Journal of Environmental Radioactivity 147 (2015) 43e50
Contents lists available at ScienceDirect
Journal of Environmental Radioactivity journal homepage: www.elsevier.com/locate/jenvrad
Maternal transfer of anthropogenic radionuclides to eggs in a small shark Ross A. Jeffree a, *, Francois Oberhansli a, Jean-Louis Teyssie a, Scott W. Fowler b a b
IAEA Environment Laboratories, 4 Quai Antoine, MC 98000, Monaco Institute Bobby, 8 Allee des Orangers, 06320, Cap d'Ail, France
a r t i c l e i n f o
a b s t r a c t
Article history: Received 14 January 2015 Received in revised form 7 May 2015 Accepted 10 May 2015 Available online
Maternal transfer of radionuclides to progeny is one of the least known sources of contamination in marine biota and more information is needed to assess its radiological significance. A radiotracer study on spotted dogfish, Scyliorhinus canicula, evaluated the hypothesis that four anthropogenic radionuclides (Cobalt-60, Zinc-65, Americium-241 and Cesium-134) could be maternally transferred to eggs and each of their major components during maternal ingestion of radiolabelled food. The linear regressions between cumulative radioactivity that had been maternally ingested and the level in subsequently laid eggs were used to derive maternal-to-egg transfer factors (mTFs). These maternal transfers varied over an order of magnitude and were ranked 134Cs > 65Zn > 60Co > 241Am. This ranking was the same as their relative assimilation efficiencies in radiolabelled food consumed by adults. Among these four radionuclides the potential radiological exposure of embryos is accentuated for 65Zn and 134Cs due to their predominant transfer to egg yolk where they are available for subsequent absorption by the embryo as it develops prior to hatching from the egg capsule. Thus, for cartilaginous fish like shark, the potential radioecological consequences of a pulsed release of these radionuclides into the marine environment may extend beyond the temporal duration of the release. © 2015 Elsevier Ltd. All rights reserved.
Keywords: Maternal Transfer Radionuclides Shark Egg Bioaccumulation
1. Introduction The maternal-to-embryo exposure pathway has been a topic of importance within the International Commission on Radiological Protection with regard to human transplacental transfer of radionuclides (ICRP 88, 2001), because of the extreme radiosensitivity of the embryo and foetus (Von Zallinger and Tempel, 1998). Studies among other mammal species have emphasised the speciesspecific differences in their transplacental radionuclide dynamics (Von Zallinger and Tempel, 1998). More recent studies on levels of radioactive cesium in cattle from the evacuation zone of the Fukushima Daiichi NPP have drawn attention to elevated levels in the organs of foetuses and infants compared to their mothers (Fukuda et al., 2013). Given the continued interest in this transfer pathway of demonstrated significance (Von Zallinger and Tempel,
* Corresponding author. Present address: School of Life Sciences, C3, Faculty of Science, University of Technology, Sydney, P.O. Box 123, Broadway, NSW, 2007, Australia. E-mail address: [email protected] (R.A. Jeffree). http://dx.doi.org/10.1016/j.jenvrad.2015.05.009 0265-931X/© 2015 Elsevier Ltd. All rights reserved.
1998) then exploratory studies in other taxa, in addition to terrestrial mammals, would seem valuable. There are increasing efforts to better assess the risks to biota in the aquatic environment that result from the releases of radionuclides from nuclear facilities, through a greater knowledge of their environmental transfers to biota and the significance of their consequent radiation exposure. This process is facilitated by the use of databases of transfer factors, which are currently dominated by values for adult life stages (eg. Howard et al., 2013; IAEA, 2004) and require better representation of many phylogenetic groups, eg. cartilaginous fishes. Moreover, fish eggs are included as ICRP RAPs (Reference Animals and Plants) (ICRP, 2008) but there are no data to determine radionuclide transfers to fish eggs (ICRP, 2009). Transfer factors derived from field-based data characteristically represent the aquatic organism's accumulation of a radionuclide via the environmental transfers from water, food and sediment. These three traditional environmental pathways of radionuclide exposure and transfer to marine fishes have also been frequently investigated under laboratory conditions, but the pathway of maternal transfer to their eggs and later developmental stages has received the least attention. Exposures of the embryonic phase, the life-stage that is typically more vulnerable to toxicological impacts of trace metals
44
R.A. Jeffree et al. / Journal of Environmental Radioactivity 147 (2015) 43e50
and radionuclides (Blaylock and Griffith, 1971; Weis and Weis, 1991), has received more attention in teleost fish (Mothersill et al., 2010), compared to their cartilaginous relatives with their characteristically reduced reproductive potential (Camhi et al., 1998). In the bull ray, Pteromylaeus bovinus, Hg transfer from mother to embryos has been found under field exposure conditions (Horvat et al., 2014) but to our knowledge no previous experimental studies have been undertaken on maternal transfer and distributions of radionuclides in the eggs of cartilaginous fish. With respect to other transfer pathways there is growing evidence that Chondrichthyans (car tilagenous fish) may be more susceptible than Teleosts (bony fish) to accumulation of trace metal contaminants from food and seawater (De Boeck et al., 2010; Jeffree et al., 2006a, 2010; Leah et al., 1991a,b; Mathews et al., 2008). We have chosen the spotted dogfish, Scyliorhinus canicula, (Order Carcharhiniformes, Family Scyliorhinidae), to test the hypothesis that anthropogenic radionuclides can be maternally transferred to progeny. The four radionuclides used in this study, Zinc-65, Cobalt-60, Americium-241 and Cesium-134, are typically associated with effluents entering the marine environment from coastal nuclear facilities. S. canicula belongs to the Family Scyliorhinidae, which includes over 100 species that occur worldwide in temperate and tropical seas (Compagno, 1999). It has a low natural capacity to increase (r), which is typical of cartilaginous fish, producing a relatively small number of large, heavily yolked eggs. The eggs of S. canicula have long periods of embryonic development, which are characteristic of dogfish species (Castro et al., 1998), and contrast with eggs of pelagic marine teleosts which typically show reduced duration of development by at least an order of magnitude (Paully and Pullin, 1988). Such a mode of reproduction is typical of oviparous Chondrichthyans which comprise about 40% of the sharks and all the skates (Wourms and Demski, 1993). In this radiotracer uptake experiment, we determined whether there was maternal transfer of four anthropogenic radionuclides (65Zn, 60Co, 241Am and 134Cs) to eggs of S. canicula, during 61 days of the maternal exposure to radio-labelled food, and their distributions among three egg components (case, yolk and jelly). 2. Materials and methods 2.1. Radioanalysis of tissues and water samples A high-resolution gamma spectrometry system was used for all radiometric analyses of mussel, dogfish, egg and seawater samples, consisting of four coaxial Germanium (N- or P-type) detectors (EGNC 33-195-R, Intertechnique; 40e70% efficiency). The detectors were connected to a multi-channel analyser and a personal computer employing spectral analysis software (Interwinner 6, Intertechnique). The radioactivity levels of samples were determined by comparison with known standards of appropriate geometry, including a phantom dogfish egg, and were corrected for background and physical decay of a given radioisotope. Counting times were adapted to obtain count rates with relative propagated errors of less than 5%, viz. typically 10e30 min for whole egg radioanalyses and 1e12 h for seawater. The radioanalytical techniques used in this study were identical to those used in previous experiments on shark juveniles, eggs and their components (Jeffree et al., 2006a,b; 2007, 2008). 2.2. Acclimation and experimental feeding of female dogfish with radiolabelled food Eight female dogfish provided by the Roscoff Biology Station, Britanny, France, were chosen within a total length range of 70e80 cms to ensure their sexual maturity (Ivory et al., 2004;
Rodriguez-Cabello et al., 1998). These individuals were transported overnight in seawater within plastic bags which were inflated with oxygen and that were held within insulated boxes. Following their arrival at the IAEA Monaco laboratory, they were maintained with no mortality in running, carbon-filtered Mediterranean seawater in a 3000 L aquarium for 30 days at 16 ± 1 C. Acclimation conditions were virtually identical to those used during the subsequent long-term exposure experiment, and with a similar feeding regime. All animals used in this experimental study were treated in accordance with the animal care and ethics standards required by the Principality of Monaco and the IAEA. It is typical of oviparous Chondrichthyans that a large proportion of the females are not laying eggs in any one year (Hamlett and Koob, 1999). Therefore, since not all sexually mature females of S. canicula would be expected to contain egg capsules (RodriguezCabello et al., 1998), they were scanned by ultrasound to determine those individuals that contained developing eggs within their ovaries, and a subset of four with developing eggs was chosen for the experimental exposures to radiolabelled food. Under natural conditions, dogfish are known to feed opportunistically on various macrobenthic fauna, bivalve and gastropod molluscs being the dominant prey items and increasing in dietary frequency with increasing size of dogfish (Lyle, 1983). For this experiment, we chose Mytilus edulis as the experimental food both because of the importance of bivalves in the dogfish diet, and the well-established ability of M. edulis to accumulate a range of trace elements and radionuclides in its soft tissues (Farrington et al., 1983; Fisher et al., 1996). Assimilation efficiencies (AE) for the uptake from food (from a single feeding) of the four radionuclides used in the present investigations have previously been determined for S. canicula by Mathews et al. (2008). For dogfish fed with radiolabelled juvenile Mediterranean sea bream (Sparus auratus), the following percent AE and average loss rate constants in % loss. day1, in brackets, were reported as follows: Am, 6.0 (0.3); Co, 11.0 (nd); Cs, 73.0 (0.8); and Zn, 20.0 (1.7), i.e. AE's varied by up to more than a factor of ten. 2.2.1. Radio-labelling of mussels Approximately 1000 M. edulis (5e6 cm length) were exposed to seawater labelled with high specific activity 241Am, 60Co, 134Cs and 65 Zn for one month in a 300 L aquarium at 16 ±1 C, pH 8.00. These radionuclides had been purchased from Amersham, UK (241Am, 60 Co, 134Cs) and Isotope Product Laboratory, USA (65Zn), and were added to the experimental aquarium every second day for 4 weeks in order to maintain constant activities. No change in pH was detectable after the addition of these isotope solutions. To ensure their more realistic and thorough radio-labelling, mussels were also fed every second day with spikes of 104 cells/ml of unlabelled Isochrisis galbana that the mussels effectively filtered from the radiolabelled water within 30 min. The activity concentrations of each radioisotope were checked daily both before and after each seawater renewal, so as to determine their mean time-integrated values, which were as follows: 60 Co; 0.5 kBq/L, 65Zn; 0.6 kBq/L, 134Cs; 0.6 kBq/L, and 241Am; 0.1 kBq/L. Their mean accumulated activities per individual mussel after one month of exposure were: 150 kBq of 60Co, 180 kBq of 65Zn, 180 kBq of 134Cs, and 30 kBq of 241Am. At the end of this uptake phase, the soft tissues of all mussels were dissected, homogenised and then divided into four aliquots that were then weighed and frozen. Just prior to beginning the feeding experiment, the respective radioisotopic activity concentrations in mussel tissues were quantified. The average mussel tissue (wet weight)-to-water concentration factors attained were: 65 Zn, 333; 60Co, 35; 241Am, 61; and 134Cs, 2. At each feeding time portions of mussel tissue (c. 20 gms) were taken from each of the
R.A. Jeffree et al. / Journal of Environmental Radioactivity 147 (2015) 43e50
four aliquots, weighed and their radioisotopic activities were determined. Variability in the radionuclides contents between the homogenized food aliquots were negligible. 2.2.2. Maternal feeding regime with radio-labelled mussel tissue The four out of the eight female dogfish that were identified by ultrasound to be gravid were maintained in a 3000 L tank under experimental conditions that were the same as those used for their acclimation, but which was then partitioned into corrals of 750 L volume under separate confinement so that their individual rates and patterns of ingestion of food and radionuclide activities could be accurately quantified. For the 61 days of the experimental exposure the dogfish were offered radio-labelled food for five days (Monday-Friday) per week. Within this 61 day feeding regime, the period of exposure of the maternal reproductive tract to radionuclides absorbed from radiolabelled food was defined as extending from the first ingestion minus a lag phase (as determined below) to the last ingestion plus the lag phase, where the maximum absolute period of ingestion extended for 46 days within the 61 day period. Following the delivery of food portions of measured mass and radioisotopic activity concentrations to each individual dogfish, the residual food that was not consumed within 15 min was then retrieved by pipette in order to determine, by subtraction, both the food mass and total radiotracer activity concentrations that were actually ingested by each fish. The small amount of radioactivity that may have been lost from food not ingested during this period was flushed out of the experimental bath by the high through-put rate of seawater. This was confirmed by repeated analyses of seawater taken from the corral of each fish, post-feeding. Moreover, two control dogfish of similar size which were fed only with mussels not containing radionuclides were maintained in experimental water to evaluate total isotope redistributions from labelled food to fish during the feeding phase. No radioisotopes were detected in these controls. The feeding and radioisotopic ingestion patterns for each dogfish that ingested food varied considerably between individuals. Three individuals repeatedly ingested radio-labelled mussel tissue during the feeding phase. The remaining individual only fed once during the feeding phase and the levels of radionuclides in the 5 eggs that it subsequently laid were below the limits of detection for each of the four radionuclides studied. These data on the total weight of food ingested for each dogfish within the 61 days were used to calculate their daily rations, as the percentage of food ingested per body weight per day. 2.3. Data calculations and statistical analyses 2.3.1. Estimation of the lag period in the maternal transfer of radionuclides to eggs during maternal uptake To investigate the relationship between the activity concentrations of radioisotopes ingested by the mother and the resulting levels in eggs, it was first necessary to adjust these measured data for a lag period between the time of the maternal ingestion of a radioisotope and when it first appeared in the laid egg. Such a lag results from a sequence of time dependencies following the initial maternal ingestion of radiolabelled food, including; food digestion and absorption of radionuclides into the blood, their transfer to those organs involved in the production of egg components, and time from egg production to oviposition. To empirically estimate the appearance of each radiotracer in the egg, we determined for the three egg-laying females i) the maximum number of days that elapsed since the beginning of radiolabelled feeding when no radiotracer was yet detected in an egg; and ii) the number of days that elapsed before the first egg was laid with detectable radiotracer.
45
2.3.2. Radionuclide transfer during maternal uptake After correction of data for the lag phase, the total activity and the activity concentration of each radionuclide in each egg at oviposition (laying) were linearly regressed against the cumulative activity ingested by the mother prior to its oviposition. These regressions were appropriately constrained through the origin so that the slope of the significant (P 0.05) fitted regression line could then be interpreted as a measure of the maternal-to-egg transfer factor (mTF), based on the cumulative dietary intake of these radionuclides. 2.4. Distributions of radionuclides among egg components Following radioanalysis of each whole egg to measure its total activities and calculation of the wet weight activity concentration of each radio-isotope, the egg was dissected into case, glycosaminoglycan jelly and yolk with embryo for determination by gamma radiometry of the individual radioisotope content and wet weight activity concentrations. Directly following oviposition, the embryo has reached the developmental stage of a blastodisc (disc-shaped layer of cytoplasm) on the surface of the yolk (viz. Stage 4; Ballard et al., 1993), which is still very small and difficult to visually identify. As a result the embryo was included with the yolk and analysed for its radionuclide content. These data were used to discern patterns of radionuclide distributions in the eggs during the experiment. 3. Results 3.1. Maternal rates of ingestion of radiolabelled food The temporal frequencies of feeding were similar between individual dogfish. However the quantities of radio-labelled mussel tissue ingested per meal, were quite variable among individuals., To quantify this variability the total mass of food ingested was divided by the total number of days over which feeding took place, within the 61 day experiment. The following average daily rations for each maternal dogfish, expressed in terms of the average mass of food ingested daily as a percentage of total mass of maternal dogfish, were 1.5, 2.8 and 0.03% food ingested/body weight/day. The two higher values fell within the range of values determined for various other Chondrichthyans (Wetherbee and Cortes, 2004a,b; Holmgren and Nilsson, 1999). 3.2. Maternal production of eggs and their radiolabelling The three dogfish that repeatedly consumed radiolabelled food subsequently laid 19 eggs collectively during the maternal uptake phase. These eggs ranged between 5.3 and 7.2 g (wet weight) and in total length between 5.5 and 6.3 cm. Dogfish repeatedly laid either single or paired eggs at time points which were evenly distributed over 45e60 days from the beginning of the experiment, with similar total numbers for each mother. Small but significant amounts of each radionuclide were detected in whole eggs as early as day 4, as well as for each egg subsequently laid. The setting of the lag phase at 4 days was also consistent with information on S. canicula regarding its time scales of digestion (Sims et al., 1996), yolk production (Craik, 1978) and oviposition (Mellinger, 1983). 3.3. Rates of maternal transfer of radionuclides to eggs 3.3.1. Whole eggs Table 1 provides the results of simple linear regression analyses between the total activity and activity concentration in each egg laid during maternal uptake, as a function of the total amount of activity ingested by the mother prior to its oviposition. These
46
R.A. Jeffree et al. / Journal of Environmental Radioactivity 147 (2015) 43e50 Table 1 Results of linear regression analysis between total radionuclide ingested by maternal dogfish (Bq) and the resultant radionuclide contents (Bq) and concentrations (Bq g1 wet weight) in their eggs.a Radionuclide Regression slope 60
Co Zn 241 Am 134 Cs 65
60
Co Zn Am 134 Cs 65
241
SE of slope
r2
P for regression
Total content in egg (Bq) 3e-4 5e-5 0.67
Contents lists available at ScienceDirect
Journal of Environmental Radioactivity journal homepage: www.elsevier.com/locate/jenvrad
Maternal transfer of anthropogenic radionuclides to eggs in a small shark Ross A. Jeffree a, *, Francois Oberhansli a, Jean-Louis Teyssie a, Scott W. Fowler b a b
IAEA Environment Laboratories, 4 Quai Antoine, MC 98000, Monaco Institute Bobby, 8 Allee des Orangers, 06320, Cap d'Ail, France
a r t i c l e i n f o
a b s t r a c t
Article history: Received 14 January 2015 Received in revised form 7 May 2015 Accepted 10 May 2015 Available online
Maternal transfer of radionuclides to progeny is one of the least known sources of contamination in marine biota and more information is needed to assess its radiological significance. A radiotracer study on spotted dogfish, Scyliorhinus canicula, evaluated the hypothesis that four anthropogenic radionuclides (Cobalt-60, Zinc-65, Americium-241 and Cesium-134) could be maternally transferred to eggs and each of their major components during maternal ingestion of radiolabelled food. The linear regressions between cumulative radioactivity that had been maternally ingested and the level in subsequently laid eggs were used to derive maternal-to-egg transfer factors (mTFs). These maternal transfers varied over an order of magnitude and were ranked 134Cs > 65Zn > 60Co > 241Am. This ranking was the same as their relative assimilation efficiencies in radiolabelled food consumed by adults. Among these four radionuclides the potential radiological exposure of embryos is accentuated for 65Zn and 134Cs due to their predominant transfer to egg yolk where they are available for subsequent absorption by the embryo as it develops prior to hatching from the egg capsule. Thus, for cartilaginous fish like shark, the potential radioecological consequences of a pulsed release of these radionuclides into the marine environment may extend beyond the temporal duration of the release. © 2015 Elsevier Ltd. All rights reserved.
Keywords: Maternal Transfer Radionuclides Shark Egg Bioaccumulation
1. Introduction The maternal-to-embryo exposure pathway has been a topic of importance within the International Commission on Radiological Protection with regard to human transplacental transfer of radionuclides (ICRP 88, 2001), because of the extreme radiosensitivity of the embryo and foetus (Von Zallinger and Tempel, 1998). Studies among other mammal species have emphasised the speciesspecific differences in their transplacental radionuclide dynamics (Von Zallinger and Tempel, 1998). More recent studies on levels of radioactive cesium in cattle from the evacuation zone of the Fukushima Daiichi NPP have drawn attention to elevated levels in the organs of foetuses and infants compared to their mothers (Fukuda et al., 2013). Given the continued interest in this transfer pathway of demonstrated significance (Von Zallinger and Tempel,
* Corresponding author. Present address: School of Life Sciences, C3, Faculty of Science, University of Technology, Sydney, P.O. Box 123, Broadway, NSW, 2007, Australia. E-mail address: [email protected] (R.A. Jeffree). http://dx.doi.org/10.1016/j.jenvrad.2015.05.009 0265-931X/© 2015 Elsevier Ltd. All rights reserved.
1998) then exploratory studies in other taxa, in addition to terrestrial mammals, would seem valuable. There are increasing efforts to better assess the risks to biota in the aquatic environment that result from the releases of radionuclides from nuclear facilities, through a greater knowledge of their environmental transfers to biota and the significance of their consequent radiation exposure. This process is facilitated by the use of databases of transfer factors, which are currently dominated by values for adult life stages (eg. Howard et al., 2013; IAEA, 2004) and require better representation of many phylogenetic groups, eg. cartilaginous fishes. Moreover, fish eggs are included as ICRP RAPs (Reference Animals and Plants) (ICRP, 2008) but there are no data to determine radionuclide transfers to fish eggs (ICRP, 2009). Transfer factors derived from field-based data characteristically represent the aquatic organism's accumulation of a radionuclide via the environmental transfers from water, food and sediment. These three traditional environmental pathways of radionuclide exposure and transfer to marine fishes have also been frequently investigated under laboratory conditions, but the pathway of maternal transfer to their eggs and later developmental stages has received the least attention. Exposures of the embryonic phase, the life-stage that is typically more vulnerable to toxicological impacts of trace metals
44
R.A. Jeffree et al. / Journal of Environmental Radioactivity 147 (2015) 43e50
and radionuclides (Blaylock and Griffith, 1971; Weis and Weis, 1991), has received more attention in teleost fish (Mothersill et al., 2010), compared to their cartilaginous relatives with their characteristically reduced reproductive potential (Camhi et al., 1998). In the bull ray, Pteromylaeus bovinus, Hg transfer from mother to embryos has been found under field exposure conditions (Horvat et al., 2014) but to our knowledge no previous experimental studies have been undertaken on maternal transfer and distributions of radionuclides in the eggs of cartilaginous fish. With respect to other transfer pathways there is growing evidence that Chondrichthyans (car tilagenous fish) may be more susceptible than Teleosts (bony fish) to accumulation of trace metal contaminants from food and seawater (De Boeck et al., 2010; Jeffree et al., 2006a, 2010; Leah et al., 1991a,b; Mathews et al., 2008). We have chosen the spotted dogfish, Scyliorhinus canicula, (Order Carcharhiniformes, Family Scyliorhinidae), to test the hypothesis that anthropogenic radionuclides can be maternally transferred to progeny. The four radionuclides used in this study, Zinc-65, Cobalt-60, Americium-241 and Cesium-134, are typically associated with effluents entering the marine environment from coastal nuclear facilities. S. canicula belongs to the Family Scyliorhinidae, which includes over 100 species that occur worldwide in temperate and tropical seas (Compagno, 1999). It has a low natural capacity to increase (r), which is typical of cartilaginous fish, producing a relatively small number of large, heavily yolked eggs. The eggs of S. canicula have long periods of embryonic development, which are characteristic of dogfish species (Castro et al., 1998), and contrast with eggs of pelagic marine teleosts which typically show reduced duration of development by at least an order of magnitude (Paully and Pullin, 1988). Such a mode of reproduction is typical of oviparous Chondrichthyans which comprise about 40% of the sharks and all the skates (Wourms and Demski, 1993). In this radiotracer uptake experiment, we determined whether there was maternal transfer of four anthropogenic radionuclides (65Zn, 60Co, 241Am and 134Cs) to eggs of S. canicula, during 61 days of the maternal exposure to radio-labelled food, and their distributions among three egg components (case, yolk and jelly). 2. Materials and methods 2.1. Radioanalysis of tissues and water samples A high-resolution gamma spectrometry system was used for all radiometric analyses of mussel, dogfish, egg and seawater samples, consisting of four coaxial Germanium (N- or P-type) detectors (EGNC 33-195-R, Intertechnique; 40e70% efficiency). The detectors were connected to a multi-channel analyser and a personal computer employing spectral analysis software (Interwinner 6, Intertechnique). The radioactivity levels of samples were determined by comparison with known standards of appropriate geometry, including a phantom dogfish egg, and were corrected for background and physical decay of a given radioisotope. Counting times were adapted to obtain count rates with relative propagated errors of less than 5%, viz. typically 10e30 min for whole egg radioanalyses and 1e12 h for seawater. The radioanalytical techniques used in this study were identical to those used in previous experiments on shark juveniles, eggs and their components (Jeffree et al., 2006a,b; 2007, 2008). 2.2. Acclimation and experimental feeding of female dogfish with radiolabelled food Eight female dogfish provided by the Roscoff Biology Station, Britanny, France, were chosen within a total length range of 70e80 cms to ensure their sexual maturity (Ivory et al., 2004;
Rodriguez-Cabello et al., 1998). These individuals were transported overnight in seawater within plastic bags which were inflated with oxygen and that were held within insulated boxes. Following their arrival at the IAEA Monaco laboratory, they were maintained with no mortality in running, carbon-filtered Mediterranean seawater in a 3000 L aquarium for 30 days at 16 ± 1 C. Acclimation conditions were virtually identical to those used during the subsequent long-term exposure experiment, and with a similar feeding regime. All animals used in this experimental study were treated in accordance with the animal care and ethics standards required by the Principality of Monaco and the IAEA. It is typical of oviparous Chondrichthyans that a large proportion of the females are not laying eggs in any one year (Hamlett and Koob, 1999). Therefore, since not all sexually mature females of S. canicula would be expected to contain egg capsules (RodriguezCabello et al., 1998), they were scanned by ultrasound to determine those individuals that contained developing eggs within their ovaries, and a subset of four with developing eggs was chosen for the experimental exposures to radiolabelled food. Under natural conditions, dogfish are known to feed opportunistically on various macrobenthic fauna, bivalve and gastropod molluscs being the dominant prey items and increasing in dietary frequency with increasing size of dogfish (Lyle, 1983). For this experiment, we chose Mytilus edulis as the experimental food both because of the importance of bivalves in the dogfish diet, and the well-established ability of M. edulis to accumulate a range of trace elements and radionuclides in its soft tissues (Farrington et al., 1983; Fisher et al., 1996). Assimilation efficiencies (AE) for the uptake from food (from a single feeding) of the four radionuclides used in the present investigations have previously been determined for S. canicula by Mathews et al. (2008). For dogfish fed with radiolabelled juvenile Mediterranean sea bream (Sparus auratus), the following percent AE and average loss rate constants in % loss. day1, in brackets, were reported as follows: Am, 6.0 (0.3); Co, 11.0 (nd); Cs, 73.0 (0.8); and Zn, 20.0 (1.7), i.e. AE's varied by up to more than a factor of ten. 2.2.1. Radio-labelling of mussels Approximately 1000 M. edulis (5e6 cm length) were exposed to seawater labelled with high specific activity 241Am, 60Co, 134Cs and 65 Zn for one month in a 300 L aquarium at 16 ±1 C, pH 8.00. These radionuclides had been purchased from Amersham, UK (241Am, 60 Co, 134Cs) and Isotope Product Laboratory, USA (65Zn), and were added to the experimental aquarium every second day for 4 weeks in order to maintain constant activities. No change in pH was detectable after the addition of these isotope solutions. To ensure their more realistic and thorough radio-labelling, mussels were also fed every second day with spikes of 104 cells/ml of unlabelled Isochrisis galbana that the mussels effectively filtered from the radiolabelled water within 30 min. The activity concentrations of each radioisotope were checked daily both before and after each seawater renewal, so as to determine their mean time-integrated values, which were as follows: 60 Co; 0.5 kBq/L, 65Zn; 0.6 kBq/L, 134Cs; 0.6 kBq/L, and 241Am; 0.1 kBq/L. Their mean accumulated activities per individual mussel after one month of exposure were: 150 kBq of 60Co, 180 kBq of 65Zn, 180 kBq of 134Cs, and 30 kBq of 241Am. At the end of this uptake phase, the soft tissues of all mussels were dissected, homogenised and then divided into four aliquots that were then weighed and frozen. Just prior to beginning the feeding experiment, the respective radioisotopic activity concentrations in mussel tissues were quantified. The average mussel tissue (wet weight)-to-water concentration factors attained were: 65 Zn, 333; 60Co, 35; 241Am, 61; and 134Cs, 2. At each feeding time portions of mussel tissue (c. 20 gms) were taken from each of the
R.A. Jeffree et al. / Journal of Environmental Radioactivity 147 (2015) 43e50
four aliquots, weighed and their radioisotopic activities were determined. Variability in the radionuclides contents between the homogenized food aliquots were negligible. 2.2.2. Maternal feeding regime with radio-labelled mussel tissue The four out of the eight female dogfish that were identified by ultrasound to be gravid were maintained in a 3000 L tank under experimental conditions that were the same as those used for their acclimation, but which was then partitioned into corrals of 750 L volume under separate confinement so that their individual rates and patterns of ingestion of food and radionuclide activities could be accurately quantified. For the 61 days of the experimental exposure the dogfish were offered radio-labelled food for five days (Monday-Friday) per week. Within this 61 day feeding regime, the period of exposure of the maternal reproductive tract to radionuclides absorbed from radiolabelled food was defined as extending from the first ingestion minus a lag phase (as determined below) to the last ingestion plus the lag phase, where the maximum absolute period of ingestion extended for 46 days within the 61 day period. Following the delivery of food portions of measured mass and radioisotopic activity concentrations to each individual dogfish, the residual food that was not consumed within 15 min was then retrieved by pipette in order to determine, by subtraction, both the food mass and total radiotracer activity concentrations that were actually ingested by each fish. The small amount of radioactivity that may have been lost from food not ingested during this period was flushed out of the experimental bath by the high through-put rate of seawater. This was confirmed by repeated analyses of seawater taken from the corral of each fish, post-feeding. Moreover, two control dogfish of similar size which were fed only with mussels not containing radionuclides were maintained in experimental water to evaluate total isotope redistributions from labelled food to fish during the feeding phase. No radioisotopes were detected in these controls. The feeding and radioisotopic ingestion patterns for each dogfish that ingested food varied considerably between individuals. Three individuals repeatedly ingested radio-labelled mussel tissue during the feeding phase. The remaining individual only fed once during the feeding phase and the levels of radionuclides in the 5 eggs that it subsequently laid were below the limits of detection for each of the four radionuclides studied. These data on the total weight of food ingested for each dogfish within the 61 days were used to calculate their daily rations, as the percentage of food ingested per body weight per day. 2.3. Data calculations and statistical analyses 2.3.1. Estimation of the lag period in the maternal transfer of radionuclides to eggs during maternal uptake To investigate the relationship between the activity concentrations of radioisotopes ingested by the mother and the resulting levels in eggs, it was first necessary to adjust these measured data for a lag period between the time of the maternal ingestion of a radioisotope and when it first appeared in the laid egg. Such a lag results from a sequence of time dependencies following the initial maternal ingestion of radiolabelled food, including; food digestion and absorption of radionuclides into the blood, their transfer to those organs involved in the production of egg components, and time from egg production to oviposition. To empirically estimate the appearance of each radiotracer in the egg, we determined for the three egg-laying females i) the maximum number of days that elapsed since the beginning of radiolabelled feeding when no radiotracer was yet detected in an egg; and ii) the number of days that elapsed before the first egg was laid with detectable radiotracer.
45
2.3.2. Radionuclide transfer during maternal uptake After correction of data for the lag phase, the total activity and the activity concentration of each radionuclide in each egg at oviposition (laying) were linearly regressed against the cumulative activity ingested by the mother prior to its oviposition. These regressions were appropriately constrained through the origin so that the slope of the significant (P 0.05) fitted regression line could then be interpreted as a measure of the maternal-to-egg transfer factor (mTF), based on the cumulative dietary intake of these radionuclides. 2.4. Distributions of radionuclides among egg components Following radioanalysis of each whole egg to measure its total activities and calculation of the wet weight activity concentration of each radio-isotope, the egg was dissected into case, glycosaminoglycan jelly and yolk with embryo for determination by gamma radiometry of the individual radioisotope content and wet weight activity concentrations. Directly following oviposition, the embryo has reached the developmental stage of a blastodisc (disc-shaped layer of cytoplasm) on the surface of the yolk (viz. Stage 4; Ballard et al., 1993), which is still very small and difficult to visually identify. As a result the embryo was included with the yolk and analysed for its radionuclide content. These data were used to discern patterns of radionuclide distributions in the eggs during the experiment. 3. Results 3.1. Maternal rates of ingestion of radiolabelled food The temporal frequencies of feeding were similar between individual dogfish. However the quantities of radio-labelled mussel tissue ingested per meal, were quite variable among individuals., To quantify this variability the total mass of food ingested was divided by the total number of days over which feeding took place, within the 61 day experiment. The following average daily rations for each maternal dogfish, expressed in terms of the average mass of food ingested daily as a percentage of total mass of maternal dogfish, were 1.5, 2.8 and 0.03% food ingested/body weight/day. The two higher values fell within the range of values determined for various other Chondrichthyans (Wetherbee and Cortes, 2004a,b; Holmgren and Nilsson, 1999). 3.2. Maternal production of eggs and their radiolabelling The three dogfish that repeatedly consumed radiolabelled food subsequently laid 19 eggs collectively during the maternal uptake phase. These eggs ranged between 5.3 and 7.2 g (wet weight) and in total length between 5.5 and 6.3 cm. Dogfish repeatedly laid either single or paired eggs at time points which were evenly distributed over 45e60 days from the beginning of the experiment, with similar total numbers for each mother. Small but significant amounts of each radionuclide were detected in whole eggs as early as day 4, as well as for each egg subsequently laid. The setting of the lag phase at 4 days was also consistent with information on S. canicula regarding its time scales of digestion (Sims et al., 1996), yolk production (Craik, 1978) and oviposition (Mellinger, 1983). 3.3. Rates of maternal transfer of radionuclides to eggs 3.3.1. Whole eggs Table 1 provides the results of simple linear regression analyses between the total activity and activity concentration in each egg laid during maternal uptake, as a function of the total amount of activity ingested by the mother prior to its oviposition. These
46
R.A. Jeffree et al. / Journal of Environmental Radioactivity 147 (2015) 43e50 Table 1 Results of linear regression analysis between total radionuclide ingested by maternal dogfish (Bq) and the resultant radionuclide contents (Bq) and concentrations (Bq g1 wet weight) in their eggs.a Radionuclide Regression slope 60
Co Zn 241 Am 134 Cs 65
60
Co Zn Am 134 Cs 65
241
SE of slope
r2
P for regression
Total content in egg (Bq) 3e-4 5e-5 0.67
