IRSN, Institut de radioprotection et de sûreté nucléaire

Search our site :


Contact us :

En Fr

All our expertise to protect you


Research programmes

Scientific/technical achievements of the previous Medium and Long Term Plan (2009-2012)



Scientific results


As for the environmental risks posed by low doses, the research undertaken produced findings which could be used directly to design or improve models that simulate the development and the effects of radionuclides within ecosystems:


1. For soils and sediments, the reservoirs of radionuclides in terrestrial and aquatic ecosystems which are subject to chronic exposure, the reactions of microbial chemical transformations, the influence of the roots of higher plants and even the bioturbation caused by macrofauna influence to various degrees the speciation, transport and transfer of uranium and selenium. These findings have let us identify the accumulation areas within these 'reservoir' compartments, the importance of interactions with the cycle of the natural organic material and the associated residence times.


2. These findings condition the bioavailability of radionuclides as well as the occurrence and intensity of the organisms' exposure. Bioavailability and toxicity depend on the speciation of the element in the environment (or in food), which depends on the quality criteria of the latter and the physiology of the organism (state of membrane surfaces). Various approaches to the modelling of uranium-organism and selenium-organism interactions were therefore published.


3. Once ingested, the contaminant is distributed and biotransformed according to a wide range of metabolic processes specific to the element and/or the species. Only a fraction of the ingested amount is biologically active in terms of toxicity. From an operational perspective, awareness of this 'critical' amount and its biological target marks the transition between the no-effect domain and the toxicity domain. These 'internal load' criteria have been quantified for uranium (and selenium) for various chronic exposure scenarios and various species.


4. The biological effects that are likely to appear in response to exposure to the ingested radionuclide's biologically active fraction can arise at different levels of biological organisation: from the sub-cellular level to the individual level and even the population. Dose-response relationships which are likely to affect the demography of a population of diverse plant and animal species have been modelled for the main functions of an individual. The responses studied are mainly survival at different stages of a species’ life cycle, its reproductive capacity and age of maturity for first reproduction. Sub-cellular research, which is restricted to understanding the basic toxicity mechanisms which would affect these main functions, has been carried out on energy metabolism disorders, the fight against oxidative damage and genotoxicity.


5. The relative contribution of chemotoxicity and radiotoxicity for an element such as uranium varies in line with the sensitivity of the biological model and physiological function studied.


6. The results obtained with invertebrate models show that physiological acclimatisation phenomena, but also genetic selection and adaptation phenomena, occur in population groups exposed to radioactive substances over several generations. Depending on the mechanism put in place, the long-term consequences can be very different, particularly in terms of reversibility. These studies have been conducted with three stressors (depleted uranium, americium-241 and external gamma radiation exposure) and compared with one another.


7. Consideration of the effects of substances in a mixture makes chronic exposure situations which are in situ more realistic. Studying these effects has proven to be complex, requiring a combination of reductionist approaches for the identification of interactions (e.g. controlled environment and binary mixtures) and systemic approaches (e.g. multi-stressor mixtures and integrated 'from the gene to the population' responses) for use on an operational level. The effects of stressors in a binary mixture on a micro-alga revealed processes of synergy (greater toxicity of U in the presence of Cd at a concentration level that has no effect) and antagonism (U and protective effect of Se) in comparison with the growth inhibition obtained for uranium on its own. The interactions of tritium (HTO) and Cu on a micro-alga were also studied. It is thought that the effect of tritium observed on the entry kinetics of copper in the cells can be put down to the increased production of reactive oxygen species (ROS) in the alga exposed to tritium.


8. The ERICA method, which assesses the environmental risk associated with radioactive substances, was adapted to the context of uranium-bearing mining sites by undertaking a parallel and coherent assessment of the chemotoxic and radiotoxic risks for uranium and its daughter products.


9. The data acquired on the link between the physico-chemical speciation of uranium, bioavailability and ecotoxic effects in fresh water was used to put forward a method to derive a predicted no effect concentration (PNEC) guideline value in fresh water. Acute and chronic PNEC grids specific to large physico-chemical domains of watercourses are therefore put forward (areas defined by three quality parameters: pH, alkalinity and hardness). These grids, whose purpose is to be used as an environmental management tool, integrate the most recent findings on this element's bioavailability variation and therefore the variation of its ecotoxicity in accordance with the physico-chemical characteristics of watercourses.



Data for the expert appraisal


  • Suggestion of guideline values for uranium in watercourses to support the specification of an environmental quality standard
  • Expert appraisal on the ecotoxic impact of the SOCATRI incident
  • Expert appraisal on the environmental impact of the Fukushima accident


Greater interaction with society


  • Participation in the multi-expert group (GEP) on mines to assess the impact of former uranium mines on ecosystems.







1. The biokinetics of uranium is not what was expected: The distribution of uranium in tissues does not comply with the usual bioaccumulation models. The research undertaken shows that uranium accumulates in most organs according to a complex process. In a chronically-contaminated rat, uranium concentration levels in the colon appear to increase gradually over time. This complies with the models of the International Commission on Radiological Protection (ICRP). However, the accumulation of uranium observed in other tissues does not follow this pattern. In the kidneys, skeleton, small intestine, brain, muscle, liver and finally the whole body, the contamination profiles are very different (successive waves of accumulation and elimination). In the kidney, the accumulation of uranium extends to the whole of the nephron. Chronic contamination is not therefore equivalent to - in terms of accumulation in tissues and organs - the sum of repeated acute exposures. Nevertheless, the two major organs in which uranium accumulates are the same for low-dose chronic contamination as for high-dose acute contamination: the kidney and bone tissue. As for the dose, the dose calculated is very low, around 10 mGy, and is not traditionally considered to be toxic.


2. Tissue, cellular and sub-cellular distribution of uranium is more heterogeneous than expected: With regards to tissue, we have been able to show that in a chronic exposure situation, uranium spreads throughout the nephron and does not only affect the proximal convoluted tubule, as thought until now. Furthermore, we have shown that uranium tends to be present in the cell nucleus. This is especially noted at the time of exposure of human kidney cells to non-cytotoxic concentrations of uranium and during the chronic exposure of rats to uranium.


3. Standards on the amount of uranium in drinking water aim to protect consumers: The amount of uranium in drinking water is generally around one microgram per litre. However, this amount can be 100 to 1000 times higher in some regions. These uranium levels raise the question of whether the health of human populations living in these areas is being affected. To answer this question, we contaminated animals for nine months with drinking water containing depleted uranium at different concentration levels. We were able to show that the chronic ingestion of uranium at exposure levels that are higher than those in the environment (from < 2 mg/l up to 120 mg/l) does not have any harmful effects on rats, as proven by the absence of clinical signs and histological lesions in all of the organs studied (kidney, liver, brain, intestine, etc.). Therefore, the NOAEL (No-Observed-Adverse-Effect Level) of uranium following chronic ingestion is above 120 mg/l. Based on the results of these experiments, the tolerable daily intake calculated produces a guideline value for humans of 1250 mg/l of uranium in drinking water when based on harmful chemical toxicity in the kidneys. The results of these experiments suggest that the 30 mg/l reference value that the WHO prescribes in its directive on uranium in drinking water is highly protective.


4. Uranium's single target concept must be abandoned: Different LOEL (Lowest-Observed-Effect Level) thresholds can be deduced on the basis of the molecular effects observed: 20 mg/l for the intestine, >10 mg/l for the kidneys, >2 mg/l for the liver, >0.2 mg/l for the brain. These results show that a single target concept has given way to a multi-organ target concept (kidney, liver, brain, gonads). They also indicate that the kidneys are not the most sensitive organ to chronic contamination caused by the ingestion of uranium. The brain appears to be the most sensitive. Surprisingly, these results have shown that there is no dose-effect relationship between the accumulation of uranium and its biological effects. The organs most affected by exposure to uranium are those that have not accumulated it (1 ng/g in the brain, 2.5 ng/g in the liver and 350 ng/g in the kidneys for a drinking water contamination value of 120 mg/l). This finding shows how sensitive the brain is to the chemical toxicity of uranium.


5. Juveniles are more sensitive: Various studies have shown that the chronic ingestion of a small quantity of uranium has subtle and significant biological effects on different metabolic systems such as modifications to gene and protein expression. Our results raise a number of questions. For example, we have shown that uranium contamination is capable of modifying blood vitamin D levels, a hormone associated with bone metabolism, suggesting that this radionuclide could also affect bone remodelling. A recent study has revealed that ingested natural uranium affects bone growth across the width of the femur in growing individuals without, however, resulting in bone disease. These results therefore suggest that specific attention must be given to children at the time of chronic exposure to this radionuclide.


6. The effects observed show that organisms can adapt: Over the past few years, a number of studies have shown that uranium could have multiple but minor biological effects and thereby slightly disrupt the equilibrium of several major metabolic processes such as the metabolism of xenobiotics, cholesterol, vitamin D, acetylcholine, iron, steroid hormones and bone remodelling. These studies have revealed effects on a molecular level: modification of gene expression, protein concentrations and enzyme activity. Nevertheless, these metabolic modifications are not accompanied by the appearance of diseases. The biological effects observed therefore probably reflect the organism's adaptive response to the presence of internal contamination.


7. Some of the effects observed are irreversible: Not every biological effect poses a threat to health. A biological effect is usually reversible when it is situated within the limits of homeostasis. It can simply display the normal 'adaptive' response of the cell, tissue or organism to a stimulation and, in our case, to exposure to uranium. To check this hypothesis, an experiment using a so-called reversion model was carried out. In this study, rats were contaminated with depleted uranium for six months and then subject to a reversion period of three or six months. The results show that for the liver, certain modifications of gene expression caused by depleted uranium are irreversible like that of the enzymes associated with the metabolism of cholesterol and vitamin D. However, other modifications of gene expression in the brain and kidneys are reversible. Our study shows that for the first time, depleted uranium has irreversible biological effects in rats that the organism stores 'in its memory'. These effects could be passed down to offspring via epigenetic mechanisms. 


In conclusion, uranium can interfere in living organisms' different metabolic systems. However, the biological effects of this radionuclide are minor and do not result in acute toxicity and do not therefore cause phenotypic change in an animal's life cycle. On the other hand, the various biological effects observed on a molecular and genetic level following contamination could have an irreversible genetic imprint following internal and chronic exposure to this radionuclide. To check this hypothesis, internal uranium contamination experiments will have to be undertaken over several generations. The molecular effect observed in the first generation could result in phenotypic damage in the second or third generation of individuals. The multi-generational model should therefore enable us to check whether the hypothesis of a 'genetic imprint' caused by the chronic ingestion of a low level of uranium is valid.


Number of publications: 39 original articles, 7 journals and 7 book chapters




1. At the time of low-level chronic contamination, caesium-137 does not accumulate in tissue. Only an increase in the concentration of caesium-137 in tissue, which takes place in accordance with the growth of the animals (first three months), was observed. Caesium-137 spreads heterogeneously in the organism. Muscles and kidneys are the tissues in which the greatest concentration is measured while blood and skin are the tissues which are least contaminated.


2. Few biological and harmful effects are observed during experiments on rodents. In most cases, these effects remain limited and do not trigger diseases. This has been observed for the immune system, the haematopoietic system, the central nervous system, the digestive system and steroid metabolism.


3. Nevertheless, the uncovering of effects on the cardiovascular system (modification of arterial pressure and plasma cardiac markers) is an initial indication of a possible health effect caused by the chronic ingestion of caesium-137. Furthermore, modification of the metabolomic profile of the plasma of the contaminated animals lets us distinguish between the groups of animals contaminated by low caesium-137 concentration levels and non-contaminated animals.


4. The cardiovascular effects observed following caesium-137 contamination under the Envirhom programme have led to the continuation of studies in this area. Multiple factors contribute to the appearance of cardiovascular diseases and at low dose levels it is difficult to separate the risk posed by radiation from other risk factors. It was therefore necessary to implement experimental in vivo and in vitro studies in order to identify the pathophysiological mechanisms involved after low dose exposure. In vivo studies shall identify the effect of chronic caesium-137 contamination at low doses on the development of atheromatous disease, which is the main cause of morbidity and mortality in industrialised nations. The in vitro studies form part of the work of the European DOREMI network, whose aim is to identify the mechanisms involved after umbilical cord endothelial cells are exposed to low doses. This study shall make it possible to differentiate between the effect of caesium-137 radiation at different doses and caesium-137 contamination at different concentration levels.

Number of publications: 17 original articles, 2 journals and 7 book chapters





1. Biokinetic studies have confirmed the specific accumulation of strontium in bones. These studies have also shown that the accumulation of strontium in bones occurs far more quickly during the bone growth phase, in other words, among juvenile animals. The doses of radiation produced by this contamination process are low, around 10 mGy after 20 weeks of chronic ingestion of 75 Bq of 90Sr per day and per animal.


2. The location of strontium in the bones suggests that two physiological systems could be affected: the haematopoietic system, situated in the bones, and bone physiology. Initial studies revealed that strontium contamination did not affect the haematopoietic system.


3. On the other hand, there was a modification in the bone's physiology, namely, an increase in bone resorption, which did not, however, result in the appearance of any diseases among the animals during the contamination period studied.


Number of publications: 2 original articles


Link to other IRSN research programmes


Envirhom-santé is connected to a clinical research programme (EPICE) and epidemiological studies developed at IRSN in the field of low exposure doses:


1. A research programme called EPICE ('Assessment of diseases induced by caesium-137 contamination'), which is a prospective clinical study involving a population of around 18,000 children living in areas in Russia contaminated by the fall-out of the Chernobyl accident. The aim of this programme is to identify through adapted diagnostic examinations non-cancerous diseases (heart rhythm disorders in the first phase of the study) affecting these children who ingest on a daily basis food grown on soil contaminated by caesium-137 and to assess the extent to which their bodies have been contaminated with caesium. 


2. Epidemiological studies carried out among workers (AREVA, EDF, CEA) involved in the fuel cycle (from the uranium mine to fuel reprocessing) and who are particularly exposed to uranium.



Data for the expert appraisal


  • Participation in national and international working groups
    > Afssa (Anses): recommendations on uranium concentration levels in drinking water in France (2009)
    > WHO: revision 'Drinking water guidelines' (2009-2010)


  • Drinking water recommendations (pending)
    > Uranium: calling into question of the WHO value defined on the basis of uranium's chemical toxicity for kidneys
    > Tritium: contribution of data based on the type of tritium (HTO or OBT) for the WHO value


  • Specification of the thresholds at which the various effects, whether harmful or not, appear
    > Response to the current issue regarding weak signals in the case of environmental health between alerts and decisions


Greater interaction with society


  • Participation in CLIs (Local information committees)
    > Cadarache CLI every two years (September 2007, November 2009, October 2011)
    > ITER CLI: Tritium problems (May 2012)


  • General public
    > Science festival: November 2008 (presentation and workshops)
    > Secondary schools: International radiation protection days (April 2009; February 2010; February 2012) (presentation and workshops)


  • Specialised audience
    > Lavoisier publication: Toxicologie nucléaire environnementale et humaine co-coordination IRSN-CEA - Environmental and human nuclear toxicology, jointly coordinated by IRSN and CEA (published in September 2009) 
    > Scientific website: paper on low doses (October 2009)


Send to a friend

The information you provide in this page are single use only and will not be saved.
* Required fields

Recipient's email:*  

Sign with your name:* 

Type your email address:*   

Add a message :

Do you want to receive a copy of this email?