Patterns of DNA damage accumulation and repair in aquatic organisms

9.5.1 Evidence from laboratory studies

Much information on DNA damage induction in aquatic organisms is based on incubation experiments under artificial UVR. Unfortunately, the ecological relevance of such studies is restricted, mainly due to the differences in spectra between artifical sources (lamps) and solar radiation. As a result, the effects of UV-B may be overestimated even when weighted UV-B irradiances are realistic. On the other hand, laboratory experiments may provide valuable information on species specific differences in UV vulnerability (see also section 9.6), the potential for repair, or effects of various environmental factors on UV vulnerability. Joux et al. [68] clearly showed species specific DNA damage accumulation in marine bacteria, mainly as a result of differences in photoreactivation potential. Earlier work showed the potential for CPD accumulation and photo-enzymatic repair capacity in cyanobacteria and diatoms [9,55,69-71]. Pakker et al. [72] demonstrated 6-4 PP and CPD accumulation in the marine red macro-alga Palmaria palmata and showed the importance of photoreactivation for successfully overcoming UV-B stress. Another study focussed on marine red algae [73] related UV-B vulnerability with patterns of CPD accumulation and repair. They found that littoral red algae were less vulnerable to CPD accumulation as compared with sub-littoral species and related UV-B vulnerability primarily with photorepair capacity.

Without directly measuring DNA damage induction in marine organisms, many studies have focussed on the potential for photoreactivation in marine organisms. Han and Kain [74] demonstrated photoreactivation by blue light in young sporophytes of several brown seaweeds. They stressed that visible light resulted in recovery from UV damage that would otherwise have caused high mortality in these organisms. Huovinen et al. [75] studied the effects of UVR on early developmental stages of the giant kelp Macrocystis pyrifira by exposing gametophytes to UV-B before and after germination. They found that nuclear division and translocation in zoospores after germination were most sensitive for UVR. Furthermore, recovery took place in the light, possibly as a result of photoreactivation of damaged DNA in the zoospores. Huovinen et al. [75] concluded that short exposures to UV-B may perturb or delay the development and recruitment of the giant kelp gametophytes by inhibiting their nuclear division and translocation. Coohill and Deering [76] found a clear relationship between photoreactivating wavelengths and survival of zoospores of the water mold Blastocladiella emersonii. Moreover, they also found that temperature largely affected photoreactivation rates, with higher temperatures causing higher survival in the spores.

CPD formation and the potential for photoreactivation have been described for fish cell lines, living fish and fish embryos [77-80], Kouwenberg et al. [81] argued that UV induced mortality in marine zooplankton and fish could be attributed to DNA damage. Biological weighting functions for egg mortality in Atlantic cod showed a high resemblance with the wavelength dependency for DNA damage induction in DNA solutions, as published by Setlow [33]. They also demonstrated decreased egg mortality rates in the presence of photoreactivating wavelengths. In the marine planktonic copepod Acartia omorii, eggs were more susceptible than older life stages [82]. UV-B induced damage was alleviated under simultaneous irradiance with enhanced PAR, indicating that photoenzymatic repair plays a major role in determining UV stress in these organisms. Kouwenberg et al. [81] concluded that younger stages may be primarily affected by incident UVR in the field. Zagarese et al. [37,83] demonstrated species specific efficiency of photoenzymatic repair for various zooplankton species. Hays et al. [84] measured amphibian egg mortality as a function of photolyase activity. They found photolyase to vary 100-fold among eggs of 10 amphibian species. Moreover, hatching success was closely related with photolyase activity [84],

It is clear from these laboratory studies that photoreactivation may be a key process in determining the final UV-B response in aquatic organisms. Most organisms that have been investigated have the potential for photorepair, i.e. they exhibit photoenzymatic activity in the presence of "photoreactivation light". However, this potential for photoreactivation may not be fully exploited by aquatic organisms in their natural environment. First of all, the spectral conditions under water are highly dynamic and may not always support the photoenzymatic process. At greater depths, for instance, UV-A or PAR may remain below the thresholds for repair. Also, chemical and energetic requirements for repair may not always be sufficient. Therefore, in situ accumulation and repair patterns are needed to further clarify the effectivity of photorepair and its consequences for DNA damage accumulation in aquatic organisms.

9.5.2 Depth related CPD patterns, diel cycles and mixing effects

In pelagic organisms, UV-B exposure is to a large extent determined by mixing phenomena. As a result of (wind induced) vertical mixing, organisms are moved up and down in the upper water column. Thus, the exposure of the organisms to the different wavelengths will strongly fluctuate and will be extremely difficult to determine (see Chapter 4). With respect to DNA damage, the induction to repair ratio will also fluctuate during mixing because both processes are strongly, but in different ways, wavelength dependent. The consequences of vertical mixing for DNA damage accumulation and repair can be assessed using models such as proposed by Huot et al. [85]. They combined irradiance, mixing, and biological models to predict net DNA damage levels in the mixed layer. Model predictions were generally consistent with measurements of DNA damage levels in the field. The model showed that the depth of the mixed layer strongly influences the amount and distribution of DNA damage in the mixed layer but that the average amount of DNA damage in the whole euphotic zone was only little affected by mixing. Moreover, model calculations revealed that photoreactivation plays a significant role in preventing rapid CPD build-up in marine bacterioplankton.

Recently, many observations on in situ UV-B related DNA damage accumulation in marine organisms have been published. The smallest organisms inhabiting aquatic ecosystems are viruses. Viral lysis can be as important as flagellate grazing in causing the mortality of marine bacterial communities. UV-induced destruction of viral infectivity may, therefore, reduce virally mediated mortality of bacterio- and phytoplankton. Viruses have been shown to accumulate DNA damage in marine waters, thereby reducing viral infectivity. Both CPDs as well as 6-4 PP showed a clear depth pattern in the Gulf of Mexico [12] (6-4 PP profile shown in Figure 5). Similar patterns of damage were observed in bacterio-plankton in the Southern Ocean (Jeffrey et al., unpublished results). In their study Weinbauer et al. [12] described the induction of both CPDs and 6-4 PPs in viral isolates, incubated at several water depths, but also in natural viral communities sampled at the end of the day (concentrated from specific depths). 6-4 PPs and CPDs showed a significantly similar distribution with depth. The percentage of the 6-4 PP relative to total measurable DNA damage averaged 3.1% in natural viral communities in the Gulf of Mexico. The generally higher damage levels in fixed isolates as compared with natural in situ samples indicated that mixing rates minimized photodamage accumulation in viral DNA. This conclusion was further supported by the CPD depth profiles of bacterioplankton from the Southern Ocean [86] and the Northern Gulf of Mexico [11,87] (Figure 6). During calm seas, damage was highest in the surface, decreased with depth and could be detected down to 10 m in the Gulf of Mexico. On moderately mixed days, however, no net accumulation of damage was observed, not even at the surface. In the Gerlache Strait, Antractica, damage was observed to accumulate below 20 m on a calm day. When seas approached 1 m, however, DNA damage at the end of the day was less than it was at sunrise [86]. It was postulated that damaged cells may be moved to deeper waters by mixing, where rates of photoreactivation may outweigh rates of UV-B induced damage, resulting in net decreases of DNA photoproducts.

Although to a lesser extent (see also section 9.6), microalgae also accumulate

Figure 5. Depth profiles of 6-4 PP in phage DNA, at two stations (station B and F) in the Gulf of Mexico. Filled circles: PWH3a-Pl phage isolate, incubated at various water depths for an entire daily period. Open triangles: natural viral communities concentrated from specific depths at the end of the solar day. [Redrawn from Weinbauer et al. 12.]

Figure 5. Depth profiles of 6-4 PP in phage DNA, at two stations (station B and F) in the Gulf of Mexico. Filled circles: PWH3a-Pl phage isolate, incubated at various water depths for an entire daily period. Open triangles: natural viral communities concentrated from specific depths at the end of the solar day. [Redrawn from Weinbauer et al. 12.]

Cyclobutane Dimers/Mb DNA (bacterioplaiikton)

Cyclobutane Dimers/Mb DNA (bacterioplaiikton)

Cyclobutane Dimers/Mb DNA (Dosiineters)

Figure 6. CPD depth profiles in bacterial assemblages and dosimeters in the Northern Gulf of Mexico, measured at 6.30h and 19.00h. Left panel: September 7,1994. Right panel: September 8,1994. [Redrawn from Jeffrey et al. 11.]

Cyclobutane Dimers/Mb DNA (Dosiineters)

Figure 6. CPD depth profiles in bacterial assemblages and dosimeters in the Northern Gulf of Mexico, measured at 6.30h and 19.00h. Left panel: September 7,1994. Right panel: September 8,1994. [Redrawn from Jeffrey et al. 11.]

CPDs in the water column or in sea ice. Prezelin et al. [88] demonstrated the induction of CPDs in frazil ice algae from the Antarctic after exposure to near-surface radiation for 4 h during the morning. No damage was induced under UV-A + PAR (Photosynthetically Active Radiation) or PAR alone. A recent study carried out in the temperate Bahia Bustamante (Chubut, Argentina) area demonstrated both photosynthetic inhibition and DNA damage formation in pro- and eukaryotic pelagic organisms due to UYR [53]. In that study, plankton samples were incubated at various depths in the water column, after which the accumulated damage was measured at the end of the afternoon. CPDs accumulated rapidly in the microbial community maintained at the surface, but at depth (3 m and 6 m) a decrease was observed, indicating that damage had been removed by repair processes. Clear depth dependent relationships in CPD abundance were found during early summer in Antarctic marine bacteria and phytoplankton assemblages, the latter mainly consisting of diatoms [89]. During early summer, shelf ice melting caused stabilization of the upper water layers in these coastal waters. Here, bacteria were found to contain much higher levels of damage as compared with the eukaryote fraction (>10 /mi). Nevertheless, phytoplankton exhibited CPD abundance during the whole summer in this Antarctic bay. Towards the end of the summer lower solar angles, decreasing water temperatures and higher mean wind speeds caused the water column to become deeply mixed. At that time CPDs were still present in all size fractions, but damage was distributed evenly throughout the water column.

Sessile organisms do not have the opportunity to be mixed below damaging wavelengths and are therefore likely to possess adequate strategies to reduce or prevent DNA damage. Lyons et al. [52] measured induction of DNA damage in microbial communities associated with coral mucus and observed increased CPD abundance with depth, an apparently anomalous result. DNA damage in coral mucus samples was also consistently lower than in organisms sampled in the water column at similar depths. These results suggested that sessile organisms inhabiting shallower waters rely more on photoprotective compounds such as MAAs or utilize more efficient repair systems. Unfortunately, laboratory experiments were not reported that might distinguish between these possibilities.

The depth profiles that are described above were either taken close to the UV-B peak around noon, or at the end of the afternoon, when CPD abundance was assumed to be maximal, if in agreement with the cumulative DNA effective UV-B dose. Any change in the physiology of the organism under investigation during daytime, such as the induction of repair systems, might prevent the CPD accumulation pattern from following the DNA effective dose. Therefore, diel patterns of DNA damage could give information on maximal CPD levels as they might occur during the day but also provide information on in situ repair or induction of repair processes.

Jeffrey et al. [11] demonstrated accumulation of DNA damage during the day in bacterioplankton assemblages in the Gulf of Mexico. They showed that CPD levels decreased immediately after sunset. During this period synthesis of new DNA (measured as thymidine incorporation) was not high to enough to explain the decrease in CPDs suggesting that removal of CPDs was mainly caused by excision (dark) repair and that DNA damage was not "diluted" by growth [11]. In addition, measurements of recA gene expression in these samples [90] and bacterioplankton from the Gerlache Strait, Antarctica [91,92], was shown to follow a clear daily pattern with maximal expression after sunset. These data collectively indicated that excision repair is essential for daily recovery from solar exposure in marine bacterioplankton.

Although there is significant evidence for the potential for photoreactivation in marine bacterioplankton [91,93], further supported by model calculations [85], direct evidence of its effectiveness in situ is limited. Boelen et al. [94] incubated CPD containing picoplankton samples at 10 m depth, where biologically effective UV-B levels were only 6% of surface levels, but UV-A and PAR (involved in photorepair) were still high (see also Table 1). A significant decrease in CPD levels was not detected, however, during the light period. Although it could not be completely ruled out that UV-A and PAR levels at 10 m depth were not sufficient to support photorepair, photoreactivation did not seem the prevailing pathway for CPD removal in these organisms. In addition, experiments carried out in the Gulf of Aqaba, Red Sea, also indicated absent or negligible photorepair, but mainly repair during the night (Figure 7). Bacterioplankton incubated in bags at the water surface as well as in situ samples showed clear daily patterns with maxima at the end of the afternoon, although the in situ samples had lower damage levels throughout [95] (Figure 7). DNA damage was also found to accumulate during the day in temperate marine bacterial and phytoplankton assemblages from Bahia Bustamante, Argentina. CPDs accumulated rapidly when samples were exposed to full solar radiation, even on cloudy days [96]. The > 10 ¬°im fraction (phytoplankton, mainly diatoms) also accumu-

solar time (h)

Figure 7. Diel cycles of CPD abundance in two bacterioplankton size fractions in the Gulf of Aqaba. (A) cell numbers of the major plankton groups; (B) in situ sampling at water surface 0.2-0.8 /im; (C) in situ sampling at water surface 0.8-10 /xm. [Redrawn from

solar time (h)

Figure 7. Diel cycles of CPD abundance in two bacterioplankton size fractions in the Gulf of Aqaba. (A) cell numbers of the major plankton groups; (B) in situ sampling at water surface 0.2-0.8 /im; (C) in situ sampling at water surface 0.8-10 /xm. [Redrawn from

lated damage, although at much lower rates as compared with the picoplankton fraction. Moreover, this work also showed that photosynthetic inhibition and CPD accumulation followed different daily patterns. This was also found previously for a tropical fresh water assemblage [97] (Lake Titicaca, Bolivia), when incubated at the surface for an entire light cycle. Here, CPD accumulated throughout the day, whereas photosynthetic inhibition by UV-B was virtually constant. In incubation experiments conducted at Bahia Bustamante, Argentina, no repair was found when marine microbial assemblages were incubated in photoreactivating light. Damage accumulated during morning hours was not removed during afternoon hours when samples were exposed to UV-A -f PAR or PAR alone. The absence of significant photorepair, therefore, was argued to contribute to the observed rapid accumulation of CPDs during the day [96]. These observations were in accordance with similar experiments conducted in the Antarctic (Buma unpublished results) where photorepair was minor or absent in experiments where UV-B was excluded. In contrast, repair was found in situ in Bahia Bustamante in samples incubated at 3 and 6 m in the water column, where UV-B levels were low, but UV-A and PAR were obviously favoring photorepair [53].

Weinbauer et al. [12] showed diel patterns of CPDs and 6-4 PPs in natural virus communities in surface waters of the Gulf of Mexico. CPD concentrations generally increased during the day and highest concentrations were found between 15:00h and 18:00 h. Samples taken the next morning showed that the damage was removed during the night. The 6-4 PP showed a comparable trend to that found for CPDs, also showing removal during the subsequent night. The authors, however, argued that this decrease might not only be due to hostmediated repair but also to dilution of damage as a result of virus replication. In another study, Wilhelm et al. [98] showed clear differential dose responses under photoreactivating and non-photoreactivating conditions in viral infectivity. It was demonstrated that host-mediated repair was able to restore infectivity for a significant proportion of the viruses, thereby allowing the viruses to complete their lytic cycle. Most recently, in situ measurements have demonstrated that up to 52% of solar radiation-inactivated viruses may be photoreactivated in coastal marine environments.

In summary, all the published diel cycles for DNA damage induction in marine bacterio- and phytoplankton indicate that, even when photorepair occurs, it will play a rather limited role. Photorepair does not hinder rapid build up of damage during the day. As a result, damage accumulation patterns are often found to roughly follow the DNA effective UV-B dose, especially after noon.

Eggs and larvae of pelagic fish may be susceptible for UV-B induced DNA damage because they are small, transparent and occur in the upper layers of the ocean. UV-B induced CPD formation and the capacity for repair were studied in newly spawned eggs and yolk-sack larvae of northern anchovy, Engraulis mor-dax, exposed to natural radiation [99], Eggs and larvae died when exposed to full solar irradiance. At lower levels, i.e. more natural conditions for the water column, there was a clear diel cycle of dimer concentration. This pattern closely followed solar intensity (Figure 8) and not the DNA effective dose, as found for bacteria and phytoplankton (see above). This diel cycle was thought to be due to the instantaneous interaction of damage and true photorepair, whereas dark (excision) repair was shown to be of minor importance. Photoreactivation could be stopped when samples were transferred to the dark. Unhatched embryos, spawned in the dark, also exhibited a strong photorepair response, indicating that photolyase expression in these organisms is not dependent on the previous UVR regime. Vetter et al. [99] concluded that CPD concentration at the time of sampling is a good indicator of dose rate and not of the cumulative dose and that anchovy have a highly efficient photoenzymatic repair system. In agreement with this, efficient photorepair capability in anchovy had been described before [100] showing increasing larval survival under photoreactivating conditions.

It is clear that more information is needed on the conditions affecting photo-reactivation in aquatic organisms. First of all, field measurements give contradic-

Dna Photorepair

time (cumulative hours from midnight day 1) Figure 8. CPD concentrations in unhatched embryos of northern anchovy: (A) Newly spawned eggs, exposed to full solar irradiance (bold solid line) or with 50% reduction in UV-B (lighter solid line). (B) UV-B dose rates, measured as the mean of four scans (Optronics OM 752 spectroradiometer) per hour, integrated between 280 and 320 nm.

[Redrawn from Vetter et al. 99.]

time (cumulative hours from midnight day 1) Figure 8. CPD concentrations in unhatched embryos of northern anchovy: (A) Newly spawned eggs, exposed to full solar irradiance (bold solid line) or with 50% reduction in UV-B (lighter solid line). (B) UV-B dose rates, measured as the mean of four scans (Optronics OM 752 spectroradiometer) per hour, integrated between 280 and 320 nm.

[Redrawn from Vetter et al. 99.]

tory results with respect to the importance and effectiveness of photoreactiva-tion. At the same time, highly efficient photoreactivation potential has been reported for a large variety of aquatic organisms, when exposed to artificial (light) conditions. Therefore, even when the potential for photoreactivation is present, other factors like nutrient and energy supply or differences in artificial lamp and solar spectra may interfere with the induction of repair systems. In addition, UV-B exposure might cause viability loss in aquatic organisms in situ, thereby shutting off CPD removal. This possibility of DNA damage induction beyond the capability of repair will be addressed in the next section.

9.5.3 Residual DNA damage and viability

Almost all field studies that have been carried out so far have demonstrated the presence of so-called residual DNA damage before sunrise. The occurrence of residual DNA damage indicates that damage induced during previous UV-B exposure events was not completely removed by the various repair pathways before a new UV-B exposure cycle (next day).

Residual DNA damage was found in early morning samples by Jeffrey et al.

[11] in bacterioplankton from the Gulf of Mexico and in Antarctic marine bacteria and phytoplankton [89]. The CPD levels found at the end of summer in Antarctic waters seemed to support the presence of an unrepairable fraction: significant residual CPD levels were detected despite low incident UV-B levels and a deeply mixed water column which further decreased the mean UV-B experienced by the cells. Incubation experiments also showed that CPDs induced during morning hours were not removed by photoreactivating light during the afternoon (UV-B excluded, UV-A and PAR admitted, Buma, unpublished results) or by any other repair process (i.e., dark repair). Similar results were found in the temperate marine plankton assemblages from Bahia Bus-tamante, Argentina [96]. In fact, residual DNA damage has been reported for every location where CPD abundance was studied. In the Gulf of Aqaba for instance (Figure 7), picoplankton size fractions retained residual DNA damage (between 14 to 43 CPD MB"1) at the end of the night despite the fact that the number of CPDs decreased during the dark. This residual damage suggested that dark repair processes were not able to remove all CPDs. For a typical Synechoc-cocus or bacterial cell containing circa 2.1 x 10~15 g DNA per cell [101], this would imply that between 50 and 160 CPDs per cell were still present, blocking DNA replication and cell division. Quaite et al. [34] found that at low initial dimer frequencies (less than circa 30 CPD MB-1) alfalfa seedlings did not use excision repair. This level (30 CPD MB-1) closely matches the residual levels that were detected in the morning samples in the Gulf of Aqaba. Another suggestion to explain the residual damage levels is that DNA damage is not uniformly distributed over all cells, but that most of the damage accumulates in a few heavily damaged cells, which are or become incapable of repair. These cells would then lose viability and eventually disappear by lysis [102].

Laboratory studies have shown that UV-B exposure causes loss of viability in marine diatoms [103-105]. Indeed, the lack of repair as reported in a number of field experiments suggest the presence of non-viable, CPD containing cells in situ, sometimes further supported by the low assimilation rates for photosynthesis, e.g., the Bahia Bustamante area, where residual CPD levels were extremely high [53]. In general, even within a population subjected to identical UV-B doses, DNA damage is not evenly distributed over cells. Buma et al. [10] showed a clear non-uniform distribution of CPD specific fluorescence in a population of diatom cells, judging from the high standard deviation of the mean for CPD specific fluorescence, using immunochemical CPD labeling in combination with flow cytometry (Figure 9). When extrapolating this to the field, most of the damage would accumulate in a limited number of cells. Because cell division cannot be completed until all the damage is repaired, part of the population might thus replicate in a normal way, while damaged cells eventually die and disappear by lysis. If viability loss plays a role, residual CPD levels would not only be determined by viability loss rates, but also by the residence time of non-viable cells in the water column. For example in the Argentinean Sea, residual CPD levels were very high, whereas CPD induction rates were not extreme. These high initial levels, therefore, might have been caused by the accumulation of non-viable cells in the water column over a prolonged period.

Figure 9. Dose response curve for CPD specific fluorescence against the UV-B dose in Cyclotella sp. cells, measured with flow cytometry. Filled circles: G1 cells, open circles: G2 cells. Error bars, standard deviations of the mean for G1 and G2 cells (at least 4000 cells analysed per condition/cell cycle stage). [Redrawn form Buma et al. 10.]

Figure 9. Dose response curve for CPD specific fluorescence against the UV-B dose in Cyclotella sp. cells, measured with flow cytometry. Filled circles: G1 cells, open circles: G2 cells. Error bars, standard deviations of the mean for G1 and G2 cells (at least 4000 cells analysed per condition/cell cycle stage). [Redrawn form Buma et al. 10.]

Within this context, it would be interesting to consider potential interactions between nutrient/substrate limitation and UV-B stress. Nutrient additions have been reported to decrease the sensitivity of bacterioplankton production to UVR [91]. As has been demonstrated [106] a large fraction of the bacterioplankton in marine waters is metabolically inactive as a result of substrate limitation. Phytoplankton may also experience nutrient (N, P, Fe) limitation in the open ocean or in a post-bloom situation. Suboptimal metabolic activity in these cells would hamper DNA repair, and would thereby contribute to rapid accumulation of DNA damage in these cells. In turn, damage accumulation would then decrease viability in this fraction of the community. The low repair rates and residual DNA damage levels (morning samples) in combination with low growth estimates in many bacterioplankton field studies seem to support this hypothesis.

9.5.4 Effects of varying ozone concentrations

So far, very few studies have addressed the effects of ozone depletion on DNA damage accumulation in marine organisms. Malloy et al. [107] followed DNA damage in Antarctic ichtyoplankton with ambient UV flux during austral spring. They showed that natural levels of UV-B during ozone depletion caused measurable damage to multicellular organisms occupying higher trophic levels in the Antarctic ecosystem. In particular, icefish eggs were shown to be vulnerable to CPD accumulation. Furthermore, they showed that CPD concentrations in icefish eggs closely followed the daily UV-B flux, suggesting that damage was repaired readily within a day, in accordance with the findings of Vetter et al. [99] for anchovy eggs.

More recently, Maedor et al. [108] followed daily levels of DNA damage in planktonic microorganisms incubated under ambient solar conditions, including ozone fluctuations, at Palmer Station, Antarctica. Although the patterns are complex, there does appear to be a relationship between changes in incident solar radiation caused by ozone depletion and DNA damage in plankton. Huot et al. [85] developed a model of DNA damage induction and repair in bacterio-plankton in mixed and non-mixed environments. Also, effects of ozone depletion in mixed and non-mixed systems were incorporated. It was found that ozone thickness caused the largest effect on DNA damage accumulation, when compared with the effects of mixing, DOM concentration or chlorophyll concentration [85],

Was this article helpful?

0 0

Post a comment