Tim Mousseau – latest Chernobyl paper in International Journal of Plant Sciences:
Oct 05, 2016
Pollen viability is an important component of reproductive success, with inviable pollen causing failure of reproduction. Pollen grains have evolved mechanisms to avoid negative impacts of adverse environmental conditions on viability, including the ability to sustain ionizing radiation and repair DNA. We assessed the viability of 109,000 pollen grains representing 675 pollen samples from 111 species of plants in Chernobyl across radiation gradients that spanned three orders of magnitude. We found a statistically significant but small and negative main effect of radiation on pollen viability rates across species (Pearson’s r = 0.20). Ploidy level and the number of nucleate cells (two vs. three) were the only variables that influenced the strength of the effect of radiation on pollen viability, as reflected by significant interactions between these two variables and background radiation, while there were no significant effects of genome size, pollen aperture type, life cycle duration, or pollination agent on the strength of the effect of radiation on pollen viability.
Most organisms are susceptible to environmental perturbations—such as climate change, extreme weather events, pollution, changes in nutrient availability, and changes in ionizing radiation levels—but the effects of such perturbations on individuals, populations, and ecosystems are variable (Candolin and Wong 2012; IPCC 2013; Møller and Mousseau 2013). In order to better understand these effects and to predict how a given species would respond to environmental disturbances, a study of the specific effects at different stages of organisms’ life cycles is required. Since reproduction is a key phase in the life cycle of any organism, reproductive effects are of particular interest. In the case of the effects of ionizing radiation, the negative consequences for reproduction in response to acute irradiation have been studied for decades and are well established (review in Møller and Mousseau 2013). However, the effects of long-term chronic exposure to low dose radiation are poorly understood.
Pollen grains are susceptible to the effects of environmental perturbations, which can have significant negative consequences for plant reproduction through pollen limitation (Delph et al. 1997; Ashman et al. 2004). Potential negative environmental effects include those resulting from elevated levels of ionizing radiation (Koller 1943). Therefore, plants have mechanisms to protect themselves from such effects, such as DNA repair, bi- or trinucleate cells, or redundancies in the genome resulting from duplications.
The area around Chernobyl in Ukraine has proven particularly useful for studying the effects of radioactive contamination on ecological and evolutionary processes at a large spatial scale. The Chernobyl nuclear accident in April 1986 led to the release of between 9.35 × 103 and 1.25 × 104 petabecquerel of radionuclides into the atmosphere (Møller and Mousseau 2006; Yablokov et al. 2009; Evangeliou et al. 2015). These radioactive contaminants were subsequently deposited in the surrounding areas of Belarus, Russia, and Ukraine but also elsewhere across Europe and even in Asia and North America. The pattern of contamination is highly heterogeneous, with some regions having received much higher levels of radionuclides than others, owing to atmospheric conditions at the time of the accident (fig. 1). To this day, the Chernobyl area provides a patchwork of sites that can differ in radioactive contamination level by up to five orders of magnitude across a comparatively small area. Even decades after the accident, the amount of radioactive material remaining around Chernobyl is enormous (Møller and Mousseau 2006; Yablokov et al. 2009).
Fig. 1. Map of the distribution of radioactive contamination in the Chernobyl region, with pollen sampling locations marked. Adapted from DeCort et al. (1998).
Because of the unprecedented scale and global impact of the Chernobyl event, it is not surprising that it generated significant interest in both the scientific community and the general public. As a result, studies have been conducted to assess the consequences of Chernobyl for human health and agriculture as well as its biological effects, ranging from the level of DNA to entire ecosystems. Since ionizing radiation has long been well established as a mutagen (Nadson and Philippov 1925; Muller 1950), a large proportion of the research effort has focused on examining changes in mutation rates in areas that have been radioactively contaminated to different degrees as a result of the accident. Although there is considerable heterogeneity in the results of these studies, most have detected significant increases in mutation rates or genetic damage following the Chernobyl disaster, with the rates remaining elevated over the following 2 decades (reviewed in Møller and Mousseau 2006). For example, the mean frequency of mutations in Scots pine (Pinus sylvestris) is positively correlated with the level of background radiation, and it is 10 times higher in contaminated areas compared with control sites (Shevchenko et al. 1996). A study of Scots pine seeds detected elevated mutation rates within the exclusion zone over a period of 8 yr following the accident (Kal’chenko et al. 1995). In wheat (Triticum aestivum), the mutation rate was six times higher in radioactively contaminated areas compared with controls (Kovalchuk et al. 2000). Likewise, the frequency of chromosomal aberrations in two varieties of wheat grown within the Chernobyl exclusion zone 13 yr after the disaster was elevated compared with the spontaneous frequency of chromosomal aberrations in these cultivars (Yakimchuk et al. 2001). The levels of chromosome aberrations in onions (Allium cepa) were also positively correlated with the intensity of radioactive contamination in plants grown 20 yr after the accident (Grodzinsky 2006). Therefore, there is considerable evidence showing increased mutation rates in plants in the most contaminated sites (Møller and Mousseau 2015).
On the basis of the results of these studies, one might expect that a similar relationship between radiation level and the frequency of abnormalities would be seen in pollen. Indeed, Kordium and Sidorenko (1997) reported that the frequency of meiotic anomalies in microspore formation and the frequency of pollen grain viability was reduced in 8%–10% of the 94 plant species studied as a function of the intensity of gamma radiation 6–8 yr after the accident. In violets (Viola matutina), the proportion of viable pollen was negatively correlated with background radioactive contamination (Popova et al. 1991). While it is evident that plants differ in their susceptibility to ionizing radiation, the reasons for this variation are not entirely clear. It is likely that some species develop tolerance and/or resistance to mutagenic effects of radiation to a greater extent than others (Baer et al. 2007). For example, pollen of silver birch (Betula verrucosa), which grows in areas contaminated by the Chernobyl accident, showed elevated DNA repair ability compared with pollen from control areas, consistent with adaptation or epigenetic responses to increased radiation (Boubriak et al. 2008). There are also indications that genome size might affect the response of different species to radiation. Among the plants studied by Kordium and Sidorenko (1997), the rate of pollen viability decreased with increasing radiation to a higher degree in plants with smaller genomes (Barnier 2005), although the actual mechanism remains unknown. One potential explanation is that a larger genome might contain multiple copies of some genes as a result of duplication, rendering mutations in one of these copies less deleterious than if there were only a single copy present, although this explanation may not universally apply (Otto 2003).
In order to assess the effects of radioactive contamination on plant reproduction and to further assess species-specific differences in the effects of ionizing radiation on pollen viability, we analyzed pollen samples from plants growing in the Chernobyl region. We expected that the effects of radiation would differ among species, with some plants showing higher pollen inviability rates than others as a result of elevated radiation levels. A second objective was to test whether observed differences in pollen viability rates could be attributed to differences in phenotype among species, with possible explanatory factors including pollen size, the number of pollen apertures, ploidy, genome size, bi- or trinucleate cells, life span (annual vs. perennial), and pollination agent. We hypothesized that each of these factors could be related to the plants’ ability to resist or to tolerate radiation-induced mutations. Pollen size, genome size, and ploidy are all related to the amount of DNA and the number of copies of genes contained in the pollen grain. Because the pollen aperture—as the site of pollen germination—could be particularly susceptible to radiation-induced damage, we included the number of apertures as a potential explanatory variable. Furthermore, whether a plant is annual or perennial is related to individual longevity and, consequently, to the number of mutations that can accumulate over its lifetime as well as to the number of generations from the time of the Chernobyl accident until the time of sample collection. This may be particularly relevant for plants, given that germ tissue is derived from somatic tissues during each reproductive event as opposed to most animals, in which germ cells terminally differentiate very early during embryonic development (Buss 2006). Pollen viability depends on the ability of pollen to assess the integrity of its DNA and to repair the DNA of the generative nuclei before division (Jackson and Linskens 1980). This process is particularly important for binucleate pollen cells in which this happens during pollen germination, which is in contrast to trinucleate pollen cells, in which the need for DNA repair during pollen germination is less evident. DNA repair efficiency and adaptation of plants to chronic irradiation may also depend on the composition of radiation at the contaminated sites (Boubriak et al. 1992, 2008).
Across all plant species, we found a statistically significant relationship between radiation and the frequency of viable pollen of an intermediate magnitude (Cohen 1988). We also documented significant interactions between species and radiation, radiation and cell number, and radiation and ploidy. However, the significant effect of ploidy disappeared when both ploidy and whether cells were bi- or trinucleate were entered simultaneously in a single model. Most effects were small to intermediate in magnitude, as is commonly the case in studies of living organisms (Møller and Jennions 2002). We emphasize that our study included by far the largest sample size so far reported to detect effects of chronic radiation on pollen viability. However, we also emphasize the limits of our study. Many plant species could not be included simply because we could not locate multiple flowering specimens during our fieldwork. These and other sampling limitations reduced the number of pollen grains and the number of species that could be included.
Species differ in their susceptibility to radiation, as demonstrated for birds at both Chernobyl and Fukushima (Møller and Mousseau 2007; Møller et al. 2013; Galván et al. 2014), and in terms of adaptation to radiation (Galván et al. 2014; Møller and Mousseau 2016; Ruiz-González et al. 2016). The observed interspecific differences in radiation effects reported here for the proportion of viable pollen could be due to adaptation to radiation through tolerance of radiation-induced mutations or through induction of increased DNA repair in organisms living in contaminated areas. Another possibility is that some species are more resistant to radiation because of historical exposure in radiation hotspot areas with high natural levels of radiation (Møller and Mousseau 2013).
We observed a significant relationship between the proportion of viable pollen and the interaction between ploidy and radiation. Such a finding might suggest that resistance to deleterious effects of radiation is based on redundancy in the genome, where species with higher ploidy levels have an advantage if they have multiple copies of a given gene. We failed to detect an effect of selected physical attributes of pollen grains—such as genome size, pollen size, and aperture type—on the susceptibility of pollen to radiation. Furthermore, whether a plant was annual or perennial or whether it was insect or wind pollinated did not affect the proportion of viable pollen. Finally, whether plants produced bi- or trinucleate pollen had a significant effect on pollen viability, and the interaction between radiation and cell number was also significant.
While we confirmed the general finding of Kordium and Sidorenko (1997) that in approximately 10% of species the proportion of viable pollen is negatively correlated with radiation level, we were unable to reproduce their findings with respect to the overall magnitude of this effect. Our observed effect size was much smaller, and the slopes for individual species differed significantly from those reported by Kordium and Sidorenko (1997). Because more than 10 yr have passed between the two studies, we suggest that a change in radiation effects has taken place over time, for example, as a result of adaptation or accumulation of mutations. Another possible explanation for the discrepancy has to do with sample size, since our study included a much larger number of pollen samples and sampling locations than the study by Kordium and Sidorenko (1997). These explanations are not necessarily mutually exclusive.
Whereas other studies have demonstrated significant negative effects of radioactive contamination around Chernobyl on mutation rates and fitness in general, our study of pollen viability shows a very small effect, and some species even show positive relationships between pollen viability and radiation that is suggestive of adaptation to increased levels of radiation. However, on the basis of the current study, it is not possible to determine whether the observed heterogeneity reflects evolved adaptive responses or is the consequence of unmeasured selective effects on characters correlated with pollen viability, which could in part explain an overall positive effect of radiation (for a discussion of evolutionary responses in Chernobyl, see Møller and Mousseau 2016). Experimental approaches would be needed to decipher the mechanisms underlying the heterogeneity in plant responses observed here (Mousseau 2000).
The observed variability in susceptibility to radiation is a common finding in studies of the effects of radiation from Chernobyl (Møller and Mousseau 2007; Galván et al. 2011, 2014; Møller et al. 2013). While our results are consistent with earlier findings that DNA repair mechanisms may play an important role in adaptation to life in radioactively contaminated environments—especially for plants, which are sessile and hence cannot move to less contaminated areas—further research is required to test this explicitly. Finally, because of the observed differences in resistance to radiation among species, it is likely that even small overall effects of radiation—such as the one on the proportion of viable pollen described here—can have significant consequences for species composition and abundance at a given location and, therefore, for ecosystem characteristics and functioning.
In conclusion, we have found a statistically significant overall negative relationship between radiation intensity and the frequency of viable pollen in plants growing in contaminated areas around Chernobyl. The magnitude of this effect across species included in our study was intermediate. We only found a significant relationship between the proportion of viable pollen and ploidy × radiation interaction, bi- or trinucleate cells, and bi- or trinucleate cells × radiation interaction. This suggests that DNA repair mechanisms could play an important role for the ability of plants to resist increased radiation, at least when it comes to pollen formation.
We thank Puri López-García for use of a microscope for pollen counts. This work has benefited from the facilities and expertise of the cytometry platform of Imagif (Centre de Recherche de Gif; http://www.imagif.cnrs.fr). We thank Spencer Brown and Mickaël Bourge for their help with the flow cytometry measurements and Srdan Randić for help with pollen counts. Field collections for this study were supported in part by the Centre National de la Recherche Scientifique (France), the North Atlantic Treaty Organization Collaborative Linkage Grant program, the Fulbright program, the University of South Carolina College of Arts and Sciences, and the Samuel Freeman Charitable Trust. Two reviewers provided constructive criticism.
Read full paper at: