E L S E V I E R The Science of the Total Environment 154 (1994) 237-247

the Science of the Total Environment

Am I m m l ~ m d Jsm=~ tw ~ I tmm~t ~ t * ~ ~ t m P t ~ m m

Abnormally high polychlorinated biphenyl levels in striped dolphins ( SteneUa coeruleoalba) affected by the 1990-1992

Mediterranean epizootic

Alex Aguilar*, Asunci6n Borrell

Department of Animal Biology (Vertebrates), Faculty of Biology, University of Barcelona 08071 Barcelona, Spain

Abstract

PCB concentrations and total lipid content were determined in the blubber and liver of striped dolphins affected by the 1990 morbillivirus epizootic in the Mediterranean Sea, and in the blubber of striped dolphins from the same area sampled with a biopsy dart in 1987-1989 and 1991. PCB levels were found to be significantly higher in the individuals that succumbed to the epizootic than in the 'healthy' population sampled before or after the event. Although recent mobilization of lipid reserves was found to have occurred in some of the diseased dolphins, this had little effect on their PCB blubber concentrations and cannot explain the observed difference with the healthy individuals. Three hypotheses are put forward to explain the apparent link between high PCB levels and mortality caused by the epizootic: (i) depressed immunocompetence caused by PCBs leading to an increase in individual susceptibility to the morbillivirus infection, (ii) mobilization of fat reserves leading to increased PCB levels in blood which, in turn, may produce a liver lesion capable of increasing the individual's susceptibility to the morbillivirus infection, and (iii) previous existence of an unspecific hepatic lesion producing impairment of the liver function which, in turn, could lead to an increase both in tissue PCB concentrations and in individual susceptibility to the morbillivirus infection.

Keywords: Striped dolphin; Stenella coeruleoalba; Epizootic; Pollution; Polychlorinated biphenyls; Distemper; Mediterranean Sea; Mass mortality; Immunosupression

I. Introduction

During 1990 and 1991, an epizootic affected the striped dolphin (Stenella coeruleoalba) popula- tion in the Mediterranean Sea (Aguilar and Raga, 1993). The process started in mid-eastern Spain in July 1990 but soon extended to most of the west- ern Mediterranean basin. From July to December of 1990, more than 700 carcasses of this species

* Corresponding author.

were collected on the beaches of Spain, France and northwestern Italy. However, the total num- ber of animals that died because of the process was estimated to be much higher because no information was available from the coasts of northern Africa, which were also affected by the epizootic, and also because the striped dolphin is an offshore species which reaches its highest den- sity far away from the coast. For this reason, many corpses remained in open waters never reaching the shore and were therefore not in- cluded in the statistics. In 1991 and 1992, sec-

0048-9697/94/$07.00 © 1994 Elsevier Science BV. All rights reserved. SSDI 0048-9697(94)04232-C

238 A. Aguilar, A. Borrell / Sci. Total Environ. 154 (1994) 237-247

ondary outbursts affected the coasts of southern Italy, Sicily, the Greek Islands and northern Turkey again producing substantial mortality of striped dolphins.

A morbillivirus infection producing fatal lesions in the lungs, nervous system and lymph nodes was found to be the primary cause of the death of the dolphins affected by the epizootic (Domingo et al., 1990; Bressem et al., 1991). In recent years, morbillivirus has frequently been recognized as an agent capable of producing major epizootics in marine mammal populations (Kennedy et al., 1988a,b; Osterhaus and Vedder, 1988; Grachev et al., 1989). However, the mortality level associated with each event varied greatly. The reasons for these dissimilarities probably lay in differences in climatic and environmental conditions, previous exposure to the disease, genetic predisposition, and overall population resilience to the viral in- fection.

In the striped dolphins examined during the 1990 outburst, a number of unusual pathological or physiological conditions, difficult to explain by the morbillivirus infection, were observed besides the lesions and physiological alterations undoubt- edly caused by it. Thus, the load of ectoparasites in these dolphins was abnormally high when com- pared to individuals of the same species and region stranded in previous years (Raga et al., 1992). Their nutritive condition was very poor during the first, strongest phase of the epizootic but became normal afterwards (Aguilar et al., 1992), and many individuals presented a lesion of uncertain origin and nature in the hepatic parenchyma but apparently not caused by the action of the morbillivirus (Domingo et al., 1992; Duignan et al., 1992). These unexplained condi- tions have given rise to considerable discussion about the possible existence of triggering factors at the onset and later development of the epi- zootic (Aguilar and Raga, 1990).

Environmental degradation and chemical pol- lution, in particular, have been claimed to play a role in these marine mammal die-offs (Brody, 1989; Eis, 1989; Simmonds and Johnston, 1989; Kuehl et al., 1991). Polychlorinated biphenyls (PCBs), a major group of organochlorine com-

pounds of industrial origin, have been the center of most of these discussions owing to their docu- mented capacity to depress the immune system of mammals and their hepatotoxicity (Loose et al., 1977; Brouwer et al., 1989; Vos and Luster, 1989). However, in none of the cases so far studied could a link be established between these pollu- tants and the die-offs.

It has long been known that the striped dolphin population in Mediterranean waters is exposed to high levels of organochlorine pollutants and that the PCB levels in their tissues are extremely high (Alzieu and Duguy, 1979, Aguilar et al., 1989). Because of this, and also because preliminary analyses in dolphins succumbing to the epizootic showed high levels of PCBs, it has been suggested that the liver lesion observed in these animals may have been produced by a sudden increase in their PCB circulating levels originating after a massive mobilization of lipid reserves rich in these compounds (Domingo et al., 1992).

The objectives of the present paper are three- fold: (i) to establish whether PCB levels in the liver of diseased dolphins were higher than those in the hypodermic fat tissue or blubber, therefore indicating recent mobilization of the pollutant blubber loads and subsequent shedding of PCB into the main blood stream, (ii) to compare the PCB concentrations in the blubber of striped dolphins affected by the 1990 outburst in Spain with those of dolphins of the same species and area sampled before and after the die-off, and (iii) to ascertain whether the poor nutritive condi- tion observed in some diseased dolphins was re- sponsible for significant changes in the PCB lev- els carried by the dolphins.

2. Material

The corpses of 72 striped dolphins dying during the first outbreak of the Mediterranean epizootic in 1990 along the coast of central and northern Spain and the Balearic Islands were necropsied, and biological and tissue samples were collected from them (Fig. 1). To avoid undesired variations in organochlorine pollutant levels because of post-mortem volatilization or decomposition of

45 N -

5W

A. Aguilar, A. Borrell / Sci. Total Entiron. 154 (1994)237-247

0 5E I

10E

I

-45 N

239

40 N - -40 N

35 N - ! ~ AREA COVERED BY THE 1990 OUTBREAK

STRANDINGS

BIOPSIES

-35 N

I I I I 5W 0 5E 10E

Fig. 1. Map showing the region affected by the 1990 epizootic outbreak, the coastline along which diseased striped dolphins were collected, and the areas where biopsies were obtained.

compounds (Borrell and Aguilar, 1990), only fresh carcasses were examined and sampled in the study.

Blubber samples, containing a complete and homogeneous representation of all strata, were excised from the region posterior to the dorsal fin, as is usually recommended for monitoring organochlorine pollutants in cetaceans (Anony- mous, 1985). These tissue samples were used both for organochlorine analyses and for lipid quanti- fication to estimate nutritive condition. Liver samples from the same individuals were collected from the distal portion of the hepatic lobes. Sex was determined upon examination of the gonads.

For comparative purposes, blubber samples ob-

tained from 109 free-ranging striped dolphins us- ing biopsy techniques were also analyzed. The dart used for collecting these samples was of the butterfly valve type, as designed by Aguilar and Nadal (1984), fired from a boat with a spear gun or a crossbow into bowridding dolphins. Biopsies were collected during the period 1987-1989 and in 1991 from dolphins from the Mediterranean waters off central and northeastern Spain, the same area from which the diseased dolphins origi- nated (Fig. 1).

All samples were wrapped in aluminum foil and preserved at -20°C until the moment of the analysis.

240 A. Aguilar, A. Borrell / Sci. Total Environ. 154 (1994) 237-247

3. Methods

3.1. Anatytical For the organochlorine determination, tissue

samples were ground with anhydrous sodium sul- phate in a mortar. The mixture was extracted with n-hexane for 4 h in a Soxhlet apparatus and the extract was concentrated to 40 ml, from which a 10-ml subsample was taken to determine tissue fat content. An aliquot of the remaining extract containing 10 mg of lipid was mixed with sul- phuric acid for the clean-up, following the proce- dures described by Murphy (1972).

After centrifugation and phase separation, the lipid-free extract was concentrated to 1 ml and was injected into a Perkin Elmer Sigma 3B gas chromatograph (injector temperature 250°C), equipped with an electron capture detector of 63Ni (temperature, 3500C), and a Perkin Elmer Sigma 15 computing integrator. For all the analy- sis, a fused-silica capillary column of 0.25-mm i.d., 60-m length, and a stationary phase SPB-1 with a film thickness of 0.25/zm was used. Pure nitrogen at a flow rate of 1 ml/min was used as carrier gas. Temperature was programmed according to the following sequence: Oven temperature at in- jection was 40°C. Oven steady for the first 2 min and then increased from 40 to 160°C at a rate of 25°C/min. Oven maintained at steady tempera- ture for 1 min and then increased from 160 to 250°C at a rate of 2°C/min. From this point until the end of the analytical run, the column re- mained isothermal at a temperature of 250°C.

Heptachlor was used as the internal standard to calibrate fluctuations in working conditions. The identity of the DDT group compounds was confirmed by an alkali conversion to their respec- tive olefins and re-analysis by gas chromatogra- phy. PCBs were identified and quantified by their peak characteristics and retention times in rela- tion to a 1:1 standard mixture of Aroclors 1254 and 1260, and confirmed by their resistance to the chemical derivations detailed above. Eight repli- cates of samples fortified with standards gave the following percentages of recovery for the whole analytical process (mean + coefficient of varia- tion): p,p'-DDE: 72.56 + 15.8%; p,p'-TDE: 87.41

+ 8.4%; o,p'-DDT: 81.91 ___ 8.3%; p,p'-DDT: 98.13 + 7.6%; PCB: 90.97 + 8.0%. Ten replicate chromatographic runs of a mixture of the DDT and PCB standards gave the following coefficients of variation in the quantification: p,p'-DDE: + 7.9%; p,p'-TDE: + 6.5%; o,p'-DDT: + 5.5%;

p,p'-DDT: + 8.2%; PCBs (mean of the eight peaks commonly used in the quantification): + 8.7%.

The nutritive condition was measured through the quantity of lipids present in the blubber, which is considered to be the most reliable condi- tion index in the striped dolphin (Aguilar et al., 1992). For the quantification of lipids, the blubber sample was ground with anhydrous sodium sul- phate and its lipid fraction extracted with n- hexane for 4 h in a Soxhlet apparatus. The extract was concentrated, and the lipid content from an aliquot determined by evaporation under a cold air stream and gravimetry. This method is ap- propriate for extracting neutral lipids, which typi- cally represent up to 95% of all blubber lipid constituents in adult cetaceans (Lockyer, 1991).

3.2. Computational and statistical Because the epizootic followed two distinct

phases, the sample of diseased dolphins was di- vided into two subsamples: one composed of indi- viduals that died during the first 70 days, and another composed of those that died later. PCB levels and nutritive condition were studied sepa- rately for each subsample.

To avoid harm to calves or youngsters when collecting biopsies, biopsy darts were only shot into large individuals. Because of this selective sampling, the comparison of PCB concentrations with diseased dolphins was only carried out con- sidering dead individuals longer than 170 cm in order to reduce potential heterogeneities in body size between the two subsamples.

Because data distributions were found to de- part from normality according a Shapiro-Wilks test (Sprent, 1989), sample groups were compared using Mann-Whitney tests and nonparametric statistics. All statistical calculations were carried out using the SPSS-X statistical package (SPSS Inc., 1983).

A. Aguilar, A. Borrell /Sci. Total Environ. 154 (1994) 237-247 241

4. Results and discussion

4.1. PCB concentration in the blubber of epizootic dolphins

Results of the PCB analyses carried out on the blubber and liver samples collected from the dol- phins that died during the 1990 outbreak show that, expected, overall concentrations were higher in males than in females. This difference con- forms to the pattern of variation of organochlo- rine compounds usually found in marine mammal populations and is explained by the fact that reproductive females transfer substantial amounts of lipophilic pollutants to their offspring during gestation and lactation, therefore reducing their own pollutant loads in relation to males (Aguilar, 1987).

The PCB concentrations found in the overall sample were extremely high, possibly among the highest ever found in wild mammals. Indeed, the western Mediterranean population of striped dol- phins is known to have been carrying, at least during the last decade, high levels of PCBs in their blubber, usually over 300 ppm (Alzieu and Duguy, 1979; Aguilar et al., 1989). Such levels are much higher than those typically found in popula- tions of the same species inhabiting other water masses (Taruski et al., 1975; O'Shea et al., 1980; Tanabe et al., 1984; Kawai et al., 1988; Morris et al., 1989; Loganathan et al., 1990; Cockcroft and Ross, 1991).

On the other hand, striped dolphins appear to accumulate greater organochlorine concentra- tions in their blubber than other small odon- tocetes also inhabiting the Mediterranean Sea, like bottlenose dolphins (Tursiops truncatus), com- mon dolphins (Delphinus delphis) or Risso's dol- phins (Grampus griseus), which are of similar size, food regime and many physiological and biologi- cal traits (Vicente and Chabert, 1978; Alzieu and Duguy, 1979; Aguilar, 1983; Aguilar et al., 1989). The reason for these differences is unclear, al- though it may be caused by differences in the composition of their diet or in a lower capacity of striped dolphins to degrade and excrete xenobi- otics. In the western Mediterranean, striped dol- phins feed mainly on bony fish such as blue

whiting ( Micromesistius poutassou), anchovy ( En- graulis encrasicholus) or bogue (Boops boops), cephalopods such as Histioteuthis reversa, Histio- teuthis bonnellii, Heteroteuthis dispar, Todarodes sagittatus or Ancistroteuthis lichtensteini, and crus- taceans such as Pasifaea multidentata (Wurtz and Marrale, 1991; Pulcini et al., 1992).

Isomer-specific analysis of the congeners that compose the PCB load of these animals, as well as an evaluation of their toxicity, is reported in Kannan et al. (1993).

4.2. Influence of nutritive condition on the PCB blubber concentration of epizootic dolphins

As can be seen in Fig. 2, the 1990 epizootic outburst followed two phases: the first of high virulence (first 70 days) and the second, more protracted, during which the mortality rate was much lower (after the first 70 days). Monitoring of the beaches and collection of carcasses re- mained essentially constant throughout the whole period, for which reason these two phases are considered to reflect an actual variation in the mortality associated with the epizootic.

The characteristics of the dolphins in each pe- riod appeared to differ: those from the first phase showed a poorer body condition than those from the second (Aguilar and Raga, 1993). Because of these differences, it was considered convenient to initially divide the total sample into two subsets, one corresponding to each of these two phases.

% Frequency of strandings 25

15

10

5

0 30 60 90 120 150 180 210 240 270 300

Days

Fig. 2. Distribution of the frequency of striped dolphin strandings during the first outbreak of the Mediterranean epizootic. The arrow shows the estimated cutpoint between the two phases of the event.

242 A. Aguilar, A. Borrell / Sci. Total Environ. 154 (1994) 237-247

PCBs are highly lipophilic compounds that con- centrate in fat-rich tissues. According to Tanabe et al. (1981), about 90% of all organochlorines present in the body of striped dolphins are stored in the hypodermic fat or blubber, a tissue that constitutes the external cover of the animal and serves for energy storage and thermoregulation. In the blubber, PCBs are molecularly bound mostly to neutral lipids and, when these are mobilized, the pollutants may also be partially or totally mobilized (Aguilar, 1987). When this oc- curs, both lipids and lipophilic pollutants are drawn to the liver for catabolism. The liver is the main center for detoxification of xenobiotic subs- tances in the body and, therefore, an increased concentration of both lipids and xenobiotic com- pounds is to be expected there when mobilization of fat reserves and pollutants bound to them o c c u r s .

Fig. 3 shows the relationship between PCB concentrations in the liver and in the blubber of epizootic dolphins. As can be seen, most studied individuals fall above the 1:1 line. This indicates that the distribution of PCBs in the body was not in equilibrium between the different compart- ments at the time of the dolphin's death, accord- ing to the simple mammillary model of distribu- tion that organochlorines typically follow in the body of mammals (Aguilar, 1985). Such an im- balance of PCB concentrations towards the liver suggests recent mobilization of lipid reserves and subsequent shedding of significant amounts of

P C B L i v e r c o n c e n t r a t i o n ( p p m ) 3500

3000 •

2500 • •

2000 • • •

• • 00 • • 1500 • •

O • • • • 1000 ~ • • •

5O0

0 i , = i I i

500 1000 1500 2000 2500 3000 3500 PCB Blubber concentration (ppm)

Fig. 3. Relationship between PCB concentrations (expressed as ppm calculated on a lipid basis) in the blubber and in the liver of striped dolphins which died during the epizootic.

PCBs to the main blood stream in recent times. It is unclear, however, how important such mobiliza- tion had been, and to what extent the PCB levels in the liver of the diseased dolphins had been increased because of that mobilization.

Nevertheless, PCB levels detected in dolphins from the two periods of the epizootic could not be statistically distinguished (according to a Mann-Whitney test). This result indicates that changes in nutritive condition throughout the epi- zootic, although found, had little effect on the concentrations found in the blubber of the dis- eased dolphins.

4.3. Differences between epizootic and 'healthy' dol- phins

Although the nutritive condition of the dol- phins that died during the first phase of the outbreak (first 70 days) was below the levels usu- ally found in apparently healthy dolphins sampled with biopsy darts in the years before and after the epizootic (P < 0.001, according to the Mann- Whitney test), that of the individuals examined during the second phase (after the first 70 days) was indistinguishable from them (Aguilar et al. 1992).

However, irrespective of this variation in nutri- tive condition, the blubber PCB concentration of dolphins dying during the two phases presented, both separately and combined, significant differ- ences with that of the apparently healthy popula- tion (P < 0.001 in all cases, according to the Mann-Whitney test). Because of these results, PCB data from the two phases were combined in subsequent analyses.

Fig. 4 shows the medians and their associated interquartile ranges (the middle 50% values in a group) for the PCB concentrations detected in the striped dolphins found during the epizootic and in the 3 years prior to the event and 1 year after it. As can be seen in the figure, PCB levels in biopsies, which are considered to come from 'healthy' animals, did not show any trend through the years, nor was any significant difference de- tected between them (according to a Kruskal- Wallis test). Levels were usually situated between 100 and 500 ppm (median, 282 ppm), and had a reduced range of variation. In contrast, concen-

A. Aguilar, A. BorreU /ScL Total Entiron. 154 (1994) 237-247 243

PCB (ppm) 3000

2500

2000

1500

1000 ~

5OO

0 1987 1988 1989 1990 1991

Year

Fig. 4. Variation of PCB concentrations (expressed as ppm calculated on a lipid basis) in the dorsal blubber of Mediter- ranean striped dolphins during the period 1987-1991. Samples from 1987-1989 and 1991 were collected from apparently 'healthy' dolphins using a biopsy dart. Samples from 1990 were collected from dolphins affected by the morbillivirus epizootic. For each year, the central horizontal line represents the median, and the upper and lower boundaries of the box the upper and lower quartiles (middle 50% of values in a group). The vertical lines emanating from each box extend to the smallest and largest observations in each sample group that are not considered outliers, i.e. that are less than one quartile range from the end of the box.

trations of PCBs in the epizootic dolphins were much higher (median, 778 ppm) and also had a much higher range of variation; in some of the diseased dolphins, PCB concentrations reached levels over 1000 ppm.

This difference indicates that PCB levels were actually higher in the striped dolphins affected by the epizootic than in the 'healthy' population sampled before or after the event. Because only fresh carcasses were sampled for this study, poor preservation conditions of the diseased dolphins cannot be adduced to explain such differences. Moreover , if any post-mortem changes in organochlorine loads indeed occurred, it should be expected that the residue levels in the dead animals would tend to decrease in relation to the live-sampled ones, rather than to increase (Bor- rell and Aguilar, 1990).

On the other hand, dissimilarities in age or sex composition between the epizootic and healthy dolphins are also unlikely to justify the observed difference in pollutant levels. The sample from the epizootic contains both males and females of variable ages, but neither of these two variables is

known from dolphins sampled with the biopsy dart. However, even considering the most ex- treme of the situation, that is, that all biopsies were taken from adult females, which are the less contaminated segment of the population, the dif- ference cannot yet be explained. Thus, a Mann- Whitney test revealed significant differences (P < 0.001) between epizootic females and the biopsy sample.

Therefore, and taking into consideration the apparent lack of effect of the variation in blubber lipid richness on these results, it can be con- cluded that the difference in PCB concentrations between epizootic and 'healthy' dolphins observed is not an artifact of the sample and cannot be attributed only to variations in nutritive condi- tion. Rather, the cause must reside in some char- acteristic peculiar to the dolphins affected by the epizootic.

4.4. Explanations for the high PCB levels observed in epizootic dolphins

Although a link appears to exist between epi- zootic mortality and abnormally increased blubber PCB levels, it is unclear which mechanism actu- ally establishes that link. It is even unclear which of the two factors is indeed the cause or effect of the other. As shown in Fig. 5, three possible mechanisms have been envisaged to explain the relation observed between increased pollutant levels and susceptibility to the epizootic process.

The first hypothesis considers the recognized fact that PCBs can act as depressors of the im- mune system of mammals (Loose et al., 1977; Thomas and Hinsdill, 1978; Kunita et al., 1985; Vos and Luster, 1989; Brouwer et al., 1989). Thus, the stereochemistry of some PCB congeners and that of the tetrachlorodibenzo-p-dioxins, which at extremely low doses may cause atrophy of the thymus cortex with consequent reduction of resis- tance to viral infections and overall effects on the immune system of mammals (Neubert et al., 1990, 1991), is very similar. Functionally, PCB-mediated immunotoxicity in laboratory animals is charac- terized by suppressed humoral immunity, and administration of PCBs to mammals inhibits anti- body production (Dean et al., 1989). More con- cretely, organochlorine-induced higher suscepti-

244 A. Aguilar, A. Borrell / ScL Total Entiron. 154 (1994) 237-247

Increased PCB levels: cause or effect ?

Cause

1 PCB I levels I

mobilization of fat reserves

depressed immune reaction

liver lesion by PCB

\ /

increased relative mortality of highly polluted individuals

Effect :

chronic or old liver lesion

impaired --., liver

function

increased PCB levels in diseased dolphins

Fig. 5. Hypothetical mechanisms to explain the observed relationship between abnormally high PCB blubber concentrations in striped dolphins affected by the epizootic and their susceptibility to the disease (for explanation, see the text).

bility to viral diseases has been well demonstrated in mammals, including man (Koller and Thigpen, 1973; Imanishi et al., 1980; Wu et al., 1984; Dean et al., 1989). This led Kendall et al. (1992) to suggest that, in common and grey seals surviving the morbillivirus epizootic that devastated north- ern European populations in 1987 and 1988, indi- viduals with higher than average PCB levels may have taken longer to recover from the immuno- suppressive effects of the morbillivirus infection.

The levels detected in many of the diseased dolphins in the Mediterranean were extremely high, and it is therefore quite feasible that, when the virus reached the population, those individu- als with higher PCB levels and their depressed immune function were, for this reason, more sus- ceptible to the disease and suffered a higher mortality. This is supported by the fact that pla- nar PCBs, which are the congeneric fraction with

a higher immunotoxic capacity (Silkworth and Grabstein, 1982) were detected in relatively high concentrations in the tissues of these dolphins (Kannan et al., 1993). Such selective mortality of highly polluted dolphins would explain why the levels detected in epizootic individuals were much higher than in the normal population and, partic- ularly, than the survivors sampled with the biopsy dart in 1991. In this scenario, PCBs would have played a significant role in the development of the disease.

However, anatomopathology studies also showed the presence of a liver lesion in the dolphins affected by the epizootic (Domingo et al., 1992). The role of this lesion in the process has not been ascertained, and it is not known either whether this lesion is generally present in the overall population or was only a characteristic of the dolphins that died during the epizootic. At

A. Aguilar, A. Borrell / Sci. Total Ent~ron. 154 (1994) 237-247 245

high levels of exposure, PCBs, as well as DDT and other organochlorine pollutants, are known to produce various types of liver lesions in mam- mals (Jonsson et al., 1981; Tryphonas et al., 1984; Ward, 1985; Elangbam et al., 1991). Taking this into account, a second hypothesis that can be put forward is that, when dolphins chronically ex- posed to high levels of PCBs faced lack of food and mobilized fat, PCBs went into the main blood stream, reached the liver and produced the hep- atic lesion described (Domingo et al., 1992). It is unclear whether such a lesion produced hepatic dysfunction but, if this were the case, it would not benefit the capacity of the animal to overcome the morbillivirus infection. In this situation, the mortality rate of the animals with higher PCB levels and, therefore, with a severe lesion would be greater than in those carrying lower levels of pollutants. Again, this second hypothesis con- siders that PCBs were at least a partial cause for the high mortality and that these pollutants had an active role in the extent and development of the die-off.

The third hypothesis is almost the reverse of those discussed above. This considers the liver lesion to be the origin of the process and the increased PCB levels of the victims of the epi- zootic a consequence of it. Metabolization of PCBs is mediated by the cytochrome P-450 de- pendent microsome monooxigenase, an enzymatic system which is mainly available in the liver. Because of this, the liver is the main regulator of the degradation and excretion of organochlorines (Safe, 1984; Sipes and Schnellmann, 1987). If a liver lesion, originating from an unknown cause but not necessarily linked to the epizootic process, produced a chronic impairment of the liver func- tion, it is probable that this would lead to long- standing deficiencies in the degradation of PCBs and, therefore, to an increase in the body loads of these compounds.

In this scenario, abnormal PCB levels would be a consequence of a previous disease, rather than a cause. However, this would mean that the liver lesion observed had a significant role in the devel- opment of the disease, that the liver lesion was anterior to the viral infection, and that the sur- vivors of the epizootic did not suffer it, otherwise

their PCB levels would also be abnormally high. Moreover, this third hypothesis appears to be contradicted by the fact that the high ratio 3,3',4,4',5,5'H6CB /3,3',4,4',5-PsCB found in the epizootic dolphins suggests a strong induction of mixed function oxidase enzymes (Kannan et al., 1993), which in turn indicates an active, if not normal, function of the liver.

It is not possible at this stage to decide which of these three hypotheses is the best reason for the higher concentrations of PCBs observed in diseased dolphins. The link may even rest on mechanisms that have not been considered here and which are far from the physiological and pathological conditions taken into consideration in our study. However, whatever the case, abnor- mally high PCB levels were also found in marine mammals that succumbed to other mass mortali- ties although, as in the present case, the interre- lating effect of other conditions impeded clarifi- cation of the actual role of these compounds in the development of the process (Geraci, 1989; Kuehl et al., 1991; Hall et al., 1992). This is sufficient reason to continue further research di- rected at clarifying the role of PCBs in these events and to establishing their mechanism of action in marine mammals.

Acknowledgements

Recovery of carcasses and tissue sample collec- tion was funded by specific grants from Spain's Institute for the Conservation of Nature (ICONA) and the Autonomous Governments of Catalonia and Valencia. Many individuals contributed to the strandings network with their efforts and cooperation, and we would like to express our sincere gratitude to all of them. Pollutant and lipid analysis were possible through funding granted by the CICYT (projects NAT90-1255-E and NAT91-1128-C04-02), ICONA and the secre- tariat for the Environment (MOPT).

References

Aguilar, A., 1983. Spain. Progress report on cetacean research June 1981 to May 1982. Rep. Int. Whaling Commn., 33: 245-247.

Aguilar, A., 1985. Compartmentation and reliability of sam-

246 A. Aguilar, A. Borrell / Sci. Total Environ. 154 (1994) 237-247

piing procedures in organochlorine pollution surveys of cetaceans. Residue Rev., 95: 91-114.

Aguilar, A., 1987. Using organochlorine pollutants to discrimi- nate marine mammal populations: a review and critique of the methods. Mar. Mamm. Sci., 3(3): 242-262.

Aguilar, A. and J. Nadal, 1984. Obtenci6n de biopsias hipod6rmicas de cet~ceos en libertad. Inv. Pesq., 48(1): 23-29.

Aguilar, A. and J.A. Raga, 1990. Mortandad de delfines en el Mediterr~ineo. Politica Cientifica, 25: 51-54.

Aguilar A. and J.A. Raga, 1993. The striped dolphin epizootic in the Mediterranean Sea. Ambio, in press.

Aguilar, A., A. Borrell and W.F. Perrin, 1989. Differentiation of Population Units of Striped Dolphins, Stenella coeruleoalba, and Common Dolphins, Delphinus delphis, in the Eastern North Atlantic Ocean and Western Mediter- ranean Sea. U.S.-Spain Committee for Scientific and Tech- nological Cooperation, Rep. CCB-8609/031 (unpublished), 34 pp.

Aguilar, A., A. BorreU, N. Calzada and E. Grau, 1992. Body fat reserves in striped dolphins examined during the west- eru Mediterranean die-off. In: X. Pastor and M. Simmonds (Eds.), Proceedings of the Mediterranean Striped Dolphin Mortality International Workshop, 4-5 November 1991, Palma de Mallorca, Greenpeace Mediterranean Sea Pro- ject, pp. 47-52.

Alzieu, C. and R. Duguy, 1979. Teneurs en compos6s organochlor6s chez les C6tac6s et Pinnip~des frequentant les c6tes franqaises. Oceanol. Acta, 2(1): 107-120.

Anonymous, 1985. Recommendations for procedures in sur- veys of organochlorine pollution in cetaceans. Rep. Int. Whaling Commn., 35: 150.

Borrell, A. and A. Aguilar, 1990. Loss of organochlorine compounds in the tissues of a decomposing stranded dol- phin. Bull. Environ. Contam. Toxicol., 45: 46-53.

van Bressem, M.F., I.K.G. Visser, M.W.G. Van de Bilt, J.S. Teppema, J.A. Raga and A.D.M.E. Osterhaus, 1991. Mor- billivirus infection in Mediterranean striped dolphins (Stenella coeruleoalba). Vet. Rec., 129: 471-472.

Brody, M., 1989. Explaining sea mammal deaths proves chal- lenging. Am. Soc. Microbiol. News, 56(11): 595-598.

Brouwer, A., P.J.H. Reijnders and J.H. Koeman, 1989. Poly- chlorinated biphenyl (PCB)-contaminated fish induces vi- tamin A and thyroid hormone deficiency in the common seal (Phoca vitulina). Aquat. Toxicol., 15: 99-106.

Cockcroft, V.G. and G.J.B. Ross, 1991. Occurrence of organochlorines in stranded cetaceans and seals from the east coast of southern Africa. In: Cetaceans and Cetacean Research in the Indian Ocean Sanctuary. UNEP Mar. Mamm. Tech. Rep., 3: 271-276.

Dean, J.H., J.B. Cornacoff, G.J. Rosenthal and M.I. Luster, 1989. Immune system: evaluation of injury. In: A. W. Hayes (Ed.), Principles and Methods of Toxicology, Raven Press, New York, 2nd edn., pp. 741-760.

Domingo, M., L. Ferrer, M. Pumarola, A. Marco, J. Plana, S. Kennedy, M. McAliskey and B.K. Rima, 1990. Morbillivirus in dolphins. Nature, 348: 21.

Domingo, M., J. Visa, M. Pumarola, A.J. Marco, L. Ferrer, R. Rabanal and S. Kennedy, 1992. Pathological and immuno- cytochemical studies of morbillivirus infection in striped dolphins (Stenella coeruleoalba). Vet. Pathol., 29: 1-10.

Duignan, P.J., J.R. Geraci, J.A. Raga and N. Calzada, 1992. Pathology of morbillivirus infection in striped dolphins (Stenella coeruleoalba) from Valencia and Murcia, Spain. Can. J. Vet. Res., 56: 242-248.

Eis, D., 1989. Simplification in the etiology of recent seal deaths. Ambio, 18(2): 144.

Elangbam, C.S., C.W. Qualls and A.W. Confer, 1991. Evalua- tion of ultrastructural response to environmental toxicants in wild cotton rats (Sigmodon hispidus). Bull. Environ. Contam. Toxicol., 47: 321-328.

Geraci, J.R., 1989. Clinical Investigation of the 1987-1988 Mass Mortality of Bottlenose Dolphins along the U.S. Central and South Atlantic Coasts. Final report to (U.S.) N.M.F.S., US Navy and the Marine Mammal Commission (unpublished), 63 pp.

Grachev, M.A., V.P. Kumarev, L.V. Mamaev, V.L. Zorin; L.V. Baranova, N.N. Denikina, S.I. Belikov, E.A. Petro, V.S. Kolesnik, R.S. Kolesnik, V.M. Dorofeev, A.M. Beim, V.N. Kudelin, F.G. Nagieva and V.N. Sidorov, 1989. Distemper virus in Baikal seals. Nature, 338: 209.

Hall, A.J., R.J. Law, D.E. Wells, J. Harwood, H.M. Ross, S. Kennedy, C.R. Allchin, L.A. Campbell and P.P. Pomeroy, 1992. Organochlorine levels in common seals (Phoca vit- ulina) which were victims and survivors of the 1988 phocine distemper epizootic. Sci. Total Environ., 115: 145-162.

Imanishi, J., H. Nomura, M. Matsubara, M. Kita, S. Won, T. Mizutani and T. Kishida, 1980. Effect of polychlorinated biphenyl on viral infections in mice. Infec. Immunol., 29(1): 275-277.

Jonsson, H.T., E.M. Walker, W.B. Greene, M.D. Hughson and G.R. Hennigar, 1981. Effects of prolonged exposure to dietary DDT and PCB on rat liver morphology. Arch. Environ. Contam. Toxicol., 10: 171-183.

Kannan, K., S. Tanabe, A. Borrell, A. Aguilar, S. Focardi and R. Tatsukawa, 1993. Isomer-specific analysis and toxic eval- uation of polychlorinated biphenyls in striped dolphins affected by an epizootic in the western Mediterranean Sea. Arch. Environ. Contam. Toxicol., 25(2): 227-233.

Kawai, S., M. Fukushima, N. Miyazaki and R. Tatsukawa, 1988. Relationship between lipid composition and organochlorine levels in the tissues of striped dolphin. Mar. Pollut. Bull., 19(3): 129-133.

Kendall, M.D., B. Safieh, J. Harwood and P.P. Pomeroy, 1992. Plasma thymulin concentrations, the thymus and organochlorine contaminant levels in seals infected with phocine distemper virus. Sci. Total Environ., 115: 133-144.

Kennedy, S., J.A. Smyth, S.J. McCullough, G.M. Allan, F. McNeilly and S. McQuaid, 1988a. Confirmation of cause of recent seal deaths. Nature, 335: 404.

Kennedy, S., J.A. Smyth, P.F. Cush, S.J. McCuUough and G M. Allan, 1988b. Viral distemper now found in porpoises. Nature, 336: 21.

A. Aguilar, A. Borrell / Sci. Total Environ. 154 (1994) 237-247 247

Koller, L.D. and J.E. Thigpen, 1973. Reduction of antibody to pseudorabies virus in polychlorinated biphenyl-exposed rabbits. Am. J. Vet. Res., 34: 1605-1606.

Kuehl, D.W., R. Haebler and C. Potter, 1991. Chemical residues in dolphins from the U.S. Atlantic coast including Atlantic bottlenose obtained during the 1987/1988 mass mortality. Chemosphere, 22(11): 1017-1084.

Kunita, N., S. Hori, H. Obana, T. Otake, H. Nishimura, T. Kashimoto and N. Ikegami, 1985. Biological effects of PCBs, PCQs and PCDFs present in the oil causing Yusho and Yu-Cheng. Environ. Health Perspect., 59: 79-84.

Lockyer, C.H., 1991. Body composition of the sperm whale, Physeter catodon, with special reference to the possible functions of fat depots. Rit Fiskideildar, 12(2): 1-24.

Loganathan, B.G., S. Tanabe, H. Tanaka, S. Watanabe, N. Miyazaki, M. Amano and R. Tatsukawa, 1990. Comparison of organochiorine residue levels in the striped dolphin from western North Pacific, 1978-1979 and 1986. Mar. Pollut. Bull., 21(9): 435-439.

Loose, L.D., K.A. Pitman, K.F. Bentitz and J.B. Silkworth, 1977. Polychlorinated biphenyl and hexachlorobenzene in- duced humoral immunosuppression. J. Reticuloendo. Soc., 22: 253-271.

Morris, R.J., R.J. Law, C.R. Allchin, C.A. Kelly and C.F. Fileman, 1989. Metals and organochlorines in dolphins and porpoises of Cardigan Bay, West Wales. Mar. Pollut. Bull., 20(10): 512-523.

Murphy, P.G., 1972. Sulphuric acid for the clean-up of animal tissues for analysis of acid-stable hydrocarbon residues. J. Am. Org. Anal. Chem., 55: 1360-1362.

Neubert, R., U. Jacob-Miiller, R. Stahlmann, H. Helge and D. Neubert, 1990. Polyhalogenated dibenzo-p-dioxins and dibenzofurans and the immune system. 1. Effects on pe- ripheral lymphocyte subpopulations of a non-human pri- mate (Callithrix jacchus) after treatment with 2,3,7,8-tetra- chlorodibenzo-p-dioxin (TCDD). Arch. Toxicol., 64: 345-359.

Neubert, R., U. Jacob-Miiller, H. Helge, R. Stahlmann and D. Neubert, 1991. Polyhalogenated dibenzo-p-dioxins and dibenzofurans and the immune system. 2. In vitro effects of 2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD) on lympho- cytes of venous blood from man and a non-human primate (Callithrixjacchus). Arch. Toxicoi., 65: 213-219.

O'Shea, T.J., R.L. Brownell Jr., D.R. Clark, W.A. Walker, M.L. Gay and T.G. Lamont, 1980. Organochlorine pollu- tants in small cetaceans from the Pacific and South At- lantic oceans, November 1968-June 1976. Pestic. Monit. J., 14(2): 35-46.

Osterhaus, A.D.M.E. and E.J. Vedder, 1988. Identification of virus causing recent seal deaths. Nature, 335: 20.

Pulcini, M., R. Carlini and M. Wurtz, 1992. Stomach contents of striped dolphins, Stenella coeruleoalba, (Meyen, 1933) from the south-central Tyrrhenian coast. Eur. Res. Cetaceans, 6: 194-195.

Raga, J.A., J. Aznar, J.A. Balbuena and M. Fern~indez, 1992. Parasites and epizoits in striped dolphins affected by an epizootic in the western Mediterranean. In: X. Pastor and

M. Simmonds (Eds.), Proceedings of the Mediterranean Striped Dolphin Mortality International Workshop, 4-5 Nov. 1991, Palma de Mallorca. Greenpeace Mediterranean Sea Project, pp. 39-46.

Safe, S., 1984. Polychlorinated biphenyls (PCBs) and poly- brominated biphenyls (PBBs): biochemistry, toxicology and mechanism of action. CRC Crit. Rev. Toxicol., 13: 319-393.

Silkworth, J.B. and E.M. Grabstein, 1982. Polychlorinated biphenyi immunotoxicity: dependence on isomer planarity and the Ah gene complex. Toxicol. Appl. Pharmacol., 65: 109-115.

Simmonds, M. and P.A. Johnston, 1989. Seals, sense and science. Mar. Pollut. Bull., 20(11): 580-583.

Sipes, I.G. and R.G. Schnellmann, 1987. Biotransformation of PCBs: metabolic pathways and mechanisms. Environ. Toxin Ser., 1: 97-110.

Sprent, P., 1989. Applied Nonparametric Statistical Methods. Chapman and Hall, London, 259 pp.

SPSS Inc., 1983. SPSS-X User's Guide. McGraw-Hill, New York, 806 pp.

Tanabe, S., R. Tatsukawa, H. Tanaka, K. Marnyama, N. Miyazaki and T. Fujiyama, 1981. Distribution and total burdens of chlorinated hydrocarbons in bodies of striped dolphins (Stenella coeruleoalba). Agric. Biol. Chem., 45(11): 2569-2578.

Tanabe, S., H. Tanaka and R. Tatsukawa, 1984. Poly- chlorobiphenyls, tDDT, and hexachlorocyclohexane isomers in the western North Pacific ecosystem. Arch. Environ. Contam. Toxicol., 13: 731-738.

Taruski, A.G., C.E. Olney and H.E. Winn, 1975. Chlorinated hydrocarbons in cetaceans. J. Fish. Res. Board Can., 32(11): 2205 -2209.

Thomas, P.T. and R.D. Hinsdill, 1978. Effect of polychlori- nated biphenyls on the immune responses of Rhesus monkeys and mice. Toxicol. Appl. Pharmacol., 44: 41-51.

Tryphonas, L., J. Truelove, Z. Zawidzka, J. Wong, J. Mes, S. Charbonneau, D.L. Grant and J.S. Campbell, 1984. Poly- chlorinated biphenyl (PCB) toxicity in adult cynomolgus monkeys (M. fascicularis): a pilot study. Toxicol. Pathol., 12(1): 10-25.

Vicente, N. and D. Chabert, 1978. Recherches de poUuants chimiques dans le tissu graisseux d'un dauphin ~chou~ sur la c6te mediterrane~nne. Oceanol. Acta, 1(3): 331-334.

Vos, J.G. and M.I. Luster, 1989. Immune alterations. In: R.D. Kimbrough and A.A. Jensen (Eds.), Halogenated Biphenyls, Terphenyls, Naphtalenes, Dibenzodioxins and Related Products. Elsevier, Amsterdam, pp. 295-322.

Ward, J.M., 1985. Proliferative lesions of the glandular sto- mach and liver in F344 rats fed diets containing Aroclor 1254. Environ. Health Perspect., 60: 89-95.

Wu, J.C., Y.C. Lu, H.Y. Hao, C.C. Pan and R.Y. Lin, 1984. Cell-mediated immunity in patients with polychlorinated biphenyl poisoning. J. Formosa Med. Assoc., 83: 419-429.

Wurtz, M. and D. Marrale, 1991. On the stomach contents of striped dolphins (Stenella coeruleoalba, Meyen 1933) from the Ligurian coast, central Mediterranean Sea. Eur. Res. Cetaceans, 5: 62-64.