In my opinion, you can use passive sampling technique for monitoring of water body that individual fishes live in it. By this sampling technique you should be use several absorbent substances such as Triolein oil as endocrine distrupters absorbent for a long time in water bodies. This absorbent substance can work in stead of fish tissues in absorbing pollutants. By this technique you can determine how much pollutants and which of them can absorb to fish tissues. For additional information refer to "Passive sampling" or SPMDs papers about pollutants absorption.

But... If you wish to still use a fish model you should first determine which species would most likely serve your purpose and then determine which organ(s) to choose initially ... But freeze all tissues. They just may come on handy later on. Never throw anything away!!!

I agree with Ebrahimi especially if you are looking for chemical approaches. However if you want to make sure chemicals are causing biological effects (i. e. being pollutants rather than contaminants) I would suggest you a set of chemical-biological measurements. There are different biomarkers, some are chemical specific and some are not. Some biomarkers are responding in short term and others in long-term exposures. Finally, there are biomarkers of exposure (biologically-low contaminant exposure) and biomarkers of effects (biologically high contaminant exposure).

Thus I would say that if you combine some measurements in water, bioaccumulation in gall bladder of organics and in liver of metals to a group of maybe 5 biomarkers, you will be able to likely say which contaminant is in the water, which is entering in fish, for how long it is entering and which sort/extent of effects it is causing.

You might found interesting taking a look for this subjects. There is several methods to combine chemical-multibiomarkers measurements to discriminate specific sort of contaminant and effects.

The question is also a bit vague since you'll want to specify what timescale you are wanting to observe/measure. With that said, you may want to also consider doing chronic and/or acute tests on sediment and soil, especially if you're looking at heavy metals. You may consider also testing the tissue for accumulation using fluorescence.

If you are looking for non-destructive methods (i.e. where the fish can be examined repeatedly without having to kill them), it's a bit more challenging. A general measure of "stress" might be to look at cortisol levels in the surrounding water. For examples, see http://www.cefas.co.uk/publications/files/3402.pdf, and http://www.bama.ua.edu/~rlearley/Wong_2008.pdf

If you can bleed your fish, then depression of ALA-D activity in blood is an indicator of lead exposure (and some other heavy metals). See Krajnovit-Ozretik and B. Ozrettic, 1980, Mar. Ecol. Prog. Ser. 3:187-191 . http://www.int-res.com/articles/meps/3/m003p187.pdf.

There must be other options, too, though I can't think of any at present.

Cheers,

Anne

For behavioural effects of EDCs, a variety of tests are listed in Clotfelter et al 2004

http://training.fws.gov/EC/Resources/pesticides/Aquatic%20Effects/EDCsbehavior.pdf

Vitellogenin production in males is the classic biomarker of estrogenic exposure, but it can be challenging to collect enough blood to assay, without killing the fish.

For endocrine disrupting, you can also use transgenic medaka fish to monitor endocrine disrupting level in water. Further reading:

Using a Sensitive Japanese Medaka (Oryzias latipes) Fish Model for Endocrine Disruptors Screening

or http://www.plosone.org/article/info:doi/10.1371/journal.pone.0064334

Possible methods for measuring bioaccumulation would depend on what you seek to measure, and under what conditions.

Do you seek to establish rates of accumulation/depuration in fish exposed to known ambient or dietary concentrations, and measure the scale of variation between equivalent individuals exposed to these sources? Sequential sacrificing would require a large number of matched fish with identical histories.

Or do you wish to see how tissue concentrations in particular individuals respond to variations in exposure over time or space? My guess (and only that) is that few fish produce deciduous tissues that could be used to follow pollutant burdens in an individual over the short term. Sampling a living tissue repeatedly would probably require a large species and small samples if the individual’s pollutant kinetics were not to be affected by the repeated handling.

Each of the above both schema assume lab cultures exposed to controlled doses, though your question suggests that you may also be seeking methods to follow/integrate the exposure signal from both the water and the diet.

Targeting samples of individuals according to size or age groups and according to sexes over time i.e. across seasons will give you an insight into the differential kinetics of pollutants between groups that constitute a population.

One major problem is that passive samplers have physico-chemical properties which give them specific affinity for hydrophilic vs hydrophobic, neutral vs polar etc... compounds as a consequence, you will likely need several samplers. You should give more details experimental or in the field, large water bodies, ponds...

Best,

Xavier

Thank you for your answer. Alan, I am wondering if accumulated contaminants could be passed to the eggs in females, as it is the case in birds. Of course using eggs would first require validation to investigate how such contaminants are transferred, what is the fraction transferred, etc. But could it be possible to use eggs to measure contaminant burden over time?

Personally, I will also use passive sampling technique. And more specifically PDMS coating glass fiber. The general concept are explain in "Oomen, A. G., P. Mayer, et al. (2000). "Nonequilibrium Solid-Phase Microextraction for Determination of the Freely Dissolved Concentration of Hydrophobic Organic Compounds:  Matrix Effects and Limitations." analytical Chemistry 72(13): 2802-2808." But you can find other information in "Jahnke, A., M. S. McLachlan, et al. (2008). "Equilibrium sampling: Partitioning of organochlorine compounds from lipids into polydimethylsiloxane." Chemosphere 73(10): 1575-1581."

However, you will have some issues to solve. 1/ PDMS work only with hydrophobic contaminants, not heavy metal or other polar contaminants, 2/ the dimensionning of the volume of the PDMS is very important. If the volume is too big, you will not measure the real concentration of contaminants in your fish but you will overestimate as you will deplete the matrix. 3/ timing, the passive sampling work only if PDMS and fish flesh reach equilibrium at the sampling time. 4/ put the PDMS coated glass fiber into the fish.

Nevertheless, if you succeed to solve these issues, you will have as much sampling time you want (as you can put a lot of glass fiber in your fish without killing it) on the same fish. Hope it was usefull.

I would expect you would find quite large inter-individual differences in chemical burdens in fish, even in a laboratory study using a single chemical, so what you propose will not be easy to do well or interpret. You will need to think carefully about the number of animals needed to ensure statistical robustness, and also ensure you choose a very abundant species if sampling wild fish populations. The Passive samplers provide an integrated sample of the water chemistry over time, but like the fish, you will need some very good analytical chemistry to get useful data. Caged fish obtained from a reference site and deployed in various locations might be one way of investigating chemical pharmacokinetics, and would provide a more controlled approach. This could be done alongside wild fish sampling to gain further insight into objective 2. Good luck!

Pierre,

We have used snail eggs to map toxic metals in different locations. Ostensibly, eggs are a repetitive sample from an individual, and, as such, are highly amenable to measuring variation . Oviposition represents a considerable cost to the parent and the composition of each 'unit' will be under close physiological control. Additionally, fish eggs, like snail eggs, should allow relatively straight-forward determination of their pollutant burdens. The problem, of course, is eggs shed in a single spawning will not easily provide a responsive sentinel of temporal exposure,

We found the snail eggs to have very interesting differences in toxic metal burdens between populations, with considerable variation within a brood which may reflect local conditions (see Beeby A.N. and Richmond, L. 1998. Variation in the mineral composition of eggs of the snail, Helix aspersa between populations exposed to different levels of metal contamination. Environ. Pollut. 101, 25-31) . Depending on the fish species, it may be possible to control spawning so that sequential samples are possible, to follow exposure and accumulation.

Again, the controlled conditions of the lab should allow closer scrutiny of the important variables controlling egg composition and its variation within each spawning.

Thank you Edwin and Alan for insightful replies. Using eggs in fish or snail would indeed be a good way to measure contaminant burden. I know some researcher do that in long-lived birds using one egg.

I am directly interested with this individual variation. As a behavioral ecologist, my biased view is that a non-negligible part of this variation would be imputable to behavioral differences. This would require monitoring and measuring the behavioral variation among individuals, before exposing them to a contaminated environment, and eventually testing their behavior again. The challenge would be also to monitor contaminant levels without affecting the individuals too much. Coming from other fields of research what do you think is the causes of this individual variation in contaminant levels?

Pierre,

I strongly endorse your research aims and efforts to properly quantify the sources of variation. Many published studies are simple ‘handle-cranking’, and too many papers are content to report average pollutant burdens in organisms as simple indicators of their habitat’s contamination. More useful are studies which seek to attribute the variation observed among individuals. Behavioural differences are likely to be very important and I suspect that behaviours that affect the feeding and growth rate of the juvenile fish will be worth looking at first.

For instance, juvenile snails from the same brood show considerable variation in feeding and assimilation rates of Pb. Some individuals are quiescent, feeding infrequently or very slowly, whereas others consume all that is presented to them. Some go through bursts of activity. Not surprisingly, the continual and rapid feeders grow rapidly and are able to reproduce sooner. Such variability within a brood may increase the inclusive fitness of the parent in an unpredictable environment (which a parent may facilitate by producing trophic eggs), and be a strategy selected by local conditions. These differences in growth rates and toxic metal assimilation between populations may represent local ecotypes associated with essential metal availability. As you can see, there appears to be a hierarchy of factors determining body burdens of Pb in snails, from the biochemical level to the population and ecosystem levels, and with considerable interaction between them.

Consequently, looking at intra-population and inter-population variation in assimilation rates in the fish could be very valuable, using both the eggs and juveniles (of increasing age?) as replicates to measure variation. By using second or later generation fish as the parents, taken from cultures in clean and identical conditions, you can ensure that residual pollutant burdens or previous nutritional history are not a source of variation, to allow attribution to any behavioural differences.

Alan,

Thank you once again for stimulating the discussion. I did not read thoroughly your publications on the subject, but I will take them with me on the plane tomorrow. It looks really interesting to me. Lead accumulation in the snails you study is a very promising system to tackle the research questions I outline in a submitted manuscript.

The link between foraging and lead accumulation is interesting because this is a behavior that is known to vary tremendously among individuals, both as a function of environmental variables and life history trajectory but also in a consistent way (some individuals will always be more voracious irrespective of the environmental conditions). Part of these differences in behavior may also result from individuals specializing on different behavioral strategies.

I assume lead may involve both an acute toxic effect, knocking down foraging in snails in environments where the lead concentration is too high? It also involves increased physiological costs, due to detoxification or because lead decreases digestion efficiency. Are the physiological and life history 'costs' of lead accumulation studied in these snails?

Captivating topic. A little bit on the side, have you analyzed the differences on individual "innate behavior" in relation to the variability you can found? In example, if the differences on the bioaccumulation are related to individual foraging behavior, and if this foraging behavior is related to a coping style, thus, should we expect differences between "bold" vs "shy" in their susceptibility to be affected toxicants (and subsequent physiological impact) due to the "personality"?

Pierre,

You raise some interesting questions about the toxicity of Pb, especially amongst the invertebrates frequently used to monitor environmental contamination: (i) only very high dietary concentrations (far higher than in most grossly polluted ecosystems) affect rates of food consumption in snails; (ii) of the major toxic metals (Zn, Cd, Cu, Hg) Pb is invariably reported to have the lowest impact in both aquatic and terrestrial invertebrates.

Compared to these other metals Pb has three other fascinating and contrasting properties: (a) it does not appear to induce a metallothionein; (b) despite this, and its relatively low toxicity, it is one of the most closely regulated of these metals, even in populations never exposed to significant concentrations (c).

This prompts so many questions about the selective pressures that high ambient Pb might exert…..

See for example:

Dallinger, R., Berger, B., Triebskorn-Kohler, R, Kohler, H., 2001. Soil biology and ecotoxicology. In The Biology of Terrestrial Molluscs (ed. Barker, G.M.). CABI Wallingford, Oxford pp 489-525

Laskowski, R., Hopkin, S. P., 1996. Accumulation of Zn, Cu, Pb and Cd in the garden snail (Helix aspersa): implications for predators. Environmental Pollution 91, 289-297.

Swaileh, K. M., Ezzughayyar, A., 2001. Dose-dependent effects of dietary Pb and Zn on feeding and growth rates of the landsnail Helix engaddensis. Ecotoxicology and Environmental Safety 50, 9-14

Zaldibar, B., Cancio, I., Soto, I., Marigómez, I., 2008. Changes in cell-type composition in digestive gland of slugs and its influence on biomarkers following transplantation between a relatively unpolluted and a chronically metal polluted site. Environmental Pollution 156, 367-379.

Patricia,

This is exactly what I am aiming to investigate. Shyness-boldness or other personality traits are not always 'innate' but through potential interaction between personality and bio-accumulation or exposure levels, contaminants could definitely impact the individual behavioral variation within populations. We currently submitted a framework for publication, but I am still looking for a good study system (along with collaborators!) to provide proof of concept.

To my knowledge, studies dealing with this topic are still very limited. Do you have any in mind?

Hi Pierre,

Looks like that's a rising field to integrating behavior syndrome perspective to more physiological studies. We are going to that under the frame of fish health status, but still defining protocols and experimental design. Koolhaas 2008 has a nice review www.sciencedirect.com/science/article/pii/S0889159107003054 related to that, and in this one is showed its utility integrated with gene expression analysis http://www.plosone.org/article/info%3Adoi%2F10.1371%2Fjournal.pone.0005314 . However it's still remaining the issue of proper non-invasive methods for repeated measures. Will be nice to read your framwork when available.

Hi Patricia,

The coping style concept is interesting for studies interested in individual variation in physiology. However, it is often oblivious to the ecological implications of individual variation. In our 'framework' we are suggesting the use of dynamic budget theory and the recent developments on the adaptive explanations for individual variation to model the interactions between contaminants and individual behavioral variation.

However, I must admit I have homework to do when it comes to monitoring contaminant exposure or accumulation! The original question was exactly aiming to create a discussion on this topic.

Alan,

Thank you for suggesting these interesting references. I enjoyed reading them and it provokes more troughs...

Lead accumulation is also known to affect individual behavior in a permanent way in human (higher risks of depression, changes in personality scores, anxiety). Do you think that lead could interact differently with snails having faster or slower life history trajectories? Such differences are associated with differing allocation patterns of resources to soft tissue vs shell. I am wondering if such an effect could be present at doses that do not have any detectable toxic effects. Maybe the foraging behavior, diet or life history of populations with differing evolutionary histories of lead exposure have been compared?

I also read some of your own work and was astonished by the intra-population/treatment variability in lead accumulation (in one of your 2003 study validating snails as sentinel species). Do you know whether similar patterns have been reported for marine snails (for example Littorina sp)? Could life history differences contribute to the intra treatment variation in lead content you documented?

Lead also has known effects on reproduction. It could potentially select for some particular life history... If for example lead accumulation offset the reproductive benefits of a large body size one may expect lead exposure to select for life histories associated with small body size and potentially higher reproductive effort early in life.

Pierre,

I am currently on vacation. I will give some thought to your questions and get back to you as soon as possible, hopefully within the week.

Pierre,

Although I have not analysed marine invertebrates myself, the literature does suggest that high within-sample variability in toxic metal burdens (where it is fully reported) is commonplace. (There is a couple of books that give an overview of metal ecotoxicology in marine (Philip Rainbow) and terrestrial invertebrates (S P Hopkin). Although they are both quite old, the first also reviews their long history in biological monitoring. I can send you the details if you need them).

Your research could be valuable because intra-population variation rarely attracts extended comment: many authors, myself included, have been content to declare a significant difference as their within-sample variance is less than that between sites or treatments. The need is to apportion the sources of this within-sample variation, and create an experimental design to allow this.

Part of this variability comes from analytical losses, background contamination, instrument operation/performance, all of which can be measured. Beyond these analytical sources, differences are attributable to environmental factors which govern the pollutant’s availability (salinity, pH, temperature, interactions with other metals or organic matter, etc) and between organisms according to their gender, age, physiological condition etc. One of the earliest patterns observed in marine molluscs (by C R Boyden in the 1970s) was the effect of body size of on metal concentration, and the apparent consistency of its scaling factor, the biological significance of which has been the subject of considerable debate over the years. Equivalent patterns have also been claimed for terrestrial gastropods. My understanding is that few studies that have examined behaviour other than feeding rates (more often, filtration rates) on body burdens in marine molluscs. Notably, some claim that rates of uptake by terrestrial and aquatic sentinel species are the best indicators of ambient concentrations.

Here we have to distinguish possible routes of uptake, and especially in marine molluscs, between cutaneous and digestive uptake. A full description of the sources of intra-population variation would need to quantify (or control) these, especially if the dominant route depended on the behaviour of the species.

Your questions highlight what is likely to be central to the issue of Pb assimilation - growth rates. To answer them in turn:

(i) possibly – different snail populations do have different assimilation rates, and these may well reflect alternative resource allocation patterns. Certainly, the wide variation in growth rates seen within single broods of Cantareus would allow experimentation to test this and perhaps Littorina are the same.

(ii) Few toxic effects are seen in Cantareus at doses within the range of all but the most grossly polluted environments. Growth rate is easy to measure but others include time to first reproduction, brood size, egg mass, respiration rate and scope for growth. Histological or biochemical assays make greater demands on time and resources and generally are less easily quantifiable for measuring variation.

(iii) Yes foraging history may differ and differences in body size may be important. This could be examined with a feeding system that quantifies the dietary route (I suggest keeping water concentrations close to zero in early experiments), to measure consumption and assimilation, as well as feeding duration and range. With such a system feeding preferences could also be measured.

(iv) If Pb did offset the benefits of a large body size, it would be fascinating to find how it exerted such a selective pressure. One indication, of course, would be to compare intra-population (and intra-brood) variation between habitats: we would expect to see less variation in the most contaminated populations, as well as a smaller body size at first reproduction. So many experiments (and insights) could follow if such differences were found!

Really interesting topic and great discussion.

I like Alan's comment; To complement his point of view, I would recommend Pierre-Olivier to go through the following ms concerning the delineation of Pb contamination pathway in 2 Pectinid species and the effect of food quality and quantity on assimilation of some metals in other bivalves (unfortunately not studied yet on Pb):

Metian M, Warnau M, Oberhänsli F & P Bustamante (2009) Delineation of Pb contamination pathways in two Pectinidae: the variegated scallop Chlamys varia and the king scallop Pecten maximus. Science of the Total Environment 407 (11): 3503-3509.

Hédouin L, Metian M, Lacoue-Labarthe T, Fichez R, Teyssié J-L, Bustamante P & M Warnau (2010) Influence of food on the assimilation of selected metals in tropical bivalves from the New Caledonia lagoon: qualitative and quantitative aspects. Marine Pollution Bulletin 61 (7-12): 568-575.

In addition, I also recommend the reading of Rainbow's work.

Hi Marc,

Thanks for pointing out your publications here.

Reading through the first one, I wonder whether you observed inter-individual differences in the rate of filtration, plankton consumption or burying behaviour within each species? Going back to Alan's work on accumulation between hard and soft tissues, are these different species accumulating Pb differently in these two compartments? I am interested by the differences in coefficient of variatoin : 16 vs 41 % in each species if you use the values reported in the abstract the tables also hint at a greater variability of Pb accumulation in C. pecten. Do you have thoughts on why this pattern would happen?

Also could you provide me with more details on 'Rainbow'? I get mostly things on a rock band when I search for this!

Philip S Rainbow, Head of Life Sciences Department, Natural History Museum, Cromwell Road, London.

This web site lists publications:

http://www.nhm.ac.uk/research-curation/about-science/staff-directory/life-sciences/p-rainbow/index.html

Cheers,

Anne

Dear all,

Thank you for this great discussion. Here is the final version of a manuscript we published in Animal Behaviour with Raphael Royauté.

https://www.researchgate.net/publication/258173345_Contaminants_as_a_neglected_source_of_behavioural_variation

Happy new year,

Pierre-Olivier M.

Article Contaminants as a neglected source of behavioural variation

Depending on your test organism measurements of chemicals in hair or feathers have been sucesful in determining levels of exposure both repeatedly and between individuals non-destructively.

Could we expect that fish scales can also be good samples to that type of tests?

Hi Paul and Patricia,

I am aware that hair and feathers are a possible way to measure these repeatedly, but I did not look more into that. Would you have a reference that would review the technical aspects of it?

I know some researcher also measure heavy metal accumulation in otoliths. Seems like a very promising tool! However, it is a bit like dataloggers: you need to capture the animal before you can retreive the data...