Polychlorinated biphenyls in different trophic levels from a shallow lake in Argentina

Chemosphere 48 (2002) 1113–1122 www.elsevier.com/locate/chemosphere

Polychlorinated biphenyls in different trophic levels from a shallow lake in Argentina Marıa de los A. Gonz alez Sagrario a,b, Karina S.B. Miglioranza b,c,*, Julia E. Aizp un de Moreno c, Vıctor J. Moreno c, Alicia H. Escalante b,d a

b

Laboratorio de Invertebrados, Facultad de Ciencias Exactas y Naturales, Universidad Nacional de Mar del Plata, Funes 3250, 7600, Mar del Plata, Argentina Consejo Nacional de Investigaciones Cientıficas y T ecnicas (CONICET). Av. Rivadavia 1917, 1033, Capital Federal, Argentina c Laboratorio de Ecotoxicologıa, Facultad de Ciencias Exactas y Naturales, Universidad Nacional de Mar del Plata, Funes 3350, 7600, Mar del Plata, Argentina d Laboratorio de Limnologıa, Facultad de Ciencias Exactas y Naturales, Universidad Nacional de Mar del Plata, Funes 3250, 7600, Mar del Plata, Argentina Received 14 November 2000; received in revised form 14 February 2002; accepted 25 February 2002

Abstract Polychlorinated biphenyls (PCBs) were determined in a aquatic community from Los Padres Lake, Argentina. Twenty four PCB congeners from tri- to octa-chlorinated isomers were detected and quantified using conventional gas chromatography with electron capture detector (GC-ECD). The aim of this study was to investigate the concentrations of PCBs in freshwater organisms from a shallow lake of Argentina. Stems of bulrush (Schoenoplectus californicus), whole tissues of false loosestrife (Ludwigia sp.) and grass shrimp (Palaemonetes argentinus), and liver, gonads, muscle and mesenteric fat (if present) of fish species (Rhamdia sapo) and (Oligosarcus jenynsi) were analyzed. Two areas were selected to macrophytes sampling: the input area, main PCB source of the lake (Station 1), and the output area, a potential anoxic zone (Station 2). Macrophytes from Station 1 bioconcentrated higher total PCB levels than Station 2, showing that the former have received PCBs washed down from upstream areas. Penta- and hexa-congeners were enriched relative to other congeners in animal biota and macrophytes from Station 1, consistent with commercial mixture of Aroclor 1254 used in this region. In bulrush from Station 2 a predominance of tri- and tetra-chlorinated congeners was observed. Grass shrimp showed the lowest PCB values among animal biota. PCB concentrations in fish tissues varied with the species and the gonadal development. Mesenteric fat, only present in post-spawning organisms of R. sapo, had the highest values of PCBs relative to other tissues. A clearance of total PCBs in ovaries of post-spawning females of R. sapo was observed, but not in testes. O. jenynsi/P. argentinus biomagnification factor (BMF) had a mean value of 18.7. Congeners 44, 52 and 151, showed the highest BMF values, being 64, 66 and 62, respectively. These values would be a consequence of the low depuration rate of 44 and 52 congeners with orthochlorine substitution conducted by O. jenynsi and the high depuration rate of congener 151, which lacks 4 40 - chlorine substitution, carried out by grass shrimp. Although the most of congeners have been biomagnified, they did not clearly displayed a concomitantly increasing with log Kow . Ó 2002 Elsevier Science Ltd. All rights reserved. Keywords: Polychlorinated biphenyls; Macrophytes; Grass shrimp; Fish; Biomagnification factor; Polymictic lake

*

Corresponding author. Address: Laboratorio de Ecotoxicologıa, Facultad de Ciencias Exactas y Naturales, Universidad Nacional de Mar del Plata, Funes 3350, 7600, Mar del Plata, Argentina. Tel.: +54-223-4752426; fax: +54-223-4753150. E-mail address: [email protected] (K.S.B. Miglioranza). 0045-6535/02/$ - see front matter Ó 2002 Elsevier Science Ltd. All rights reserved. PII: S 0 0 4 5 - 6 5 3 5 ( 0 2 ) 0 0 1 4 9 - 2

1114

M.A. Gonzalez Sagrario et al. / Chemosphere 48 (2002) 1113–1122

1. Introduction Polychlorinated biphenyls (PCBs) are ubiquitous contaminants in aquatic environments as a result of uncontrolled spillage, stream transport, surface runoff and atmospheric deposition. Many studies have focused on the distribution of PCBs within lake ecosystem in an attempt to predict the factors influencing the accumulation of PCBs by biota. Because of the low water solubility of PCBs, a strong relationship between the concentration of total PCBs and the lipid content of animal biota has been observed in different water bodies (Van der Oost et al., 1988; Niimi, 1996; Fisk et al., 1998). Recent research has attributed the variation in congener bioaccumulation to: (a) steric changes in the biphenyl molecule with increasing chlorination, (b) partitioning among different environmental compartments and (c) differences in Kow (Russell et al., 1999). Van der Oost et al. (1988) suggested that the congeners distribution in a particular trophic level was better described as a function of the time available to accumulate the contaminants (i.e. age of organisms) than simple partitioning among water, sediments and biota. Therefore, the distribution of PCB congeners within any particular organism in a waterbody is the result of complex interactions among the congener composition of the original PCB source, removal and transport mechanisms within the ecosystem, chemical and physical properties of the congeners which affect the uptake into the organisms, route (e.g. diet) and exposure time. From an ecological viewpoint, biomagnification theory implies that biomagnifying contaminants can play an useful role in determining feeding relationships and consequently,

the trophic community structure (Russell et al., 1999). Moreover, it has been recently stressed that the patterns and concentrations of hydrophobic pollutants are determined by internal physiological processes, such as lipid metabolism and biotransformation (Barron, 1990). In Argentina, there are few works about PCB levels in estuarine ecosystems (Lanfranchi et al., 1998; Menone et al., 2001). The aim of this study was to quantify PCBs in organisms of different trophic levels from a shallow lake in Argentina. Moreover, to know whether lipid content and trophic level influence the accumulation of PCBs in this ecosystem, and to determine the biomagnification factor (BMF) of main PCB congeners in the predation– prey relationship O. jenynsi–P. argentinus.

2. Materials and methods This work was performed in Los Padres Lake, a shallow waterbody, 2.16 km2 surface, located in the southeastern area of Buenos Aires province, Argentina, (37° 550 –38° 020 South; 57° 340 –57° 330 West), between July and November 1997 (Fig. 1). The annual average temperature was 13.5 °C, the minimum mean in July was 7 °C and the high average for January was 19.2 °C (Bocanegra et al., 1993). 2.1. Sampling Samples from different trophic levels were collected. The organisms selected were bulrush (Schoenoplectus californicus), a very common littoral macrophyte; false

Fig. 1. Map of Los Padres Lake located in the southeastern area of Buenos Aires Province, Argentina. S1 : Station 1 and S2 : Station 2.

M.A. Gonzalez Sagrario et al. / Chemosphere 48 (2002) 1113–1122

loosestrife (Ludwigia sp.), a rooted floating leaved; grass shrimp (Palaemonetes argentinus), living associated with aquatic plants, and two fish species: Oligosarcus jenynsi (Characidae), an omnivorous species feeding mainly on grass shrimp, and Rhamdia sapo (Pimelodidae), a bottom feeder. O. jenynsi and the most of R. sapo females were captured in pre-spawning stage. A few male specimens (pre- and post-spawning) and post-spawning females of R. sapo were also analyzed. Bulrush and false loosestrife samples were collected from two sites, close to Los Padres Creek, the input area (Station 1), the main PCB source of the lake, and La Tapera Creek, the output area (Station 2), using a 20  20 cm square sampler with mobile side. Grass shrimp was caught by means of a hand net and fish species using different types of bait. Immediately after catch, samples were wrapped in clean aluminum foil and ice and transported to the laboratory. Total bulrush stems, whole tissues of false loosestrife and grass shrimp, and gonads, liver, mesenteric fat and muscle from fish were analyzed. Samples from the same species and place were pooled. 2.2. Analytical methods Samples were frozen and stored at 20 °C until analysis. After determining length and weight of the whole thawed fish, tissues of liver, gonads, mesenteric fat (only present in immature organisms of R. sapo) and muscle fillets were removed. The tissues from 3 to 8 fish Table 1 were then pooled according to sex and gonad development. Samples were homogenized using a Waring blender. 2.2.1. Extraction Subsamples of 5–10 g of tissues were ground in a mortar with anhydrous sodium sulfate and extracted with a 50:50 mixture of hexane and dichloromethane in a Soxhlet apparatus for 8 h (Muir et al., 1988). Extracts were concentrated under nitrogen to 3 ml. Lipids were removed from the extracts by gel permeation chromatography (GPC) in Bio Beads S-X3 (200–400 mesh) (Bio-Rads Laboratory, Hercules, California) and extracts were subfractionated by silica gel chromatography as previously described by Metcalfe and Metcalfe (1997). The lipid fraction from the GPC was evaporated to dryness to calculate the lipid content. 2.2.2. Analysis All PCB congener analyses were performed as described by Metcalfe and Metcalfe (1997) using a Varian 3500 gas chromatograph equipped with an electron capture detector, and 60 m DB-5 column. Target analytes for the present study were 24 individual congeners: IUPAC # 18, 31, 28, 52, 49, 47, 44, 66, 101, 99, 87, 110, 118, 105, 149, 151, 153, 156, 138, 180, 170, 199, 195 and

1115

194. The sum of them is defined as total PCBs. Quantification of all PCB congeners was done using an external standard consisting of a mixture of PCB congeners (CLB-1 series), purchased from the National Research Council, Canada, supplemented with congeners 52, 99 and 105 purchased from Ultra Scientific (Rhode Island). The limits of detection for analysis of PCB analytes were calculated as three times the standard deviation of the detector response in repetitive injections of sample blanks, according to Keith et al. (1983), ranging between 0.3 and 1.0 ng/g lipid. 2.2.3. Quality control Procedural blanks, a cod liver oil standard reference material (SRM 1588) purchased from the National Institute of Standards and Technology (NIST), were analyzed for QA/QC purposes. Analysis of the NIST reference material indicated that the PCB analytes were quantified to within 10% of their certified concentration. Analysis of the NRC reference material indicated that the PCB analytes were quantified to within 7% of their certified concentration. Duplicate analyses of samples gave results that varied by less than 10%. The concentrations of individual PCB congeners were calculated on a lipid-normalized basis, which is an important factor for the bioconcentration of hydrophobic compounds in animal tissues. Various authors have suggested normalizing concentrations of pollutants to the lipid content in order to reduce intra-species and inter-species variability (Pastor et al., 1996). The BMF was calculated as mean concentration in O. jenynsi muscle/mean concentration in grass shrimp whole tissues.

3. Results The species collected, sample sizes, lengths and weights of animal biota are given in (Table 1). Lipid contents (%) and mean concentrations of total PCBs (ng/g wet wt. and ng/g lipid wt.) of the analyzed biota are shown in Table 2. Since PCB compounds are hydrophobic (high Kow ), their concentrations based on amounts of lipids were considered to be physiologically more relevant in animal biota than those on wet or dry weight basis. PCB values of different trophic levels ranged from 3.5 to 730 ng/g wet wt. and from 0.3 to 12.7 lg/g lipid in the case of the animal biota. The highest PCB levels were found in fish species, consistent with the higher trophic level. There was an apparent relationship between total PCB levels (ng/g wet wt.) and lipid content, but R. sapo muscle tissue. PCB congeners from 3 to 8 chlorine atoms were detected in tissues from all samples.

1116

M.A. Gonzalez Sagrario et al. / Chemosphere 48 (2002) 1113–1122

Table 1 General data of aquatic biota sampled in Los Padres Lake Common name

Species

Sex

N

Bulrush False loosestrife Grass shrimp – Catfish

S. californicus Ludwigia sp. P. argentinus O. jenynsi R. sapo

– – n.d. a-pre a-pre a-post  - pre  - post

4 2 3 1 1 1 1 1

pools of 20 pools of 20 pools of 45 pool of 8 pool of 5 pool of 2 pool of 2 pool of 2

Total length (mm)

Total weight (g)

– – 22.1a (19.4–24.8)b 169.8 (145–185) 430 (402–466) 427.5 (425–430) 405 (393–417) 366.3 (335–384)

– – 0.077 (0.043–0.132) 53.3 (31.8–69.3) 1013.8 (950.8–1030) 946.5 (880.9–1012) 655.4 (590.8–712.1) 550.1 (385.3–718.6)

n.d.: non-determined; a-pre: pre-spawning female; a-post: post-spawning female;  - pre: pre-spawning male;  - post: post-spawning male. a Mean. b Range.

Table 2 Gonadal development, sex, tissues and lipid content of the biota analyzed Organisms

Sex

Tissues

Lipids (%)

PCB concentration

Bulrush

–

Stema ;b

0.2 0.5

5.3 3.5

–

False loosestrife

–

Wholea ;b

1.9 1.9

16.9 4.1

–

Grass shrimp

n.d.

Whole

3.0

9.7

323.3

O. jenynsi

a-pre

Ovary Liver Muscle

4.9 4.9 0.9

135.7 42.0 29.3

2775.5 848.8 3289.9

Catfish

a-pre

Ovary Liver Muscle

4.8 1.9 1.3

43.3 18.6 106.2

894.8 1002.2 8364.6

a-post

Ovary Liver Muscle Mesenteric fat

1.5 3.6 1.6 65.4

11.8 13.2 79.2 389.7

767.5 371.6 4950.6 595.9

 - pre

Testes Liver Muscle

1.2 2.6 1.3

10.5 67.7 153.6

900.0 2400.4 12 693.4

 - post

Testes Mesenteric fat

2.0 66.2

12.9 730.6

650.3 1103.8

Wet weight (ng/g)

Lipid weight (ng/g)

Total polychlorinated biphenyls (mean values) are expressed as ng/g wet wt. and ng/g lipid wt. a-pre: pre-spawning female; a-post: post-spawning female;  - pre: pre-spawning male;  - post: post-spawning male. a Station 1. b Station 2.

3.1. Macrophytes Lipid and PCB levels expressed as ng/g wet wt. in false loosestrife were higher than those in bulrush (Table 2). The differences between macrophytes from both sampling sites were more apparent when the congener group levels were expressed as relative abundance (Fig. 2). PCB concentrations of both macrophytes found in

Station 1 were higher than those from Station 2. A clear predominance of penta- and hexa-chlorinated isomers in macrophytes from Station 1 has been registered, although differences between both species were noted regarding individual congeners (Table 3). On the other hand, in the Station 2, the bioconcentration pattern has exhibited differences in both macrophytes. Tri- and tetra-chlorinated congeners were the

M.A. Gonzalez Sagrario et al. / Chemosphere 48 (2002) 1113–1122

1117

Fig. 2. Global distribution of PCB from 3 to 8 isomers expressed as relative abundance in macrophytes collected in both stations from Los Padres Lake, Argentina. Bul: bulrush; f.loo: false loosestrife.

main groups in bulrush (74% of total PCBs), dominated by congener 18, 28 and 49 while false loosestrife showed the highest concentrations of penta- and hexa-congener groups (80% of the total PCBs) with 138, 153, 110, 118 and 101 as the main congeners. 3.2. Grass shrimp Whole tissues of grass shrimp showed the lowest PCB values among animal biota (Table 2). Animal biota of different trophic levels exhibited a common PCB signature, dominated by highly chlorinated congeners (5 and 6 Cl) representing 85% of the total PCBs. These congener groups were dominated by isomers 138, 153, 110, 118 and 101. 3.3. Fish species Pre-spawning female of O. jenynsi, a low-fat fish, accumulated the highest PCB levels (ng/g wet wt.) in ovary, followed by liver and muscle. The main congener groups in liver and muscle were tetra-, penta- and hexachlorinated while in ovary penta-, hexa- and hepta. On the other hand, post-spawning organisms of R. sapo presented mesenteric fat, with the highest PCB concentration (ng/g wet wt.) in both sexes. PCB levels in prespawning ovary of R. sapo was much higher than in post-spawning. This behavior was not observed in males. Muscle and liver of R. sapo females (pre- and post-) were dominated by tri-, penta- and hexa-chlorinated congener groups while penta-, hexa- and heptawere accumulated by males (Fig. 3). In mesenteric fat of

post-spawning organisms and pre-spawning gonads a predominance of penta-, hexa- and hepta-congener groups was observed. When PCB values were expressed on ng/g lipid, the muscle tissues of pre-spawning females of both fish species showed the highest concentrations, being R. sapo 2X-fold higher than O. jenynsi, (Table 2). Tissues of both fish species showed a similar congener group pattern when they were expressed either in ng/g wet wt. or ng/g lipid wt. (Fig. 4). Differences between fish species were noted regarding lower chlorinated congeners. Thus, congeners 49 and 52 were detected in high concentration in muscle and liver tissues of O. jenynsi, while congener 18 in muscle and liver of R. sapo prespawning females (Table 3). O. jenynsi/grass shrimp BMF had a mean value of 18.7. Congeners 44, 52 and 151 showed the highest BMF values.

4. Discussion 4.1. Macrophytes It has been shown that macrophytes are involved in the uptake, bioconcentration and movement of organochlorine contaminants in aquatic ecosystems as well cycling of essential nutrients and increasing sedimentation of suspended particles within macrophyte beds (Lovett-Doust et al., 1994; Biernacki et al., 1995b). Painter (1990) found that PCB levels could be 3–4 times higher in plants than in sediments and 6000–9000 times higher in plants than in the water around them (on

1118

M.A. Gonzalez Sagrario et al. / Chemosphere 48 (2002) 1113–1122

Table 3 Concentrations of the different PCB congeners on lipid basis (ng/g lipid wt.) in organisms from Los Padres Lake, Argentina IUPAC #

S. californicusa

Ludwigia sp.a

S1

S2

S1

S2

Tri18 31 28

0.3 0.5 n.d.

1.3 0.2 0.3

n.d. n.d. n.d.

0.1 0.1 n.d.

Tetra52 49 47 44 66

n.d. n.d. n.d. n.d. 0.1

0.1 0.3 n.d. 0.3 <0.1

n.d. n.d. n.d. n.d. 0.2

Penta101 99 87 110 118 105

0.4 0.2 0.3 0.9 0.3 0.1

0.1 <0.1 <0.1 0.1 0.1 <0.1

Hexa149 151 153 156 138

0.2 0.1 0.3 <0.1 0.9

Hepta180 170 Octa199 195 194 a b

P. argentinus

O. jenynsib

R. sapob

Ovary

Liver

Muscle

Ovary

Liver

Muscle

1.2 3.0 6.2

n.d. 13.1 n.d.

24.3 10.1 n.d.

n.d. 50.6 n.d.

n.d. n.d. n.d.

138.2 15.6 n.d.

1338.6 795.3 n.d.

n.d. n.d. n.d. 0.1 0.1

3.6 n.d. 0.8 1.0 7.2

45.2 12.3 n.d. 13.7 32.9

40.0 108.5 n.d. 11.5 13.4

231.5 157.3 n.d. 66.3 56.2

7.2 n.d. 6.2 n.d. 8.3

11.8 5.4 36.0 21.5 14.0

148.8 185.8 n.d. n.d. 79.5

0.6 0.5 0.2 0.7 1.6 0.6

0.4 0.1 0.2 0.7 0.3 0.2

21.6 8.3 9.9 41.3 32.7 13.9

155.4 82.0 40.7 206.1 220.0 62.6

45.3 19.2 16.4 61.5 53.4 18.6

210.1 75.3 82.0 146.1 221.4 77.5

42.4 21.7 14.9 57.6 76.5 25.6

41.9 18.8 16.7 62.4 64.0 22.0

301.6 147.2 97.6 276.4 443.3 138.6

<0.1 <0.1 <0.1 <0.1 <0.1

0.4 0.2 4.5 0.3 3.5

n.d. 0.3 0.4 <0.1 0.6

14.3 0.9 59.3 2.2 61.7

153.6 72.2 610.6 18.6 519.2

39.1 18.4 135.6 4.1 125.3

179.8 56.2 542.7 18.0 518.0

38.2 9.5 224.6 12.6 180.2

34.4 63.4 166.1 4.8 135.0

255.1 100.0 1603.2 45.7 1106.3

0.1 <0.1

<0.1 <0.1

2.3 1.1

0.2 0.1

19.1 10.4

292.8 144.4

61.5 27.1

246.1 113.5

95.5 51.0

82.8 27.4

918.9 203.2

<0.1 n.d. n.d.

<0.1 <0.1 <0.1

0.3 0.1 0.2

<0.1 n.d. <0.1

2.8 0.6 1.5

38.9 13.5 27.6

8.1 2.6 4.9

29.2 12.4 n.d.

11.0 3.5 8.5

9.1 3.8 6.5

81.1 22.1 76.4

ng/g wet wt. pre-spawning female.

a dry wt. basis). In our work, bulrush (S. californicus), a littoral submerged macrophyte, was exposed to significant amount of PCBs in situ, reflecting sediment, water and air loads, whereas false loosestrife (Ludwigia sp.), a rooted floating-leaved macrophyte, mainly exposed to water and air load. False loosestrife accumulated higher PCB levels (wet wt.) than bulrush, emphasizing the importance of lipid content as a determinant of PCB concentration in macrophytes. Total PCB concentrations in both macrophytes from Station 1 were higher than those from Station 2, showing that the former would receive PCBs washed down from upstream areas. Gonzalez Sagrario (1998) also found differences of this nature in the same region for organochlorine pesticides. The relative enrichment of the lower chlorinated congeners (3 and 4 Cl) in bulrush

from Station 2 could be a consequence of: firstly, higher chlorinated congeners are more hydrophobic and persistent than lower ones and tend to accumulate in sediments while low chlorinated congeners are more likely to be transferred in water column and be transported to Station 2. Secondly, bulrush roots reach up to 40–50 cm of depth being able to uptake tri- and tetra-chlorinated congeners as a product of reductive dechlorination of the more highly chlorinated congeners because of anoxic characteristics of this station. Although it has been demonstrated the existence of various Phase I and Phase II detoxification enzymes in aquatic macroalgae and terrestrial plants (Trapp et al., 1990; Pflugmacher et al., 1999), no metabolism of PCB has been recorded in vascular aquatic plants. The PCB loads in aquatic macrophytes constitute the main routes

M.A. Gonzalez Sagrario et al. / Chemosphere 48 (2002) 1113–1122

1119

4.2. Animal biota

Fig. 3. Distribution of PCB congener groups in gonads, liver, muscle and mesenteric fat of R. sapo organisms with different gonadal development, expressed in ng/g wet wt. pre-fem: prespawning female; post-fem: post-spawning female; pre-male: pre-spawning male; post-male: post-spawning male.

for the introduction of these compounds into the food web, when they are consumed as live plants by herbivores or as detritus by detritivores.

Grass shrimp, in Los Padres Lake, lives associated with bulrush and it could obtain PCBs from water, ingestion of detritus and predation over periphytic cover of macrophytes and small invertebrates. Its relatively short life cycle, as well as the known lower bioaccumulation capacity of crustaceans, comparing with that reported by Porte and Albaiges (1993) for mollusc and fish for organic contaminants, such as PCBs, would be two factors that justify the low PCB levels found in grass shrimp. Nevertheless, Porte and Albaiges (1993) have recognized the ability of crustaceans under chronic exposure to aryl pollutants, to develop or induce their mixed function oxygenases (MFO) enzymatic system, for biotransformating PCBs. They have demonstrated that the rate of metabolic activities was in the following order: crab > tuna > mullet > mussel. The levels of total PCBs found in grass shrimp were comparatively lower than those in giant red shrimp from Suruga bay, Japan (Lee et al., 1997) and in P. pugio from Purvis Creek, USA (Maruya and Lee, 1998). In our work, congeners 138, 153, 110, 118 and 101 represent 64% of the total PCBs found in grass shrimp. Excluding 110 and 101, the rest of congeners share the recalcitrant 4-40 -substitution or no vicinal Hatoms in both aromatic rings and therefore would be accumulated and not metabolized. Body burden of PCBs in O. jenynsi could arise from suspended particles and its main prey, grass shrimp while in R. sapo can arise from direct contact with sediments contaminated, sediment ingestion and food chain routes. Muscle tissue of R. sapo showed higher PCB levels than O. jenynsi likely due to the fact that R. sapo was exposed to higher PCB loads. PCB concentrations in gonads of R. sapo varied concomitantly with their lipid levels, though this fact was not found in muscle, reflecting a non-equilibrium situation in which its lipid levels have changed more rapidly than PCB load (Kelly and Campbell, 1994). Moreover, in these relatively lean tissues it can be speculated that proteins and other non-lipid cellular components contribute substantially to chemical partitioning (Bertelsen et al., 1998). The growth of fish ovaries prior to spawning and the increase of their lipid content, would enhance the capacity for the PCB uptake. This behavior would justify the high concentration found in pre-spawning ovaries of both fish species. The low PCB levels found in postspawning ovaries of R. sapo would be a consequence of eggs releasing during spawning, which have large quantities of lipid soluble contaminants. This detoxification activity would be more important than that carried out by MFOs in the liver (Von Westernhagen et al., 1995). By the other hand, since testes constitute a minor fraction of total body of fish and have a low lipid

1120

M.A. Gonzalez Sagrario et al. / Chemosphere 48 (2002) 1113–1122

Fig. 4. Distribution of PCB congener groups in ovary, liver and muscle in pre-spawning females of both fish species, expressed in ng/g lipid wt.

content with respect to the ovary, it would be justified the low detoxification capacity carried out by males (Von Westernhagen et al., 1995). As a result, generally the PCB tissue burdens are lower in females than in males (Lanfranchi et al., 1998). The same congener group patterns found in liver and muscle of R. sapo organisms would indicate that muscle tissue reflects in reliable way the PCB uptake by diet, because of the low detoxification rate of this tissue. The absence of mesenteric fat in pre-spawning R. sapo organisms would indicate that this tissue has been used as energetic source during gonad maturation period. As a result, large amounts of PCBs would be released into the bloodstream and reach the other targets organs, such as gonads, liver or muscle tissue. In fact, higher PCB levels were found in pre-spawning tissues of R. sapo females compared with post-spawning ones. Mesenteric fat and pre-spawning ovary showed the same congener group pattern, being hexa > penta > hepta. Since males did not accumulate high lipid content in prespawning gonads, the distribution of PCB congeners from mesenteric fat, was mainly delimited to muscle and liver tissues. The congener group distribution among tissues of both fish species, would be mainly determined by the lipid composition. Thus, mesenteric fat rich in non-polar lipids have largely accumulated hydrophobic congeners

(penta-, hexa- and hepta) with Kow > 6. According to, muscle tissue, with low lipid content, usually enriched by polar lipids accumulated the highest levels of tri- and tetra-congener groups compared with other tissues. Our study showed that the distribution of PCB congeners in the lake did not occur in a predictable manner, since there is no indication of an increase in the proportion of higher chlorinated congeners in higher trophic levels. BMF for predation–prey relationship O. jenynsi– grass shrimp had a mean value of 18.7. Maruya and Lee (1998) have reported mean BMF values in the order of 12 for the shrimp–mullet couple. Polychlorinated biphenyls uptake and dietary absorption efficiency have been reported to be essentially equivalent for Cl2–Cl10 congener groups (Niimi, 1996). PCB elimination rates in fish were found to decrease with increasing number of chlorine while for less chlorinated PCB elimination rates decrease with increasing ortho-chlorine substitution (Niimi and Oliver, 1983). Accordingly, in our work congener 44, 49 and 52, which have ortho-chlorine substitution in both aromatic rings, showed the highest values between low congener groups and the BMF values of congener 44 and 52 were the highest. About to congener 49, the BMF value was not determined because of in grass shrimp was not detected.

M.A. Gonzalez Sagrario et al. / Chemosphere 48 (2002) 1113–1122

Some penta- and hexa-chlorinated congeners, lacking 4,40 -chlorinated substitution like congener 151, were found to be remarkably low in crustacean (Duinker et al., 1983). In our work, congener 151 was the lowest value found in grass shrimp among the highly chlorinated congeners and the BMF value was found among the highest although the level of this congener in O. jenynsi was relatively low.

5. Conclusions First results on the PCB levels in freshwater organisms from Los Padres Lake are presented. The bioconcentration process in macrophytes appears to be influenced by a number of factors including species, lipid content and the area where they live as well as physical conditions of the lake. There was a relationship between the PCB concentrations in the animal biota and either trophic level or lipid content of the organisms. Our results indicate that the patterns of relative abundance of PCB congener groups did not change among animal biota. The highest concentrations were found for penta- and hexa-chlorinated congeners, consistent with commercial mixture of Aroclor 1254 used in this region. PCB concentrations in fish tissues varied with the species and the gonadal development. Pre-spawning ovaries of both fish species showed high PCB levels, as a result of growth and increase of lipid content priori to spawning. On the other hand, during spawning an important load of highly chlorinated congeners has been lost, becoming in a great detoxification way conducted by females. Biomagnification of the most PCB congeners has occurred, even though they did not clearly displayed an increasing trend with increasing log Kow . In order to have a more complete comprehension of the PCB behavior in Los Padres Lake, it would be necessary to extend sampling to other matrix, like bottom sediments, for understanding the precise nature of the recycling process in aquatic systems.

Acknowledgements The authors thank Mrs M onica Thomas for drafting the map. All analytical work was conducted at Environmental and Resources Studies Program, Trent University, Peterborough, Ontario, Canada. We thank Dr. C.D. Metcalfe for the high professional standards and attention to detail in overseeing the analyses. This study was supported by grants (PMT-PICT 0363 Prest. BID 802/OC-AR, EXA 072/97 and EXA 105/97) from CONICET/Agency and SIyDT (Secretarıa de Investi-

1121

gaci on y Desarrollo Tecnol ogico) of Mar del Plata University, respectively.

References Barron, M.G., 1990. Bioconcentration Environ. Sci. Technol. 24, 1612–1618. Bertelsen, S.L., Hoffman, A.D., Gallinat, C.A., Elonen, C.M., Nichols, J.W., 1998. Evaluation of Log Kow and tissue lipid content as predictors of chemical partitioning to fish tissues. Environ. Toxicol. Chem. 17 (8), 1447–1455. Biernacki, M., Lovett-Doust, J., Lovett-Doust, L., 1995b. Sediment effects on the uptake of trichloroethylene by roots and leaves in Vallisneria americana. In: Lake Huron Ecosystem: Ecology Fisheries and Management. Ecovision World Monograph Series. SPB Academic Publishing, The Netherlands. Bocanegra, E.M., Martınez, D.E., Massone, H.E., Cionchi, J.L., 1993. Explotation effects and salt water intrusion in the Mar del Plata aquifer Argentina. In: Study and Modeling of saltwater, Intrusion into Aquifer. In: Custodio, E., Galofre, E. (Eds.), Proceedings of the saltwater intrusion meeting, Barcelona, Spain, 1–6 November 1992. CIMNE Ed., pp. 177–191. Duinker, J.C., Hillebrand, M.T.J., Boon, J.P., 1983. Organochlorines in benthic invertebrates and sediments from the Dutch Wadden Sea, Netherlands. J. Sea Res. 17, 19–38. Fisk, A.T., Norstrom, R.J., Cymbalisty, C.D., Muir, D.C.G., 1998. Dietary accumulation an depuration of hydrophobic organochlorines: bioaccumulation parameters and their relationship with the octanol/water partition coefficient. Environ. Toxicol. Chem. 17 (5), 951–961. Gonzalez Sagrario, M.A., Aizp un De Moreno, J.E., Moreno, V.J., Escalante, A.H., 1998. Dynamic of organochlorine compounds in different trophic levels of Los Padres pond, Argentina. Environ. Sci. 6, 153–170. Keith, L.H., Crumett, W., Wentler, G., 1983. Principles of environmental analysis. Anal. Chem. 55, 2210–2218. Kelly, A.G., Campbell, L.A., 1994. Organochlorine contaminants in liver of cod (Gadus morhua) and muscle of herring (Clupea harengus) from Scottish waters. Mar. Pollut. Bull. 28 (2), 103–108. Lanfranchi, A.L., Aizp un de Moreno, J.E., Moreno, V.J., Metcalfe, T., Menone, M., 1998. Distribution of organochlorine compounds in tissues of croacker (Micropogonias furnieri) from Samboromb on Bay, Argentina. Environ. Sci. 6 (1), 55–67. Lee, J.S., Tanabe, S., Takemoto, N., Kubodera, T., 1997. Organochlorine residues in deep-sea organisms from Suruga bay, Japan. Mar. Pollut. Bull. 34 (4), 250–258. Lovett-Doust, L., Lovett-Doust, J., Biernacki, M., 1994. American wildcelery, Vallisneria americana as a biomonitor of organic contaminants in aquatic ecosystems. J. Great Lakes Res. 20, 333–354. Maruya, K.A., Lee, R.F., 1998. Biota-sediment accumulation and trophic transfer factors for extremely hydrophobic polychlorinated biphenyls. Environ. Toxicol. Chem. 17 (12), 2463–2469. Menone, M.L., Aizp un de Moreno, J.E., Moreno, V.J., Lanfranchi, A.L., Metcalfe, T.L., Metcalfe, C.D., 2001.

1122

M.A. Gonzalez Sagrario et al. / Chemosphere 48 (2002) 1113–1122

Organochlorine pesticides and PCBs in a southern Atlantic coastal lagoon watershed, Argentina. Arch. Environ. Contam. Toxicol. 40, 355–362. Metcalfe, T.L., Metcalfe, C.D., 1997. The trophodynamics of PCBs including mono and non-ortho congeners in the food web of north-Central Lake Ontario. Sci. Total Environ. 201, 245–272. Muir, D.C.G., Norstrom, R.J., Simon, M., 1988. Organochlorine contaminants in Arctic marine food chains: accumulation of specific polychlorinated biphenyls and chlordane related compounds. Environ. Sci. Technol. 22, 1071–1079. Niimi, A.J., Oliver, B.G., 1983. Biological half-lives of polychlorinated biphenyl (PCB) congeners in whole fish and muscle of rainbow trout (Salmo gairdneri). Can. J. Fish Aquat. Sci. 40, 1388–1394. Niimi, A.J., 1996. Evaluation of PCBs and PCDD/Fs retention by aquatic organisms. Sci. Total Environ. 192, 123–150. Painter, S., 1990. Ecosystem health and aquatic macrophytes: status and opportunities. Symposium of the aquatic ecosystem health and management society, July. Pastor, D., Fern andez, V., Albaigas, J., 1996. Bioaccumulation of organochlorinated contaminants in three estuarine fish species––Mullus barbatus, Mugil cephalus and Dicentrarcus labrox. Mar. Pollut. Bull. 32, 257–262.

Pflugmacher, S., Wiencke, C., Sandermann, H., 1999. Activity of phase I and phase II detoxification enzymes in Antarctic and Arctic macroalgae. Mar. Environ. Res. 48, 23–36. Porte, C., Albaiges, J., 1993. Bioaccumulation patterns of hydrocarbons and polychlorinated biphenyls in bivalves, crustaceans and fishes. Arch. Environ. Contam. Toxicol. 26, 273–281. Russell, R.W., Gobas, F.A.P.C., Haffner, G.D., 1999. Role of chemical and ecological factors in trophic transfer of organic chemicals in aquatic food webs. Environ. Toxicol. Chem. 18 (6), 1250–1257. Trapp, S., Matthies, M., Scheunert, I., Topp, E., 1990. Modeling the bioconcentration of organic chemicals in plants. Environ. Sci. Technol. 24, 1246–1252. Van der Oost, R., Heida, H., Opperhuizen, A., 1988. Polychlorinated biphenyl congeners in sediments, plankton, molluscs, crustaceans, and eel in a freshwater lake: implications of using reference chemicals and indicator organisms in bioaccumulation studies. Arch. Environ. Contam. Toxicol. 17, 721–729. Von Westernhagen, H., Cameron, P., Janssen, D., Kerstan, M., 1995. Age and size dependent chlorinated hydrocarbons concentrations in marine teleosts. Mar. Pollut. Bull. 30 (10), 655–659.