simulating 20% and 40% reduction in ozone layer thickness both had significant effects on the UV-B induced DNA damage (anova contrasts P = 0.038 and P = 0.002, respectively). The amounts of CPDs in C. aspera did not change significantly directly after and before the subsequent UV-B treatments (P = 0.88), indicating that no detectable repair had taken place over night.

Figure 2.

The UV-B induced DNA damage per treatment, expressed as amounts of CPDs per million nucleotides (Mb) (averages + SE, n = 6). The left bar in every treatment represents the damage directly after the UV treatment, while the right bar shows the damage before the next UV treatment. For each ozone scenario the grey bars represent UV-B treated plants; dashed white bars represent the UV-A control treatments. The PAR treatment serves as control for both UV-A and UV-B treatments (grey and dashed bar). The small accents (*) mark the significant difference (P < 0.05) between the UV-A and UV-B treatment from the a priori comparisons in the one-way anova. The ancova analysis showed a significant dose–response relation for the amounts of CPDs under elevated UV-B radiation (P = 0.008).

In all treatments that received UV radiation, thus all treatments apart from the PAR control treatment, higher levels of UV-B absorbing compounds were found. However, not the levels of UV-B radiation, but the presence of UV-A radiation, caused the marginally significant increase in the UV-B absorbing capacity of the Chara plants (ancovaPuv-B = 0.375; ancovaPuv-A = 0.056; Fig. 3). There were no significant differences within the different ozone reduction scenarios, while the three UV-A control treatments significantly differed from the PAR treatment that had received significantly lower UV-A doses (anova contrast P = 0.008). Identical results were found for the absorption in the UV-A region (data not shown).

Figure 3.

Total UV-B absorbing compounds (average + SE; n = 6) for the seven different treatments expressed as total absorbance from 280 to 315 nm per mg d. wt. For each ozone scenario the grey bars represent UV-B treated plants; dashed white bars represent the UV-A control treatments. The PAR treatment serves as control for both UV-A and UV-B treatments (grey and dashed bar). The small accent (*) marks the significant difference between the UV-A control treatments and the PAR control treatment from the a priori comparisons in the one-way anova. The ancova analysis showed marginally significant effects for UV-A levels (P = 0.056), while there was no effect of UV-B radiation.

Field measurements on UV-B attenuation

Based on significant regression between the UV-X and spectroradiometrical analysis the Caldwell weighted biologically effective incident UV-B doses were estimated as 5.7 and 2.6 kJ m−2 d−1 for 24 June and 16 July, respectively. This difference is explained by the mostly cloudless sky conditions at the Buiten Muy, while skies turned cloudy from 11.00 onwards (local time) at Lake Veluwe.

Spectroradiometrical measurements on direct incident UV-B irradiance and the amount of CPDs in DNA dosimeters showed an exponential decline with depth. The amounts of induced DNA damage were significantly correlated to Setlow weighted daily UV-B doses both above water and in the water column (P < 0.001, r2 = 0.99 and P < 0.001, r2 = 0,97). Attenuation coefficients for UV-B radiation from both methods were high for both freshwater systems (Table 1). The 1% depth (Z1%), the depth at which hardly any surface UV-B radiation penetrates varied therefore only around 10–20 cm, which is shallower than where most vegetation occurred. Moreover, no accumulation of CPDs was found in the plants during the day, independent of the Chara species sampled.


Potential UV-B effects on charophytes

The glasshouse experiment showed that 5 wk of daily repeated exposure to UVbe doses of 5.4 and 9.3 kJ m−2 d−1 caused significant growth reductions in C. aspera. This is consistent with growth reductions in response to UV-B exposure as found in plant species from terrestrial and marine environments (Caldwell et al., 1998; Searles et al., 2001; Van de Poll et al., 2001). Despite the fact that the ambient dose of 3.8 kJ m2 d−1 could not be taken into account, the significant dose–response relation indicated that also low doses of UV-B radiation affect plant performance. DNA damage corresponded to the pattern of growth reduction as increased plant CPD concentrations occurred at increasing UV-B doses. The presence of CPDs in the DNA affects DNA transcription and replication (Sauerbier & Hercules, 1978; Draper & Hays, 2000). Because the average growth reduction correlated significantly to the CPD levels in the plants (Pearson coefficient −0.57, P < 0.001), it is likely that UV-B induced DNA damage influenced plant growth in this study. Studies on higher plants, marine phytoplankton and marine macro algae also reported that reduced growth corresponded to increased levels of CPDs (Mazza et al., 1999; Buma et al., 2000; Van de Poll et al., 2001).

Several light-dependent and independent ways to repair DNA damage have been reported in plants (Britt, 1995). Light dependent repair by photolyase enzymes was shown to be the most efficient CPD repair mechanisms in higher plants and marine macrophytes (Quaite et al., 1994; Pakker et al., 2000a,b). Since first no significant CPD repair was found in between two subsequent UV-B treatments, and second CPDs were present at all times, the DNA repair systems in the charophycean algae apparently were unable to repair all damage. This unbalanced induction and repair causes a gradual accumulation of CPDs in the DNA, as was previously observed for marine macro algae (Van de Poll et al., 2002).

While damage to DNA and subsequent growth reduction occurred, no adequate protection by increased UV-B absorption was found. Only a small increase was found in UV-B and UV-A absorbing compounds in Chara aspera due to UV-A exposure, but in general the absorbance in the UV radiation region was low (this study; De Bakker et al., 2001; Rae et al., 2001). In higher plants induction of flavonoids has been reported, while in marine organisms mycosporine-like amino acids (MAAs) act as effective sunscreen (Cockell & Knowland, 1999), protecting underlying tissue from UV radiation. Neither types of secondary metabolites have been found in Chara aspera (data not shown) or in charophycean algae in general (Wegner-Hambloch, 1983; De Bakker et al., 2001), indicating that these organisms have a limited capacity to produce protective compounds under UV exposure.

UV-B penetration in freshwater systems

The glasshouse study showed that Charophycean algae are potentially sensitive to UV-B radiation. But do they have to cope with solar UV-B radiation under natural conditions?

While previous freshwater studies on UV-B radiation mainly focussed on factors affecting attenuation (e.g. Scully & Lean, 1994; Morris et al., 1995; Huovinen et al., 2003), the (potential) ecological effects for submerged vegetation have only occasionally received attention (Rae et al., 2001). The direct impact of solar UV-B radiation was studied on CPD levels in charophyte vegetation. Since the relation between UV-B exposure and growth is difficult to determine in plants growing in natural aquatic systems, UV-B induced DNA damage is a good alternative to study solar UV-B exposure and its effects on aquatic vegetation. Furthermore, this type of DNA damage is a highly specific indicator for UV-B stress, because it reflects exposure to short wavelength radiation only, as CPDs are only slightly induced by UV-A radiation and not by PAR (Quaite et al., 1992).

The accumulated CPDs in the DNA dosimeters correlated significantly to the Setlow weighted UV-B doses calculated from spectroradiometrical analysis above water and below water. This was also found in several earlier studies from different marine locations around the world (Boelen et al., 1999; van de Poll et al., 2002). The results were comparable for both methods, except for the Buiten Muy II where resuspension of the sediment during the spectroradiometer measurements might have disturbed light penetration. Therefore, the DNA dosimeters are useful tools to estimate potential UV-B exposure of plants in freshwater systems. Both methods based on incident irradiance and daily UV-B doses showed that UV-B radiation attenuated quickly in the water column. Consequently, UV-B radiation did not reach the charophycean algae. The attenuation coefficients of UV-B radiation of Dutch phytoplankton dominated freshwater systems measured by De Lange (2000) were often lower than our measurements, leading to deeper penetration of UV-B. Therefore, our results contrast with expectations that enhanced transparency due to presence of submerged charophytes would result in deeper penetration of UV-B radiation in those shallow freshwater systems.

The additional evidence that no direct damage of UV-B induced CPDs in charophycean DNA had occurred supports the findings that charophycean algae were not exposed to solar UV-B radiation in either system under natural conditions. The study of Rae et al. (2001) on attenuation of and sensitivity to UV-B radiation in a New Zealand lake supports our findings. They showed that the 1% depth of solar UV-B irradiance (305 nm) only reached the most upper limit of the vegetation of Chara fibrosa, so no significant exposure occurred to that charophycean alga either.

This picture may alter when temporal dynamics is accounted for. The water table in Lake Veluwe is controlled to an approximately constant water level in winter and summer. Therefore it is not expected that charophycean algae are exposed to UV-B radiation at any moment during the year. But, in the Buiten Muy at Texel, the water table fluctuates during the year, being lowest at the end of the summer. Figure 4 displays the water levels for nine subsequent years. When assuming constant attenuation of UV-B radiation in time and constant vegetation height, charophytes will potentially be exposed to UV-B radiation for a certain period almost each summer depending on the weather conditions. This occurs in eight out of the nine monitored years. This coincides with field observations in shallow freshwater that show that charophycean algae have a more compressed growth form compared with deeper freshwater systems.

Figure 4.

Water table in the Buiten Muy from 1994 to 2002. Black line represents the water level; the grey line the depth where 1% UV-B penetrates, assuming constant attenuation of UV-B in time. The soil level and the top of the Littorella vegetation are shown by the horizontal lines for Buiten Muy II. The water table in the Buiten Muy I is c. 13 cm higher than in Buiten Muy I. Data were provided by Staatsbosbeheer.

In some years, the Buiten Muy may even dry out and the water table will drop below the soil surface (in four out of the nine years). Charophycean algae die and have to re-establish after such dry periods from oospores. Some charophycean algae are also able to regenerate from bulbils; vegetative reproduction structures formed near the rhizoids. In a glasshouse study De Bakker et al. (2001) showed that exposure to UV-B alters the reproduction strategy of Chara aspera leading to an increase in vegetative bulbils, and a decrease of generative oospores. This shift in reproduction strategy may increase the survival chances of the algae, because re-establishment success from bulbils is higher than from oospores (Van den Berg et al., 2001). Field observations support this idea. Since, higher regeneration rates have been found in shallow freshwater (Nat, personal comment). Therefore it seems that UV-B exposure triggers a mechanism that enhances the survival chances of charophytes after dry periods.

However, note that from the limited data available it is unclear what actual level of UV-B causes significant effects on plant performance or on reproduction strategies (e.g. a threshold value). This is a topic that clearly needs further study.

In conclusion, charophytes are sensitive to UV-B radiation. Increasing levels of UV-B radiation led to DNA damage and growth reduction, while no increase in protective UV-B absorbing compounds was found. However, by contrast to the glasshouse experiment, charophycean algae are not frequently exposed to UV-B radiation under field conditions. Nevertheless, in wetlands with fluctuating water tables plants may be regularly exposed to UV-B radiation during certain periods of the year.


This study would not have been possible without the help of Staatsbosbeheer, in particular J. van Groenigen and M. Stoepker, Dr M. van de Berg and coworkers from RIZA – Lelystad, M. Allaart from the Royal Dutch Meteorological Institute (KNMI), Dr A. Buma from the department of Marine Biology of the University of Groningen, the Netherlands, W. Wolff, Dr E. Adema and B. Noordman, M. Stroetenga and the workshop of the Vrije Universiteit for technical support during the experiments.

This research was funded by EU (DG XII) within the program Environment and Climate (contract ENV4-CT97-0580), which is gratefully acknowledged.