ACTA BIOLOGICA SLOVENICA LJUBLJANA 2002 Vol. 45, Št. 2: 3 - 14 Sprejeto (accepted): 2002-05-27 Food choice experiments with cadmium nitrate dosed food in terrestrial isopod Oniscus asellus (Crustacea) Poskusi izbire hrane z različno vsebnostjo kadmijevega nitrata na kopenskem enakonožcu Oniscus asellus (Crustacea) Ulf Iskender KASCHL1, Primož ZIDAR2, Jasna ŠTRUS2, Volker STORCH1 1 Institute for Zoology, University of Heidelberg, Im Neuenheimer Feld 230, 69120 Heidelberg, Germany 2 Department of Biology, Biotechnical Faculty, University of Ljubljana, Večna pot 111, SI-1000 Ljubljana, Slovenia; e-mail: primoz.zidar@uni-lj.si Correspondence to Primož Zidar Abstract. The influence of different concentrations of cadmium nitrate on food choice behaviour was studied in the terrestrial isopod Oniscus asellus. In paired food choice tests, consumption rates were compared in relation to cadmium nitrate concentrations and duration of feeding. The consumption of contaminated food was reduced at even the lowest cadmium concentration already in the first week of feeding. In the third week the consumption rates for uncontaminated and contaminated food reached a ratio of 6:4 in ali animal groups. Consumption of contaminated food during the experiment resulted in increased cadmium content in the animals. It is presumed that O. asellus cannot distinguish food according to cadmium concentration. The difference in consumption rates between uncontaminated and cadmium-contaminated food could be based on integration of feeding behaviour and the adverse metabolic effects of cadmium. Keywords: Isopods, Oniscus asellus, food-choice, cadmium nitrate, contamination, consumption rate, accumulation Izvleček. Proučevali smo vpliv različnih koncentracij hrani dodanega kadmijevega nitrata na izbiro hrane pri kopenskem enakonožcu Oniscus asellus. V poskusih izbire smo primerjali stopnjo hranjenja z neonesnaženo in onesnaženo hrano, v odvisnosti od koncentracije kadmijevega nitrata in trajanja hranjenja. Živali so zaužile manj onesnažene hrane že v prvem tednu poskusa izbire tudi pri najnižji koncentraciji kadmijevega nitrata dodanega hrani. V tretjem tednu poskusa je bilo pri vseh skupinah živali razmerje med stopnjo hranjenja z neonesnaženo in onesnaženo hrano 6:4. Pri živalih, ki so med poskusom jedle hrano z dodanim kadmijevim nitratom je bila telesna vsebnost kadmija povečana. Iz rezultatov sklepamo, da raki enakonožci verjetno ne razlikujejo med hrano z različno vsebnostjo kadmijevega nitrata. Razlike v stopnji 4 Introduction Acta Biologica Slovenica, 45 (2), 2002 hranjenja z neonesnaženo in onesnaženo hrano so verjetno posledica prehranjevalnega vedenja živali povezanega s presnovnimi učinki kadmija. Ključne besede: enakonožci, Oniscus asellus, izbira hrane, kadmijev nitrat, onesnaževanje, stopnja hranjenja, akumulacija Faced with an ever growing input of waste substances to the chemical cycles of the biosphere due to anthropogenic activities, the importance of assessing levels of pollution by biomonitoring programmes in various ecosystems has gained increasing recognition over the last decades (BAYNE 1979, HoPKJN 1989, GoLDBERG & BERTINE 2000). The success of the 'Mussel Watch Program' in marine ecosystems has led to considerable efforts to establish similar programmes for terrestrial ecosystems. Terrestrial isopods are amongst the most promising animals for globa! terrestrial biomonitoring, and the species Porcellio scaber has been proposed as the terrestrial biomonitoring equivalent of Mytilus edulis (CoRTET & al. 1999, CoUGHTREY & al. 1997, DROBNE 1997, HoPKIN & al. 1986, HoPKIN 1990, HoPKIN & al. 1993, PAOLETTI & HASSALL 1999). However, in recent years some data on possible pollutant-dependent food selection behaviour have been presented (VAN CAPELLEVEEN & al. 1986, DALLINGER 1977, OoENDAAL & REINECKE 1999, DROBNE & al. 1995). lf terrestrial isopods are able to discriminate between differently contaminated food, their value as biomonitoring organisms would diminish. Terrestrial isopods play an important role in decomposition of organic material and fulfill most of the criteria required of a good biomonitor (HoPKIN 1989, HoPKIN & al. 1993, CouGHTREY & al. 1997, PAOLETTI & HASSALL 1999). Many aspects ofthe ecology of Porcellio scaber are well known , and it is one ofthe most studied isopods (CoRTET & al. 1999, DROBNE 1997, GuNNARSON 1987, OoENDAAL & REINECKE 1999, PAOLETTI & HASSALL 1999). Field studies and monitoring programmes have used the species successfully in assessing the bioavailability of pollutants (HAMES & HoPKIN 1989, HoPKIN 1990, RABITSCH 1995). Terrestrial isopods accumulate the highest tissue concentrations of cadmium, copper, lead, and zine known for any invertebrate (HoPKIN 1989, Ho PKIN & MARTIN 1984). The robustness against pollution with metals in isopods apparently stems from a compartmentalization mechanism dependent on metal-containing granules in s mali cells of the hepatopancreas and possibly a detoxification mechanism with binding metallothioneins, which yet remains to be verified (CROMENTUnN & al. 1994, HAMES & HoPKIN 1991, HoPKIN 1989, PROSI & al. 1983). The shiny woodlouse Oniscus asellus has been proposed as a substitute for P scaber in areas where the latter is scarce or missing from the fauna (DROBNE 1997). O. asellus shows higher accumulation rates for cadmium and other metals than P scaber, and has likewise a wide distribution, thus constituting a suitable alternative to P scaber (HAMES & HoPKIN 1991, HoPKIN 1990). In recent years, much information has been gathered about the sensoty equipment of terrestrial isopods. They are able to distinguish between and show preferences for different qualities of food, e.g. different levels of fungal permeation of their food (ZIMMER & al 1996, GuNNARSON 1987, SzLAVECZ & MAIORANA 1990). In addition, some results offood choice experiments have suggested that terrestrial isopods may also sense different levels of contaminants in their food, e.g. P laevis has been shown to discriminate against cadmium sulfate added to leaves and P scaber can apparently discriminate and avoid lead and copper in its food (VAN CAPELLEVEEN & AL. 1986, DALLINGER 1977, ODENDAAL & REINECKE 1999). If terrestrial isopods have evolved mechanisms to avoid uptake of biotoxic substances from their food, then the concentrations of these substances in their tissues do not represent levels of habitat pollution, but are the result of an integrated behavioural response to pollution (DROBNE & al 1995, DoNKER & BooERT 1991). However, one ofthe criteria for the use of an organism asa biomonitor U. l. Kaschl, P. Zidar, J. Štrus, V. Storch: Food choice experiments with cadmium nitrate ... 5 is that accumulation of a pollutant should reflect exposure. Results from biomonitoring experiments using terrestrial isopods could therefore be unreliable and should be treated with caution (DROBNE & al. 1995). In the present work we investigated the possibility of an avoidance mechanism for the consumption of cadmium-contaminated food by the terrestrial isopod O. asellus in paired food choice experiments. We aimed to show that the results from our experiments, which document preference for uncontaminated food, in correlation with previous publications, can be explained without assuming the existence of a sensory detection mechanism for cadmium in the diet, thus rendering the species an accurate biomonitor of bioavailable cadmium in its habitat. We assumed that the observed preference for uncontaminated food could be explained by the combined influence of feeding behaviour and the adverse metabolic effects of cadmium in the isopods. Materials and Methods Sampling of animals from the field About 250 specimens of Oniscus asellus were collected in October 2001 from the litter layer of a woodland area near a former smel ter site in the vicinity of Nussloch near Heidelberg, Germany. For three weeks the animals were kept in a glass container on moist plaster of Paris and fed on partly decomposed unpolluted leaves ofvarious tree species. The temperature in the glass container was kept steady at about 17°C, and exposure to direct light was prevented. To avoid desiccation, the leaves and the plaster of Paris were lightly sprayed with commercial bottled water every two days. After three weeks, ten animals were selected at random, lyophilised and weighed, and digested in a hot acid mixture (HN0 3 :HCl0 4 = 7: 1; fina! temperature l 85°C) until dryness. The residue was suspended in 1.5 ml HN0 3 (0.2%) and analysed for cadmium content by flame atomic absorption spectrophotometry (AAS), using a Perkin Elmer AAAnalyst 100 atomic absorption spectrophotometer. Ali further animal samples were digested and analysed according to this method. Sampling of soil From the same site, five soil samples of the top 1 O cm of the soil horizon were collected, cleaned of ali visible organic components, and dried for two hours at 110° C, then homogenised in a mortar and digested in hot acid mixture (HN0 3 :HCI = 1 :3) until dryness. As the samples were not digested completely by this procedure, it was repeated once. The samples were then diluted with weak nitric acid (0.2 % ), and filtered. The filtrate was subsequently analysed for cadmium content by flame AAS analysis. Experimental set-up The animals were separated, sexed and weighed, and kept individually on moist filter paper in plastic petri dishes (0 = 9 cm). Only males and non-gravid females were selected for the experiment. The petri dishes were kept in a climate chamber at a relative humidity of 100% and under a 16 hours light and 8 hours dark regi me. Temperature was kept constant at 21 °C (±1 °C). Animals were checked every two days, and the filter paper was moistened with commercial bottled water if necessary. Any dead animals were removed immediately and their data excluded from the results. Moulting animals or animals that did not produce faecal pellets were marked and counted. The animals were fed exclusively with complex food pellets designed especially for this experiment. For the production offood pellets, partly decomposed hazel leaves (Corylus avellana) were collected in an unpolluted woodland area in the vicinity of Cerknica near Ljubljana, Slovenia. After drying the leaves at room temperature for severa! days, leaf stems were removed and the leaves pulverised with a coffee mili and subsequently sieved through a 0.25 mm mesh net. The leaf powder was dried for 2 6 Aeta Biologiea Slovenica, 45 (2), 2002 hours at 60°C. Other components of the complex food pellets were commercial Dr. Oetker' s gelatine, which was dried at 40°C, and fish food for aquarium fish (JBL Novobel; JBL GmbH& Co. KG), which was homogenised by hand in a glass mortar and dried for two hours at 60°C. Dry leaf powder, gelatine, and fish food were mixed in a 63:34:3 ratio and turned into a paste with demineralized water ( 15 ml per gram gelatine ). For the preparati on of contaminated food, four different amounts of cadmium nitrate (Cd(N0 3 ) 2 ) solution (2712.5 mg Cd/1) were added to the food paste to give nominal concentrations of O, 20, 45, 200 and 450 mg Cd kg·1 dry weight of food. In order to exclude food choice due to nitrate content (Hopkin 1989), corresponding amounts of potassium nitrate (KN0 3 ) were added to the respective control food. Food pellets were formed out of the paste by depositing equal amounts in plastic blisters (V= 0.3 ml). The food pellets were allowed to solidify for 24 h at 5°C, dried at room temperature for 24 h and then dried at 70°C for the next 48 h. The actual concentrations of cadmium in the food pellets were measured by AAS analysis and compared to the nominal cadmium contents (Tab. !). The amount of water-soluble cadmium in the food pellets was measured by soaking food pellets in demineralized water for 24 h at increasing temperatures up to 40°C, centrifuging the solution and analysing the supernatant for Cd content by flame AAS (Tab. 1). The water soluble concentration of cadmium in food pellets was observed to be less than 1 O% of the actual cadmium concentration. Table 1: Concentration of cadmium in food pellets as measured by flame AAS (* = concentration below detection limit) Preglednica 1: Koncentracije kadmija v hrani izmerjene s plamensko AAS (* = koncentracija pod zaznavno mejo) Nominal Cone. (mg/kg) Measured Cone. (mg/kg) n=20 Water Soluble Cone. (mg/kg) n=IO Mean I SE Mean I SE o 0.06 0.02 * 20 20.55 0.30 1.14 0.13 45 45.34 O.SI 3.78 0.07 200 194.55 2.69 14.05 0.20 450 445.26 1.58 36.58 0.40 Food pellets were offered in small plastic dishes (0 = 2 cm, height of rim ca. 3 mm) . Prior to food choice experiments, the animals were acquainted with the form and taste of the food pellets by offering them uncontaminated food for one week. After seven days, the animals were starved for 24 h to empty their guts (HAMES & HoPKIN 1991 ). Ten animals were selected randomly and analysed for tissue cadmium content by flame AAS. Food choice experiments The isopods were assigned to five groups of 35 individuals each, giving a total number of 175 animals. Each animal was offered a choice between an uncontaminated and a Cd-contaminated food pellet; in the group "0-20" the nominal cadmium concentration in the contaminated pellet was 20 mg kg·1 dry food weight, in the group "0-45" 45 mg kg 1 dry food weight, in the group "0-200" 200 mg kg·1 dry food weight, and in the group "0-450" 450 mg kg·1 dry food weight. Animals in the control group "0- 0" were offered a choice of two uncontaminated, but differently marked food pellets. The dry weight of food pellets was determined before and after exposure to animals using a micro-balance. After seven days of the food choice experiment, the animals were starved for 24 h to empty their guts (HAMES & HoPKIN 1991 ). Eight animals of each group were randomly selected and analysed for tissue cadmium content by AAS analysis. The food pellets were collected, dried for two days at room U. I. Kaschl, P. Zidar, J. Štrus, V. Storch: Food choice experiments with cadmium nitrate ... 7 temperature, cleaned from faeces and dried at 70 °C for 48 h. Consumption of food was calculated according to the differences in weight before and after the seven day period of exposure to the animals. The remaining animals were offered a new choice between new food pellets with the same Cd- concentrations as before. The whole cycle of feeding, removal of food, starvation for 24 h, and random selection of eight animals for analysis of cadmium content by AAS analysis was repeated twice, resulting in a total exposure period of twenty-one days for the last cohort of animals. Analysis of data To compare food consumption rates (CR), the absolute consumption of food per week was di vided by the dry weight of animals. The -, -o 120 o ..o g 100 80 bO -" 60 bO 5 40 -o 20 u o ' Field '1l-O" ''0- 20" groups ''0-45" "0-200" ''0-450" Figure 1: Concentration of cadmium in whole Oniscus asellus measured by flameAAS followed over three weeks ofthe food choice experiment. Concentration of cadmium in field animals is represented on the far left (mean ± SE). Slika 1: Koncentracije kadmija v telesu raka enakonožca Oniscus asellus, izmerjene s plamensko AAS v treh tednih poskusa izbire hrane. Povprečna koncentracija kadmija v živalih iz okolja je prikazana skrajno levo (povp. ± stand. napaka). 8 Acta Biologica Slovenica, 45 (2), 2002 Consumption rates During the preliminary experiment, the mean food consumption rate was 0.52 (SE=0.03) mg dry food weight per kg dry weight of animals. In the third week of the food choice experiment, the mean food consumption rate increased to 1.24 (SE=0.05). Combined consumption rates of contaminated and uncontaminated food increased significantly between the first and third week in group "0-0" (Tukey test: p < 0.05) (Fig. 2). In group "0-20", combined consumption rates between the first and second week are significantly different (ANOVA: p<0.05); however, with the Tukey test no statistically significant increase between weekly consumption rates could be shown. Combined consumption rates in group "0-45" were significantly higher in the second and third week (Tukey test: p < 0.005) than in the first week. In group "0-200", combined consumption rates in the third week were significantly higher than in the first week (Tukey test: p < 0.05); between the first and second and the second and third week there was no statistically significant difference. In group "0-450", no statistically significant increase in weekly consumption rates could be shown. 1,8 ...................... 1,6 1,4 1,2 iz u 0,8 0,6 0,4 0,2 o "0-0" "O- 20" ··o-45" groups "0-2(XJ" "0-450" 0 Weekl [] Week2 m Week3 Figure 2: Combined consumption rates (CR) of uncontaminated and contaminated food in different groups of animals for each of the three weeks of the food choice experiment (mean ± SE). Slika 2: Stopnja hranjenja (CR) z neonesnaženo in onesnaženo hrano skupaj v posameznih tednih poskusa izbire, pri različnih skupinah živali (povp. ± stand. napaka). Consumption rates of contaminated and uncontaminated food in different groups Consumption rates show that animals within group "0-0" did not significantly discriminate between food pellets during the three weeks (Fig. 3, Tab. 2). Animals within group "0-20" and "0-45" significantly preferred uncontaminated food to contaminated food in the first and third week but not in the second week. Animals of the group "0-200" consumed significantly more uncontaminated food in ali three weeks. In group "0-450", there was no significant preference for uncontaminated over contaminated food in the first week; however, in the second and third week, animals consumed significantly less contaminated food. U. l. Kaschl, P. Zidar, J. Štrus, V. Storch: Food choice experiments with cadmium nitrate ... 9 A 100 90 80 70 i: ••• ! ... 60 i ! ... r % 50 l. 40 30 20 10 o 0-0 0-20 0-45 0-200 0-450 B 100 90 80 70 f ... 60 I T t ... % 50 t .... .L 40 30 20 10 o 0-0 0-20 0-45 0-200 0-450 C 100 90 80 70 f · 60 f ··· f ... 1· r % 50 I 40 30 20 10 o - 0-0 0-20 0-45 0-200 0-450 Figure 3: Consumption rate (CR) ofuncontaminated food shown as percentage values ( •) of combined CR of uncontaminated and Cd-contaminated food (means and 95% conf. int.): A = 1 st week, B = 2"d week, C = 3,d week of food choice experiments. Stars above bars represent significant differences between CR ofuncontaminated food and 50 % (horizontal line) (t-test: * = p < 0.05; ** = p < 0.01; *** = p < 0.005). Slika 3: Stopnja hranjenja (CR) z neonesnaženo hrano prikazana kot odstotek ( •) skupne stopnje hranjenja z neonesnaženo in s kadmijem onesnaženo hrano (povp. in 95% interval zaupanja): A = 1. teden, B = 2. teden, C = 3. teden poskusov izbire hrane. Statistična značilnost razlik med stopnjo hranjenja z neonesnaženo hrano in 50% (vodoravna črta) je prikazana z zvezdicami (t-test: * = p < 0.05; ** = p