Age dependence of the accumulation of organochlorine pollutants in brown trout (Salmo trutta) from a remote high mountain lake (Redó, Pyrenees)

Environmental Pollution 133 (2005) 343–350

www.elsevier.com/locate/envpol

Age dependence of the accumulation of organochlorinepollutants in brown trout (Salmo trutta) from a remote

high mountain lake (Redo, Pyrenees)

I. Vivesa, J.O. Grimalta,*, M. Venturab, J. Catalanb, B.O. Rosselandc,d

aDepartment of Environmental Chemistry, Institute of Chemical and Environmental Research (CSIC), Jordi Girona 18,

08034-Barcelona, Catalonia, SpainbLimnology Group (CSIC-UB), Centre for Advanced Studies of Blanes (CEAB-CSIC), Acces Cala St. Francesc, 14,

Blanes 17300, Catalonia, SpaincNorwegian Institute for Water Research (NIVA), P.O.B. 173 Kjelsaas, N-0411 Oslo, Norway

dInstitute for Biology and Nature Conservation, The Agricultural University of Norway (NLH), Norway

Received 7 July 2003; accepted 28 May 2004

Trout in high mountain lakes display age-dependent accumulation of certain organochlorine pollutants.

Abstract

Polychlorobiphenyls (PCBs), hexachlorocyclohexanes (HCHs), hexachlorobenzene (HCB) and DDT were examined in themuscle of brown trout (Salmo trutta) from a high mountain lake located in the Pyrenees (Catalonia, Spain) that was used as a model

of these lacustrine environments. Results indicate that fish age is the main factor of variability among specimens in this populationthat is subjected to atmospheric inputs of the organochlorine compounds (OC). Increases of 2- and 20-fold between fish aged 1 yearand 15 years old are found. The observed pattern cannot be explained in terms of fish size, condition factor, or muscle lipid content.Higher molecular weight compounds (higher lipophilicity) are better correlated with fish age than low molecular weight compounds.

A transformation from 4,40-DDT to 4,40-DDE occurs in fish after ingestion; this results in amplified age-dependent signals,especially in male specimens. In contrast, PCB congener #180 has lower age dependence than the general OC group, which could bedue to its high hydrophobicity (log KowO 7). In any case, selective accumulation of hydrophobic compounds is already observed

among younger fish (age, 1 year). Due to this effect, the relative OC composition does not reflect the main OC pollutants in the lakewaters.� 2004 Elsevier Ltd. All rights reserved.

Keywords: High mountain areas; Fish; Organochlorine compounds; Age dependence; Bioconcentration

1. Introduction

Organochlorine compounds (OC), such as polychlor-obiphenyls (PCBs), hexachlorocyclohexanes (HCHs),hexachlorobenzene (HCB) and DDTs, are ubiquitous

* Corresponding author. Tel.: C34 934006122; fax: C34

932045904.

E-mail address: [email protected] (J.O. Grimalt).

0269-7491/$ - see front matter � 2004 Elsevier Ltd. All rights reserved.

doi:10.1016/j.envpol.2004.05.027

contaminants on our planet. Once released into theenvironment they may be transported over long distan-ces (Wania and Mackay, 1996) and be incorporated intomany biogeochemical cycles without undergoing impor-tant degradation. Despite the discontinued use of manyof them, their presence in cold and remote sites, such ashigh mountain lakes, has been documented (Grimaltet al., 2001; Vilanova et al., 2001a,b) where they aretrapped by condensation due to low temperatures

344 I. Vives et al. / Environmental Pollution 133 (2005) 343–350

(Grimalt et al., 2001). Moreover, organochlorine com-pounds tend to accumulate in organic tissues due totheir lipophilicity and persistence to degradation (deVoogt and Brinkman, 1989). The combination of thesetwo effects may eventually result in concentration levelsthat are toxic for the organisms living in these sites. Theaccumulation of OC in fish tissues may result fromdirect water intake (bioconcentration) and/or from preyingestion (biomagnification).

Bioconcentration and biomagnification ultimatelydepend on the octanol–water partition coefficient(Kow) of each compound (Chiou, 1985; Mackay,1982; Sijm et al., 1992; Hawker and Connell, 1988;Burreau et al., 1997; Fisk et al., 1998) but as observedfrom field data, the relationship between these pro-cesses and Kow is not straightforward (Swackhamerand Hites, 1988; Thomann and Connolly, 1984).Moreover, fish biology may also be relevant for OCaccumulation, e.g. species, sex, age, reproductive stage,trophic status (Rosseland et al., 1999; Rognerud et al.,2002). Therefore, OC accumulation in fish depends ona large number of biological factors which complicatesthe environmental significance of the observed concen-trations.

High mountain lakes offer unique environments forthe assessment of some of these biological factors sincethey contain controlled populations of fish that havebeen exposed exclusively to known OC inputs. Theseecosystems offer ‘‘natural experiments’’ of exposure tolow pollution inputs in real environments. In the presentpaper, Lake Redo (Pyrenees, Catalonia, Spain) has beenselected for study (42 �380N, 0 �460E). This lake (7.7 hm3)is situated at 2240 m above sea level, is oligotrophic, hasa surface area of 24 ha, a maximum depth of 73 m, anda water residence time of 4 years (Ventura et al., 2000).The ice-free period is from May to December (Catalan,1992). Its watershed is small (155 ha) and scarcelyvegetated. Pollution inputs are exclusively related toatmospheric deposition (wet and dry) and there is onlyone outflow. Having in mind previous studies (Grimaltet al., 2001), this lake can be taken as a model exampleof these lacustrine water bodies in high mountainsystems.

The lake contains a large population of brown trout(Salmo trutta) in which specimens between 1 year and 15years have been collected. These fish are on top of thetrophic food web but do not contain piscivorousspecimens (Rognerud et al., 2002). The inputs of PCBs,HCB, HCHs and DDTs entering into this lake (Carreraet al., 2002) as well as the OC composition of the waters(Vilanova et al., 2001a,b) have been determined inprevious studies. The study of specimens (nZ 29) fromthe same lake avoids geographical differences in OCinput and provides a good case for the evaluation of theage and sex dependence of the accumulation of thesecompounds in fish.

2. Materials and methods

2.1. Sample collection and handling

Fish sampling followed standard test fishing proce-dures with multifilament gillnets (Rosseland et al., 1997)and tissue sampling followed the EMERGE protocol(Rosseland et al., 2001). All fish were measured,dissected and sexed on site. Muscle fillets were wrappedin a pre-cleaned aluminium foil and kept frozen untilanalysis. Fish were aged using otoliths and scales(Rosseland et al., 1997).

Trophic status was determined for individual fish bymeasuring the relation between the light isotopes ofnitrogen d15N (15N/14N) (Minagawa and Wada, 1984;Rognerud et al., 2002).

2.2. Chemicals

n-Hexane, dichloromethane, iso-octane, methanol,concentrated sulphuric acid 95–97%, acetone, andanhydrous sodium sulphate for residue analysis werefrom Merck (Darmstadt, Germany). Aluminium foilwas rinsed with acetone and dried at ambient temper-ature prior to use. Cellulose extraction cartridges of20 mm i.d. and 80 mm long were from Whatman(England). The purity of the solvents was checked bygas chromatography coupled to electron capture de-tection (GC–ECD). Sodium sulphate and cartridgeswere pre-cleaned by Soxhlet extraction with dichlor-omethane:methanol (2:1, v/v) for 24 h before use.Sodium sulphate was activated overnight at 400 �C.

g-Hexachlorocyclohexane (g-HCH) and tetrabromo-benzene (TBB) were from Aldrich-Chemie (Steinheim,Germany). a-HCH and PCBs (#28, #52, #101, #118,#138, #153, #180, #209) were from Promochem (Wesel,Germany), and 4,40-DDE and 4,40-DDT were fromDr. Ehrenstorfer (Augsburg, Germany).

2.3. Organochlorine compound analysis

Muscle tissues were extracted and analysed for OCusing the method described in Berdie and Grimalt(1998). Briefly, muscle tissue (5 g) was ground withactivated sodium sulphate until a fine powder wasobtained. This mixture was introduced into cellulosecartridges and Soxhlet extracted with n-hexane:dichloro-methane (4:1) for 18 h. Lipid content was determinedgravimetrically using 20% of the extract. TBB and PCB209 standards were added to the rest of the extractwhich was subsequently cleaned up with sulphuric acid(5 times). All n-hexane solutions were combined andconcentrated by vacuum rotary evaporation (20 �C, 20torr) to small volumes (ca. 500 mL), further concentratedto near dryness under a gentle nitrogen flow andredissolved in 50 mL of iso-octane.

345I. Vives et al. / Environmental Pollution 133 (2005) 343–350

Before chromatographic analysis, an internal stan-dard of tetrachloronaphthalene (TCN) and octachloro-naphthalene (OCN) was added to correct for instrumentvariability. Samples were analysed by GC–ECD(Hewlett–Packard 5890 Series II) with a 50 m!0.25 mm i.d. DB-5 capillary column (J&W Scientific,Folsom, CA) coated with 5% phenyl 95% methylpoly-siloxane ( film thickness 0.25 mm). The instrument wasoperated in splitless mode and the oven temperatureprogram started at 90 �C (held for 1 min) to 120 �C at10 �C/min, and then to 310 �C at 4 �C/min (holding time15 min). Injector and detector temperatures were 270and 310 �C, respectively. Stringent precautions wereobserved for maintenance of the injector under cleanconditions avoiding adsorptions that could deviate thesystem from linearity and increase the limits of detectionand quantification. Helium and nitrogen were used ascarrier (0.33 mL/min) and makeup (60 mL/min) gases,respectively.

Some samples were examined by negative ionchemical ionization mass spectrometry coupled to gaschromatography (GC–MS-NICI) for structural confir-mation of the analysed compounds. A GC system fromAgilent Technologies 6890A coupled to an MS detector5973N was used. The system was equipped with a HP-5MS (30 m! 0.25 mm i.d.! 0.25 mm film thickness)and run under the same oven temperature program asdescribed above. Helium was used as carrier gas (1 mL/min) and ammonia was chosen as ionization gas(1.6! 10�4 Pa). Transfer line and quadrupole temper-atures were 280 and 150 �C, respectively. The selectedion program is reported elsewhere (Chaler et al., 1998).

2.4. Quality assurance

A detailed evaluation of the method used in thepresent study is reported elsewhere (Berdie and Grimalt,1998). Procedural blanks were analysed for every set ofsix samples. The recovery of the surrogate standards(TBB and PCB #209) was calculated for each sample.Identification and quantification of all studied com-pounds were performed by injection of external stand-ards at different concentrations. The relative responsesto TCN and OCN were used in order to correct forinstrumental variabilities and this value was alsocorrected by the recovery of the surrogate standards.

3. Results and discussion

3.1. Fish characteristics

The length of the collected specimens was265G 58 mm (meanG standard deviation), weight was200G 99 g and conditioning factor was 0.98G 0.10. Acontinuous increase in weight and length is observed

when comparing specimens between 20 and 200 g(Fig. 1). Then, the rate of length increase flattens butstill an increase is observed. Similar length–weightdistributions have been observed in other fish such asperch (Le Cren, 1951; Olsson et al., 2000). Theconditioning factors of the collected specimens aregenerally low and not correlated to weight (Fig. 1).

Increases in length ( from 12 to 35 cm) are observedwhen comparing specimens between 1 and 6 years. Inthis age interval, weight also increases from 20 to 450 g(Fig. 1). However, at higher age (between 6 and 15years) both the length and weight remain constant. Norelationship between conditioning factor and age isobserved. Muscle lipid content is generally low (0.5–5%;Fig. 1). Again, no relationship between lipid content inmuscle and age is found.

Average conditioning factors of male and femalespecimens (nZ 18 and nZ 11, respectively) were0.96G 0.11 and 0.99G 0.10, respectively, involving nosignificant difference ( p! 0.05). Average ages of themale and female groups were 7.9G 4.3 yr and 8.5G 5.2yr, respectively, involving again no significant difference( p! 0.05). Correlation of length vs age gives rise tosimilar functions in males (length= 5.2 ln(age)C 18,r2O 0.66) and females (length= 5.5 ln(age)C 15,r2O 0.80). Therefore, the two groups of male andfemale specimens examined do not reflect significantpopulation size or age differences.

Typical d15N in individual fish from Redo variedbetween 2 and 6& in 1999 (Rognerud et al., 2002), witha meanG standard deviation d15N of 3.7C 0.6 and3.9G 1.0 in June and October 2001, respectively,indicating no piscivorous specimen in the population.

3.2. POP levels in muscle tissue

The mean concentrations of HCHs (a-HCH andg-HCH), DDTs (4,40-DDT and 4,40-DDE), HCB, andPCBs (congeners #28C31, #52, #101, #118, #153, #138and #180) in muscle from brown trouts are 1.6G 0.9,19G 13, 0.60G 0.4 and 8.2G 4.8 ng g�1 ww, respec-tively (Table 1). These results are comparable to theconcentrations found for the same species in otherEuropean high mountain lakes and in fish from lowaltitude freshwater systems (Swackhamer and Hites,1988; Andersson et al., 1988; Leiker et al., 1991; Grimaltet al., 2001). The relative standard deviation of mostcompounds, 40–85%, is smaller than that reported inprevious studies.

Significant differences between water and fish com-position are observed (Table 1). Thus, HCHs largelypredominate in the waters but these compounds rangeamong those in lower concentration in fish muscle. Thiscontrast is already observed in the young fish. Thus,even in the group of 1-year-old specimens there is noHCH predominance (Table 1).

346 I. Vives et al. / Environmental Pollution 133 (2005) 343–350

0

5

10

15

20

25

30

35

40

0 2 4 6 8 10 12 14 16age (yr) age (yr)

cm

CONDITIONING FACTORCONDITIONING FACTOR

0.7

0.8

0.9

1.0

1.1

1.2

1.3

0 2 4 6 8 10 12 14 16

age (yr)

LIPID CONTENT

0

1

2

3

4

5

6

0 2 4 6 8 10 12 14 16age (yr)

%

WEIGHT

0

100

200

300

400

20 64 8 10 12 14 16

g

LENGTH LENGTH

10

20

30

40

0 100 200 300 400

0 100 200 300 400

weight (g)

cm

0.7

0.8

0.9

1.0

1.1

1.2

1.3

weight (g)

cg

·cm

-3

cg

·c

m-3

Fig. 1. Characteristics of the fish from Lake Redo included in the study.

Among DDTs, the major OC group, 4,40-DDT isfound in higher abundance than 4,40-DDE in the waterbut the latter is more abundant in fish. Again, thecontrast is already observed among the 1-year-oldspecimens examined (Table 1).

The more chlorinated PCB congeners are moreabundant than the less chlorinated congeners in fishbut the predominant compounds are #52 and #101 inthe waters (Table 1). The relative abundance of the morechlorinated congeners is higher among older (13–15years old) than younger (1 year old) fish. However, evenamong younger fish PCB #138 and #153 predominateover #52 and #101. Thus, incorporation of OC into fishmuscle proceeds through a selective process thataccumulates the more hydrophobic compounds. Thisprocess already occurs at the ages of early development.

3.3. Age dependence

The logarithms of the concentrations of most OCexhibit a significant linear dependence on age (Fig. 2).This dependence is significant at the p! 0.01 level for4,40-DDE, 4,40-DDT and all PCBs except congeners #28and #52 (Table 2). In the case of HCB, the statisticalsignificance is at the p! 0.05 level (Table 2). Only theHCH and the above mentioned PCB congeners #28 and#52 are do not exhibit this trend. HCHs are less lipophilic

than the other compounds log(Kow)= 3.9, and they arenot accumulated irrespective of being the predominantOC in the lake waters (Table 1). PCB congener #28 is alsoless lipophilic than the other PCB congeners consideredin the study. In addition, these two congeners also haverelatively high vapour pressure constants, 10�1.5 and10�1.8, respectively, which are consistent with theirdifferent degree of chlorine substitution (three and fourchlorine atoms, respectively). These two vapour pres-sures indicate that congeners #28 and #52 will be lessretained in the lake waters, and ultimately in fish, thanthe heavier molecular weight PCB.

In all cases, the dependence involves an increase ofOC in the older specimens, representing increases of2–20 times between specimens aged 1 year and 15 yearsold (Table 2). Furthermore, the correlations show acontinuous trend with age. Thus, 5-year-old fish exhibithigher values than 1-year-old fish and similar increasesare observed when comparing between 10 and 5-year-old specimens (Fig. 2). These constant incrementscannot be explained in terms of fish size since lengthand conditioning factor are not continuously correlatedto age (Fig. 1). Likewise, the continuous distributionwith age is consistent with the lack of piscivorous fish inthe lake since abrupt increases of about 3–4 times wouldbe expected when comparing specimens of differenttrophic levels (MacDonald et al., 2002; Madenjian et al.,2002; Rognerud et al., 2002).

347I. Vives et al. / Environmental Pollution 133 (2005) 343–350

Table

1

Averages

andstandard

deviationsoftheconcentrationsofthemajororganochlorinecompoundsin

fish

from

LakeRedo

Compound

Allindividuals

meanG

standard

deviation(ngg�1ww)

1-year-old

fish

meanG

standard

deviation(ngg�1ww)

13–15-year-old

fish

meanG

standard

deviation(ngg�1ww)

Allmale

meanG

standard

deviation(ngg�1ww)

Allfemale

meanG

standard

deviation(ngg�1ww)

Water(dissolved

C

particulated)(pgl�

1)

(Vilanovaet

al.,2001a)

Log(K

ow)(H

awker

andConnell,1988;

Ballschmitterand

Wittlinger,1991)

a-H

CH

0.26G

0.10

0.13G

0.04

0.25G

0.23

0.26G

0.13

0.27G

0.19

410

3.9

g-H

CH

1.4

G0.80

0.60G

0.18

1.4

G1.2

1.3

G0.60

1.5

G0.98

2500

3.9

HCB

0.60G

0.40

0.26G

0.05

0.60G

0.38

0.53G

0.32

0.70G

0.41

8.4

5.5

4,4

0 -DDE

18G

13

1.7

G0.68

30G

11

19G

13

16G

13

8.9

5.6

4,4

0 -DDT

1.2

G0.60

0.41G

0.25

1.6

G0.54

1.2

G0.52

1.2

G0.71

20

6

PCB28

0.14G

0.20

0.06G

0.11

0.14G

0.04

0.15G

0.19

0.13G

0.076

8.6

5.7

PCB52

0.30G

0.30

0.38G

0.32

0.22G

0.08

0.36G

0.42

0.20G

0.13

14

5.8

PCB101

0.94G

0.50

0.38G

0.27

1.1

G0.46

1.0

G0.55

0.83G

0.43

14

6.4

PCB118

0.74G

0.60

0.15G

0.10

0.82G

0.30

0.89G

0.71

0.50G

0.31

5.5

6.7

PCB153

2.4

G1.6

0.48G

0.16

4.3

G1.7

2.5

G1.8

2.2

G1.5

8.9

6.9

PCB138

2.2

G1.4

0.41G

0.16

3.9

G1.1

2.3

G1.3

2.1

G1.5

10

6.8

PCB180

1.5

G1.1

0.35G

0.06

2.9

G0.85

1.5

G1.0

1.5

G1.1

3.6

7.4

Theconcentrationsin

thelakewaterandKoware

alsogiven

forreference.

PCB congeners show higher increments at higherdegree of chlorination ( from four to seven chlorinegroups). However, the compound that exhibits thehighest age-increase is 4,40-DDE (19 times) which hasonly four chlorine groups. The different behaviour ofthis compound suggests that its accumulation in fishfollows metabolic processes other than those for PCB.In this respect, 4,40-DDT does not exhibit an agedependent accumulation similar or higher than 4,40-DDE despite containing five chlorine atoms. Maybepart of the 4,40-DDT incorporated by fish is transformedinto 4,40-DDE and accumulated in the form of this morechemically stable molecule. This transformation mayalso explain the difference between water and fish musclecomposition (Table 1).

3.4. Influence of lipid content

Lipid content in fish muscle is not correlated to age(Fig. 1). The log-transformed lipid-normalized concen-trations show the same significant age-dependent corre-lations (Table 3) as those observed when referring to wetweight content (Table 2). The observed age dependencecannot therefore be attributed to increasing lipidaccumulation with age.

However, both r2 values and age-increases are slightlylower when they are calculated after lipid normalizationthan over wet weight. Thus, introduction of lipid contentin the calculations does not reveal a more defined agetrend but increases in the scatter of the data. This result isexpected in view of the random distribution of lipidcontent with age (Fig. 1) and indicates that OC are storedin fish muscle in all sorts of organic tissues, not only fat.

3.5. Sex differences

Females show age dependence in the accumulation ofall PCBs except congener #28 (Table 2). 4,40-DDT, 4,40-DDE and HCB are also correlated. Males exhibitsimilar trends but the less lipophilic compounds of thisseries, that is PCB congeners #28 and #52, and HCB arenot correlated (Table 2). Since there is no differencebetween the amount of muscle lipids in females(1.93G 1.0%) and males (1.97G 1.2%), the observeddifferences in OC accumulation cannot be attributed todistinct lipid content.

The OC that have age-dependent accumulation inboth sexes generally exhibit higher old/young ratios inmales than in females (Table 2). However, the curvefitted straight lines from which these ratios werecalculated are not statistically significant. Only in thecase of 4,40-DDT, the male–female lines are significantlydifferent at p! 0.1. Previous observations on fishrelated to size have also shown higher accumulationsin males than in females (Olsson et al., 2000) and

348 I. Vives et al. / Environmental Pollution 133 (2005) 343–350

PCB 28

R2 = 0.06380.01

0.10

1.00

50 10 15age

PCB 52

R2 = 0.00010.01

0.10

1.00

10.00

50 10 15age

PCB 101

R2 = 0.44050.1

1.0

10.0

50 10 15age

PCB 153

R2 = 0.78160.1

1.0

10.0

0 5 10 15age

PCB 180

R2 = 0.7625

0.1

1.0

10.0

0 5 10 15age

PCB 138

R2 = 0.78870.1

1.0

10.0

50 10 15age

PCB 118

R2 = 0.4575

0.01

0.10

1.00

10.00

50 10 15age

pp'-DDE

R2 = 0.76221.0

10.0

100.0

0 5 10 15age

HCB

R2 = 0.1997

0.1

1.0

10.0

0 5 10 15age

α-HCH

R2 = 0.0257

0.01

0.10

1.00

0 5 10 15age

γ-HCH

R2 = 0.0851

0.1

1.0

10.0

0 5 10 15age

pp'-DDT

R2 = 0.4402

0.1

1.0

10.0

50 10 15age

Fig. 2. Dependence of the concentrations of the organochlorine compounds (ng g�1 ww) on age ( years).

probably reflect a slightly higher OC detoxificationcapacity of females, e.g. during spawning.

The two DDT compounds exhibit a distinct behav-iour. Thus, whereas the old/young ratio for 4,40-DDE ishigher in males than in females (22 and 16, respectively)the ratio for 4,40-DDT is lower in males than in females

(2.2 and 6.0, respectively; Table 2). The above men-tioned higher detoxification capacity of females thanmales could explain in part these results. Thus, afterspawning, a renewed intake of OC rich in 4,40-DDTfrom the lake could go into female when building againthe egg store. On the other hand, the large difference in

Table 2

Correlation values of the age dependence of the concentrations of the main organochlorine compounds in fish (ng g�1 ww) from Lake Redo

(log(conc.) vs age)

Compound All specimens Female Male

r2 Slope Intercept Old/young

ratioar2 Slope Intercept Old/young

ratio

r2 Slope Intercept Old/young

ratio

a-HCH 0.026 –b – – 0.0093 – – – 0.048 – – –

g-HCH 0.085 – – – 0.196 – – – 0.024 – – –

HCB 0.200* 0.024 2.5 2.2 0.612** 0.040 2.4 3.7 0.029 – – –

4,40-DDE 0.762** 0.091 3.3 19 0.837** 0.087 3.3 16 0.766** 0.096 3.4 22

4,40-DDT 0.440** 0.038 2.7 3.4 0.736** 0.056 2.5 6.0 0.234** 0.024 2.8 2.2

PCB 28 0.064 – – – 0.039 – – – 0.081 – – –

PCB 52 0.0001 – – – 0.289** 0.026 2.1 2.3 0.062 – – –

PCB 101 0.440** 0.045 2.5 4.3 0.359** 0.030 2.6 2.6 0.547** 0.060 2.4 6.8

PCB 118 0.457** 0.066 2.2 8.5 0.583** 0.043 2.2 4.0 0.531** 0.089 2.1 18

PCB 153 0.782** 0.072 2.7 10 0.856** 0.068 2.6 9.1 0.77** 0.076 2.7 12

PCB 138 0.789** 0.073 2.6 10 0.925** 0.075 2.5 11 0.731** 0.072 2.7 10

PCB 180 0.762** 0.068 2.5 8.8 0.781** 0.066 2.5 8.4 0.772** 0.070 2.5 9.5

*Significant at p! 0.05. **Significant at p! 0.01.a Ratio between the concentration at 15 years and 1 year calculated from the curve fitted slopes and intercepts.b When the correlation was not significant, values are not given.

349I. Vives et al. / Environmental Pollution 133 (2005) 343–350

old/young ratio of the two compounds does not allow toexclude that males may have higher capacity of trans-formation of 4,40-DDT to 4,40-DDE than females.

3.6. Kow dependence

A consistent relationship between log(Kow) and theslopes (SCA) of the curve fitted straight lines is observedfor the compounds showing age-dependent accumula-tion in fish (Fig. 3) with the exception of 4,40-DDE. Thistrend is independent of OC water concentrations andreflects higher bioconcentration and/or biomagnifica-tion at higher Kow.

PCB congener #180 shows a lower SCA than expectedfrom its Kow according to the general trend (Fig. 3). Thislower value could be due to difficulties in the transfer ofthis highly hydrophobic compound (log(Kow)O 7) intofish tissues, e.g. reduced membrane passage in the gills(Gobas et al., 1986) and/or in the gastrointestinal tract(Opperhuizen et al., 1985; Gobas et al., 1993), and

Table 3

Correlation values of the age dependence of the lipid-normalized

concentrations of the main organochlorine compounds in fish (ng g�1

lipid) from Lake Redo (log(conc.) vs age)

Compound r2 Slope Intercept Old/young ratioa

a-HCH 0.007 –b – –

g-HCH 0.040 – – –

HCB 0.179** 0.021 4.3 2.0

4,40-DDE 0.681** 0.087 5.1 17

4,40-DDT 0.341** 0.035 4.5 3.1

PCB 28 0.035 – – –

PCB 52 0.002 – – –

PCB 101 0.273** 0.042 4.3 3.9

PCB 118 0.341** 0.063 4.0 7.7

PCB 153 0.698** 0.069 4.5 9.2

PCB 138 0.751** 0.069 4.4 9.4

PCB 180 0.615** 0.064 4.3 7.9

**Significant at p ! 0.01.a Ratio between the concentration at 15 years and 1 year calculated

from the curve fitted slopes and intercepts.b When the correlation was not significant, values are not given.

0

0,02

0,04

0,06

0,08

0,1

0,12

65 87log Kow

SC

A

PCBsDDTsHCB

Fig 3. Correlation between Fig. 2 slopes (SCA) and logarithms of the

octanol–water partition constants.

therefore a high degree of elimination through feces(Gobas et al., 1989) or difficulties for fish absorption dueto low water solubility (Chessells et al., 1992). In bothcases, the ultimate effect would involve lower biocon-centration than expected from the Kow constant.

In contrast, 4,40-DDE deviates from the general trendinvolving a higher SCA than expected from its Kow

(Fig. 3). This higher increase may indicate that 4,40-DDE is more difficult to excrete or metabolize than PCBor HCB (Olsson et al., 2000). However, it may alsoreflect the above-mentioned transformation of 4,40-DDT to 4,40-DDE as a consequence of fish metabolism.

In conclusion, age is the main factor explaining thevariance of OC levels in muscle from a non-piscivorousbrown trout population in a high mountain lake (Redo,Pyrenees) showing increases of 2–20 times between fishaged 1 year and 15 years old. This dependence isobserved for the compounds with log(Kow) O5. In fact,compounds with a low degree of chlorination, e.g. threechlorine atoms, do not show these age-increases. Theincrease for 4,40-DDE is higher than expected whenconsidering the overall dependence of the correlationslopes from log(Kow). The observed values likely reflecttransformation from 4,40-DDT to 4,40-DDE after fishingestion. In contrast, the lower slope values found forPCB congener #180 may reflect steric restrictions for fishintake. Due to this effect, the relative abundances of OCaccumulated in muscle tissue exhibit major differencesfrom those observed in the lake waters (dissolvedC par-ticulate fractions). The observed changes in concentra-tion are not related to fish size, conditioning factor ormuscle lipid content. Higher age-dependent OC accu-mulation is generally observed in males than in femalesbut the differences are not statistically significant.

Acknowledgements

Financial support from the EU project EMERGE(EVK1-CT-1999-00032) is acknowledged. One of us, IV,thanks Generalitat de Catalunya (Catalan AutonomousGovernment) for a Ph.D. fellowship and as a student ofUniversitat Autonoma de Barcelona she acknowledgesit. Partial funding for research on the Pyrenean lakes byGeneralitat de Catalunya is acknowledged. We thank R.Chaler and D. Fanjul for their assistance with the GC–MS instrument. B. Pina (Institute of Molecular Biology,CSIC, Barcelona) is thanked for useful discussion. Wethank N. Garcıa-Reyero (IBMB, CSIC, Barcelona), J.C.Massabuau (University of Bordeaux 1) and R. Lackner(University of Innsbruck) for their valuable help duringsampling and Einar Kleiven (NIVA) for fish ageing.

References

Andersson, O., Linder, C.E., Olsson, M., Reutergardh, L., Uvemo,

U.B., Wideqvist, U., 1988. Spatial differences and temporal trends

350 I. Vives et al. / Environmental Pollution 133 (2005) 343–350

of organochlorine compounds in biota from the Northwestern

Hemisphere. Arch. Environ. Contam. Toxicol. 17, 755–765.

Ballschmitter, K., Wittlinger, R., 1991. Interhemisphere exchange of

hexachlorocyclohexanes, hexachlorobenzene, polychlorobiphenyls,

and 1,1,1-trichloro-2,2-bis( p-chlorophenyl)ethane in the lower

troposphere. Environ. Sci. Technol. 25, 1103–1111.

Berdie, L., Grimalt, J.O., 1998. Assessment of the sample handling

procedures of a man power minimized method for the analysis of

organochlorine compounds in large numbers of fish sampling.

J. Chromatogr. A 823, 373–380.

Burreau, S., Axelman, J., Broman, D., Jakobsson, E., 1997. Dietary

uptake in pike (Esox lucius) of some polychlorinated biphenyls,

polychlorinated naphthalenes and polybrominated diphenyls ethers

administered in natural diet. Environ. Toxicol. Chem. 16, 2508–

2513.

Carrera, G., Fernandez, P., Grimalt, J.O., Ventura, M., Camarero, Ll.,

Catalan, J., Nickus, U., Thies, H., Psenner, R., 2002. Atmospheric

deposition of organochlorine compounds to remote high mountain

lakes of Europe. Environ. Sci. Technol. 36, 2581–2588.

Catalan, J., 1992. Evolution of dissolved and particulate matter during

the ice-covered period in a deep, high-mountain lake. Can. J. Fish.

Aquat. Sci. 49, 945–955.

Chaler, R., Vilanova, R., Santiago-Silva, M., Fernandez, P., Grimalt,

J.O., 1998. Enhanced sensitivity in the analysis of trace organo-

chlorine compounds by negative-ion mass spectrometry with

ammonia as reagent gas. J. Chromatogr. A 823, 73–79.

Chessells, M., Hawker, D.W., Connell, D.W., 1992. Influence of

solubility in lipid on bioconcentration of hydrophobic compounds.

Ecotoxicol. Environ. Saf. 23, 260–273.

Chiou, C., 1985. Partition coefficients of organic compounds in lipid–

water systems and correlations with fish bioconcentration factors.

Environ. Sci. Technol. 19, 57–62.

de Voogt, P., Brinkman, U., 1989. Production, properties and usage of

polychlorinated biphenyls. In: Kimbrough, R.D., Jensen, A.A.

(Eds.), Halogenated Biphenyls, Naphthalenes, Dibenzodioxins and

Related Products. Elsevier Science Publishers B.V. (Biomedical

Division), Amsterdam, The Netherlands, pp. 3–45.

Fisk, A.T., Norstrom, R.J., Cymbalisty, C.D., Muir, D.C.G., 1998.

Dietary accumulation and depuration of hydrophobic organo-

chlorines: bioaccumulation parameters and their relationship with

the octanol/water partition coefficient. Environ. Toxicol. Chem. 17,

951–961.

Gobas, F.A.P.C., Clark, K.E., Shiu, W.Y., Mackay, D., 1989.

Bioconcentration of polybrominated benzenes and biphenyls and

related superhydrophobic chemicals in fish: role of bioavailability

and elimination into feces. Environ. Toxicol. Chem. 8, 231–245.

Gobas, F.A.P.C., Opperhuizen, A., Hutzinger, O., 1986. Bioconcen-

tration of hydrophobic chemicals in fish: relationship with

membrane permeation. Environ. Toxicol. Chem. 5, 637–646.

Gobas, F.A.P.C., Zhang, X., Wells, R., 1993. Gastrointestinal

magnification: the mechanism of biomagnification and food chain

accumulation of organic chemicals. Environ. Sci. Technol. 27,

2855–2863.

Grimalt, J.O., Fernandez, P., Berdie, L., Vilanova, R.M., Catalan, J.,

Psenner, R., Hofer, R., Appleby, P.G., Rosseland, B.O., Lien, L.,

Massabuau, J.C., Battarbee, R.W., 2001. Selective trapping of

organochlorine compounds in mountain lakes of temperate areas.

Environ. Sci. Technol. 35, 2690–2697.

Hawker, D.W., Connell, D.W., 1988. Octanol–water partition

coefficients of polychlorinated biphenyl congeners. Environ. Sci.

Technol. 22, 382–387.

Le Cren, E.D., 1951. The length–weight relationship and seasonal

cycle in gonad weight and condition in the perch (Perca fluviatilis).

J. Anim. Ecol. 20, 201–219.

Leiker, T.J., Rostad, C.E., Barnes, C.R., Pereira, W.E., 1991. A

reconnaissance study of halogenated organic compounds in catfish

from the lower Mississippi river and its major tributaries. Chemo-

sphere 23, 817–829.

MacDonald, R., Mackay, D., Hickie, B., 2002. Contaminant amplifi-

cation in the environment. Environ. Sci. Technol. 36, 456A–462A.

Mackay, D., 1982. Correlation of bioconcentration factors. Environ.

Sci. Technol. 16, 272–278.

Madenjian, C.P., O’Connor, D.V., Stewart, D.J., Miller, M.A.,

Masnado, R.G., 2002. Field estimate of net trophic transfer

efficiency of PCBs to lake Michigan Chinook salmon from their

prey. Environ. Sci. Technol. 36, 5029–5033.

Minagawa, M., Wada, E., 1984. Stepwise enrichment of 15N along

food chains: further evidence and the relation between d15N and

animal age. Geochim. Cosmochim. Acta 48, 1135–1140.

Olsson, A., Valters, K., Burreau, S., 2000. Concentration of

organochlorine substances in relation to fish size and trophic

position: a study on perch (Perca fluviatilis L.). Environ. Sci.

Technol. 34, 4878–4886.

Opperhuizen, A., van der Velde, E.W., Gobas, F.A.P.C., Liem,

D.A.K., van der Steen, J.M.D., 1985. Relationship between

bioconcentration in fish and steric factors of hydrophobic

chemicals. Chemosphere 14, 1871–1896.

Rognerud, S., Grimalt, J.O., Rosseland, B.O., Fernandez, P., Hofer,

R., Lackner, R., Lauritzen, B., Lien, L., Massabuau, J.C., Ribes,

S., 2002. Mercury and organochlorine contamination in Brown

trout (Salmo trutta) and Artic charr (Salvelinus alpinus) from high

mountain lakes in Europe and the Svalbard archipelago. Water Air

Soil Pollut. 2, 209–232.

Rosseland, B.O., Massabuau, J.C., Grimalt, J., Hofer, R., Lackner, R.,

Rognerud, S., Lien, L., 1999. The ecophysiology and ecotoxicology

of fishes as a tool for monitoring and management strategy of high

mountain lakes and lakes and rivers in acidified areas. Zoology 190,

90–100.

Rosseland, B.O., Massabuau, J.-C., Grimalt, J., Hofer, R., Lackner,

R., Raddum, G., Rognerud, S., Vives, I., 2001. Fish ecotoxicology,

The EMERGE Fish Sampling Manual for Live Fish, The

EMERGE Project (European Mountain lake Ecosystems: Region-

alisation, diaGnostic & socio-economic valuation) http://www.

mountain-lakes.org/methods/index.html.

Rosseland, B.O., Rognerud, S., Massabuau, J.C., Hofer, R., Grimalt,

J. 1997. MOLAR. Protocol for fish sampling: testfishing, fish

physiology, fish histology, heavy metals and organic micropollu-

tants. In: Wathne, B. (Ed.), MOLAR. Project Manual. Measuring

and Modelling the Dynamic Response of Remote Mountain Lake

Ecosystems to Environmental Change. A Programme of Mountain

Lake Research, MOLAR. NIVA-Report, Serial No. 3710–97, pp.

149–166.

Sijm, D.T.H.M., Seinen, W., Opperhuizen, A., 1992. Life-cycle

biomagnification study in fish. Environ. Sci. Technol. 26, 2162–

2174.

Swackhamer, D.L., Hites, R.A., 1988. Occurrence and bioaccumula-

tion of organochlorine compounds in fishes from Siskiwit lake, Isle

Royale, Lake Superior. Environ. Sci. Technol. 22, 543–548.

Thomann, R.V., Connolly, J.P., 1984. Model of PCB in the Lake

Michigan lake trout food chain. Environ. Toxicol. Chem. 18, 65–71.

Ventura, M., Camarero, L., Buchaca, T., Bartumeus, F., Livingstone,

M.D., Catalan, J., 2000. The main features of seasonal variability

in the external forcing and dynamics of a deep mountain lake

(Redo, Pyrenees). J. Limnol. 59, 97–108.

Vilanova, R., Fernandez, P., Martınez, C., Grimalt, J.O., 2001a.

Organochlorine pollutants in remote mountain lake waters.

J. Environ. Qual. 30, 1286–1295.

Vilanova, R.M., Fernandez, P., Grimalt, J.O., 2001b. Polychlorinated

biphenyl partitioning in the waters of a remote mountain lake. Sci.

Total Environ. 279, 51–62.

Wania, F., Mackay, D., 1996. Tracking the distribution of persistent

organic pollutants. Environ. Sci. Technol. 30, 390A–396A.