The effects of cadmium on the gills of the goldfish Carassius auratus L.: Metal uptake and histochemical changes

Camp. Biochem. Physiol. Vol. 104C, No. 2, pp. 239-247, 1993 Printed in Great Britain

0306-4492/93 $6.00 + 0.00 0 1993 Pergamon Press Ltd

THE EFFECTS OF CADMIUM ON THE GILLS OF THE GOLDFISH CARASSIUS AURATUS L.: METAL UPTAKE

AND HISTOCHEMICAL CHANGES

P. BATTAGLINI,* G. ANDREOZZI,~ R. ANTONUCCI,~ N. ARCAMONE,~. P. DE GIROLAMO,~. L. FERRARA~ and G. GARGIULO~

*UniversitB degli Studi di Napoli Federico 11, Dipartimento di Zoologia Via Mezzocannone, 8, I-80134 Napoli, Italy (Tel. (081) 552-7069; Fax (081) 552-6452); SUniversiti degli Studi di Napoli Federico 11, Dipartimento di Chimica-Via Mezzocannone, 4, I-80134 Napoli, Italy and tuniversita degli Studi di Napoli Federico 11, Dipartimento di Strutture, Funzioni e Tecnologie Biologiche-Via F. Delpino, 1,

I-80137 Napoli, Italy (Tel. (081) 446317; Fax: (081) 440120)

(Received 27 July 1992; accepted for publication 11 September 1992)

Abstract-l. Uptake of cadmium and histochemical changes of mucopolysaccharides and of cytochrome- oxidase in the gill epithelium of Curussius nuratus were examined at different times after 10 ppm Cd2+ exposure in hard water (520 mg/l CaCO,).

2. All animals survived treatment and no significant behavioural changes were observed. 3. CdZ+ precipitates completely as CdCO, in the first 72 hr. Nevertheless, metal uptake in the organ

was observed beyond this period. 4. After 96 hr an increase in the amount of sulphate mucins and a reduction in intensity of the

cytochrome-oxidase reaction were observed. 5. Cadmium toxicity in very hard water is interpreted as being due to CdCO, ingestion which might

cause histochemical alterations in the gills.

INTRODUCTION

During the last few years, cadmium toxic effects on freshwater fish have been examined in several studies (Pascoe and Mattey, 1977; Calamari et al., 1980; Oronsaye and Brafield, 1984; Karlsson-Norrgren et al., 1985; Versteeg and Giesy, 1986; Sprague, 1987; Davalli et al., 1989; Gill et al., 1991). In these animals, cadmium has been clearly found to cause major histo-morphological and functional alterations in the gills which are the uptake organs of toxic substances and which play a role in the osmoregulatory mechan- ism, probably by discharging metal (Matthiessen and Brafield, 1973). Studies on the cadmium storage in different organs of freshwater fishes have confirmed the gills as the primary target organ in the absorption of this metal (Enk and Mathis, 1977; Roberts et al., 1979; Kumada et al., 1980; Karlsson-Norrgren et al., 1985; Brown et al., 1986; Norey et al., 1990).

Various reports also indicate that changes in the gills, induced by various toxic substances, usually take place together with hyperproduction or with considerable secretion of the mucus (Labat et al.,

1974; Pequignot et al., 1975). Furthermore, the sialic acid content in the gills is regarded as a biochemical factor which is extremely responsive to environmen- tal changes in Sulmo gairdneri (Arillo et al., 1979).

Cadmium toxic action mechanisms on freshwater fish are still very much discussed. Karlsson-Norrgren et al. (1985) hypothesise that in rainbow trout (Salmo

gairdneri) and in zebrafish (Bruchydanio rerio) ex- posed to cadmium, both the gill respiratory and the extrarenal functions might be impaired by the thickening of the secondary lamellae or by its surface reduction with a gas exchange alteration. Moreover, gills of green sunfish were severely affected at high Cd’+ concentrations, with excessive mucus secretion and considerable fusing of gill lamellae (Jude, 1973).

Hiltibran (1971) reported that at relatively low levels, cadmium disrupts energy production by limit- ing oxygen uptake in bluegill (Lepomis mucrochirus) liver mitochondria to such an extent that death may result.

Oronsaye and Brafield (1984) described in Gus- terosteus aculeatus an increase in the number of chloride cells at high Cd2+ concentration which is interpreted as the physiological answer to the increase in the ionic regulation at the gills level and the need to discharge the metal absorbed through the same organ. On the other hand, in Anguilla anguilla, chloride cells have been proved to be storage and discharging places of organic substances (Motais and Garcia-Romeu, 1972; Masoni and Garcia-Romeu, 1972).

The literature data on cadmium toxicity are promi- nently morphological and biochemical ones. This research study aims at analysing the histochemical changes in mucopolysaccharides and in a mitho- chondrial enzyme (cytochrome-oxidase) induced by cadmium sub-acute intoxication in the gills of the

239

240 P. BATTAGLINI et al.

freshwater fish Carassius auratus. These data will then be compared to the metal storage in the organ at different times, and to possible behavioural changes.

MATERIALS AND METHODS

Animals

A total of 110 goldfish of both sexes, 6 + 1 cm long and weighing 2.5 + 0.5 g were purchased by CARMAR SaS., Naples. The fish were kept four weeks in adaptation in the dechlorinated 600 1 glass aquaria of the Zoology Department of the University of Naples.

Eight groups each of 10 specimens randomly selected were placed in eight glass aquaria (30 x 30 x 40 cm) containing 25 1 of tap water, pre- viously aerated for 48 hr in order to eliminate chlor- ine and in which cadmium chloride CdCl, .2fH, 0 had been added at the concentration of 20 mgjl (equal to about 10 mg/l of Cd’+ ). The amount of cadmium we used in the experiments was based on the LC~~ value at 96 hr which in the case of non-salmonid fish is around 10-12 mg/l of Cd*+ (Abel and Papoutsoglou, 1987).

Three control groups of 10 specimens were contes- tually transferred in glass aquaria similar to the previous ones containing the same volume of tap water. The number of fish per tank was determined by taking into account the ratio between the fish body mass and water mass equal to 1 g/l.

After 48 hr from the beginning of the experiment, air pumps (Eheim, Germany) were applied to the aquaria and a commercial flake food (Tetramin, Tetraverke, Melle, Germany) was administered. Tem- perature was constantly 15 + 1°C and photo-period was 12 hr.

Behavioural data

Survival, position in the tank, and motor activity, were recorded every 15 min during the first 4 hr, then every 30 min for the next 12 hr, every 12 hr for the next 4 days, and every 24 hr until the end of the experiment.

To determine position in the tank, aquaria were arbitrarily divided into three zones: bottom, central and top. Motor activity was computed on the number of movements per min of the operculum and the caudal and pectoral fins.

Chemical tests

During the experiment, tap water features were determined by using the classic volume titration to measure hardness, akalinity, calcium, and mag- nesium; a pH-meter ORMA model NK 300 was used to determine the pH in the water and a dissolved oxygen-meter HI 8543 (Hanna Instruments, U.S.A.) was employed to define the amount of diluted oxy- gen. The cadmium ion permanence was followed by measuring its concentration through atomic absorp- tion spectroscopy of water samples taken from the

aquaria following the same schedule of the specimen collections.

Determination of cadmium uptake

For each group of experimental animals three gills were used to determine the cadmium uptake. Gills were weighed and mineralised by the action of a HNO, /HClO, mixture (ratio of concentrated acids 4: 1) following K. Bull’s methods (1975). The result- ing solution was brought to a volume of 10ml in a calibrated flask and analysed by an atomic absorp- tion spectrometer VARIAN AA-275 to determine the amount of cadmium.

Histology and histochemistry

Groups of three controls and five experimental animals were decapitated caudally to the opercula. Some gill arches immediately removed were fixed in Bouin’s fluid, embedded in paraffin and cut into 7 pm sections for the study of mucopolysaccharides. Some branchial arches were directly used for enzymatic reaction test. The following histochemical tests were carried out:

Periodic-acid Schzy reaction. PAS was used in order to identify neutral mucopolysaccharides.

Alcian blue procedure. At pH 2.5 (A.B. pH 2.5) and at pH 1.0 (A.B. pH 1.0) the Alcian blue procedure was used alone or in combination with other pro- cedures for the finding and subsequent localisation of acid mucopolysaccharides.

Aldehyde-fuchsinprocedure. A.F. was used alone or in combination with A.B. pH 2.5 was employed to localise the sulphated mucins.

Enzyme digestion test. Sections immediately follow- ing the ones stained with A.B. pH 2.5-PAS and A.B. pH 2.5 were digested with neuroaminidase (from Clostridium perfrigens type V, Sigma) at a concen- tration of 0.2% in phosphate buffer for 24 hr at 37°C and then stained respectively with A.B. pH 2.5-PAS and A.B. pH 2.5 in order to compare. The main features of these procedures are reported in Table 1.

Enzyme activity test. Cytochrome-oxidase (C.C.O.) activity was determined by using the method of Nachlas et al. (1958). The reaction was carried out on fresh dissected gill arches. After the reaction the gills were fixed in Baker’s formol-calcium and embedded in paraffin. At the same time the specificity of the reaction was determined by using in parallel an incubating medium without specific substrate.

To carry out the histochemical study and to deter- mine the Cd*+ uptake, samples were sacrificed after 3, 24, 48, 72 and 96 hr, and 7, 14 and 40 days from the beginning of the experiment. The same schedule was followed for the control animals.

RESULTS

Behavioural data

Survival was 100% for the whole experiment. Carassius auratus were exposed to 10ppm Cd*+,

Cadmium effect on goldfish gills 241

Methods

Table 1. Hi&chemical procedures for mucosubstances

References Results

Periodic-acid Schiff 1% Alcian blue at pH 2.5 1% Alcian blue at pH I (in 0. I N HCI) Alcian blue at pH 2.5-PAS

Alcian blue at pH I-PAS Aldehyde fuchsin

Aldehyde fuchsin-Alcian blue at pH 2.5 Neuraminidase treatment (24 hr at 37°C)

Pearse Pearse Pearse

Mowry and Winkler

Spicer and Duvenci Spicer and Meyer

Spicer and Meyer

Spicer and Duvenci

1968 I968 1968

1956

Neutral mucins = red Acid mucins = blue Sulphated mucins = blue

Acid mu&s = blue Neutral mucins = red Both = purple

1964 1960

Acid mucins non sulphated = red Sulphated mucins = purple Non-sulphated acid mucins = uncoloured

1960 Sulphated mucins = purple or blue purple

I964 Inhibits alcianophilia of sialic acid containing glycoproteins

freely and randomly moved for the whole experiment as in control animals.

Motor activity, monitored as beats per minute of the opercula and of the caudal and pectoral fins, did not show significant changes during the whole exper- iment. Values compared to controls decreased only during the first hours by about 10%. A frequent ingestion of CdCO, sediments was observed.

Chemical data

Table 2 reports chemical and physico-chemical parameters of the water used for the experiment. It is also possible to notice the very high value of water hardness. In Fig. 1 we report the time course of the cadmium ion concentration in the solution. The results show that already after a few hours, cadmium ion is subject to a strong precipitation as carbonate and decreases to values close to 0 at 72 hr. This situation remains the same until the end of the experiment.

Cadmium uptake

Data on cadmium uptake in the gills are reported in Table 3 and are correlated to the time at which the specimens were sacrificed. Results indicate much higher Cd’+ values during the first 48 hr of exposure to the toxic substance, while they are drastically reduced after this period. In the first part of the experiment cadmium in the gills progressively de- creases to a value of 0 ppm at 72 hr, while in the samples collected in the following periods a moderate storage of Cd2+ IS detected, even though at substan- tially lower values if compared to the first period.

Changes in gill structure

The gills of Carassius auratus are structurally simi- lar to those of other teleosts (Newstead, 1967; Lau-

Table 2. Chemical and chemo-physical features of water

Hardness (mg/l CaCO,) 520 Alcalinity (mg/l HCO;) 570 Conductivity (@/cm at 20°C) 840 Chloride (mg/l Cl-) 32 Sulphate (mg/l SO:-) 25 PH 7.5 Dissolved oxygen (mg/l 0,) 77 Temperature CC) 15+ I

rent and Dunel, 1980). The pectinated gill filaments are attached to the bony gill arches and from there the respiratory lamellae (secondary lamellae) emerge perpendicularly to both sides (Kikuchi, 1977).

The gill filaments, crossed by a cartilaginous axis, are covered, in goldfish, by a stratified epithelium (primary or germinative epithelium), which in its uppermost layer is mainly made up of pavement, mucous and chloride cells. Non-differentiated epi- thelial cells also occupy the epithelium (Ishihara and Mugiya, 1987).

Mucous cells, different in size and rich in granules, are sparsely located in the primary epithelium, some at the base and some on the surface.

Chloride cells are located alone or in groups in the inter-lamellar epithelium. Some of the cells are cov- ered by a layer of pavement cells while others have the apical membrane directly exposed to the external environment (Ishihara and Mugiya, 1987). The shape of goldfish chloride cells varies according to their location in the gill filament epithelium: cuboidal in the inter-lamellar areas, and flattened in the respirat- ory lamellae (Kikuchi, 1977). Secondary lamellae are crossed by the pillar capillary and covered by the secondary epithelium formed by respiratory cells.

In the specimens exposed to Cd2+ action a dilata- tion in the pillar capillaries and a detachment of the secondary epithelium from the capillary are observed

0 ?A 4s 72 8m

Time/hours

Fig. 1. Time course of Cd2+ concentration in ppm found in solution in the experimental water.

242 P. BATTAGLINI et al.

Table 3. Cd2+ concentrations in ppm of fresh organ, found in gills of the Carmsius ~~uratus at different

collection times

Time

3 hr 1 day 2 days 3 days 4 days 1 days

14 days 40 days

Cadmium content in the gills @pm)

361 158 I04

0 6 I

I5 23

already after 3 hr. In the following hours the primary epithelium is hyperplasic and sometimes surrounded by the secondary lamellae which seem to be merged.

After 72 hr a strong reduction in the primary epithelium thickness is detected and the respiratory epithelium is entirely detached, sometimes open and floating in some places. The pillar capillaries look like rosary beads.

After 7 and 14 days the primary epithelium, which surrounds the turgid and taut pillar capillaries, is hypertrophic and hyperplasic. The secondary epi- thelium is found to be strongly thickened and fringed with adherent material.

At 40 days, hyperplasia of the primary epithelium is observed and capillaries are found in similar conditions to the controls.

Mucosubstances

In control animals the large ball-like or elongated cells displaced at the base of the primary epithelium are found to be strongly coloured with A.B. pH 2.5 procedure (Fig. 2). Some cells of average size A.B. pH 2.5 positive are placed in the apical part of the primary epithelium, at the base of the secondary lamellae. Small cells, also positive, are sometimes located along the secondary epithelium. The same distribution appears with the PAS reaction which highlights the neutral mucins. Moreover, the A.B. pH 2.5-PAS reaction showed that PAS-positive and A.B. pH 2.5positive mucins usually coexist in the same cells. Alcian- and PAS-positive mucus is also detected in the gill lamellae along the external border, often packed against the wall of the secondary epithelium.

After digestion with neuroaminidase and staining with A.B. pH 2.5, a clear reduction of the positive cells is observed at the base of the primary epithelium (Fig. 3). A small amount of cells remains with poorly stainable scattered granules. On the other hand, cells located at the top of the primary epithelium and on the secondary lamellae result to be still strongly A.B. pH 2.5-positive. Even after staining with A.F. a very small number of weakly positive cells is found in the basic part of the primary epithelium, while cells found on the surface are positive. These cells are A.B. pH l.O-positive. It is possible to hypothesise that in Carassius auratus the A.B. pH 2.5-positive mucins of the cells at the bottom of the primary epithelium

consist mainly of sialomucins (Spicer and Duvenci, 1964).

After 3 hr from the beginning of the exposure, the epithelium is completely devoid of both PAS- and Alcian-positive mucous cells. Only the small cells are still positive at the apex of the epithelium and at the bottom of the secondary lamellae.

After 24 hr PAS and A.B. pH 2.5 strongly positive cells are found in great number mostly at the bottom of the primary epithelium (Fig. 4). These cells nor- mally found in great amounts in inter-lamellar areas with strongly clustered granules, are almost com- pletely negative after digestion with neuroaminidase and staining with A.B. pH 2.5. The same results are found with A.F.

In the following hours another massive void of the mucous cells takes place reaching its peak after 72 hr, when lamellae are found totally empty except for some small cells at the apex of the primary epi- thelium, which is now very thin. The pillar capillaries look like rosary beads (Fig. 5).

After 96 hr, a strong increase in the Alcian-blue and PAS positiveness is found in the mucous cells. After 14 days the number of A.B. pH 2.5-positive cells is clearly higher than in the controls, especially at the bottom of the primary epithelium (Fig. 6). In many inter-lamellar spaces there are many different cells at different levels, while cells located at the bottom of the secondary lamellae seem to be few. A certain amount of A.B. pH 2.5- and PAS-positive thick and granulous material is detected on the secondary epithelium and among the lamellae. In the specimens obtained after neuroaminidase digestion a moderate number of cells presents an alcianophilia resistant to neuroaminidase in the apical part but mainly at the base of the primary epithelium (Fig. 7). It should be noted that compared to the controls, they have a high amount of sulphate mucins. These data were confirmed by the great number of A.F.- positive cells detected.

After 40 days cells placed at the bottom of the primary epithelium are less and smaller than those described above (Fig. 8). Reactivity is similar to controls; in fact, after digestion with neuroaminidase and A.B. pH 2.5 basal cells are barely positive or completely negative (Fig. 9).

Cytochrome-oxidase reaction

In controls the C.C.O. reaction is highly positive and appears as brown grains at the level of numerous rounded cells mostly located in the apical part of the primary epithelium which very often reach the apex of the pillar capillary (Fig. 10). In an inter-lamellar area a great number of positive cells can be detected. They resemble chloride cells in shape and distri- bution, as shown by Petrik and Bucher (1969), Garcia-Romeu and Masoni (1970) and Kikuchi (1977). In the first hours of exposure the reaction intensity dramatically decreases and cells appear vac- uolised. This reduction reaches its peak after 48 hr

(Fig. 11). After 96 hr many cells appear to be again at different levels in the primary epithelium. Further- strongly positive and with sparsely gathered grains more, much granulous material is gathered on the (Fig. 12). A few vacuolised cells may also be found secondary lamellae which are positive in some areas.

Cadmium effect on goldfish gills 243

Fig. 2. Gills of Carassius auratus, control animals, stained with Alcian blue at pH 2.5. Arrows indicate the strongly blue coloured mucous cells at the base of the primary epithelium. x 800.

Fig. 3. Gills of Carassius auralus, control animals, treated with neuroaminidase and post-stained with Alcian blue at pH 2.5. Arrows indicate mucous cells in pale blue very weakly coloured at the base of the

primary epithelium. x 800.

Fig. 4. Gills of Curussius auratus, after 24 hr of exposure to Cd?+ stained with Alcian blue at pH 2.5. Note the great number of strongly coloured cells at the bottom of the primary epithelium. Arrows indicate the

secondary lamellae which surround the hyperplasic primary epithelium. x 640.

Fig. 5. Gills of Carassius aurafus, after 72 hr of exposure to Cd?+ stained with Alcian blue at pH 2.5. Note the decreased number of cells, which are now less coloured. Arrows indicate the pillar capillaries which

look like rosary beads. x 640.

CBP(C) 104/2-D

After 7 days, cells appear still positive but small and packed with strongly massed granules (Fig. 13), while in the following periods there is a further reduction of the positive cells and of reaction intensity.

DISCUSSION

Our results indicate that Carassius aura&s exposed to the toxic action of cadmium (10 mg/l of Cd*+ ) in very hard tap water (520 mg/l of CaCO,) does not die within the 40 days period. The behavioural activity (position in the tank, beats of the opercula and of

caudai and pectoral fins) does not show any signifi- cant changes compared to the controls.

Contrasting data in the literature report that: A 48 hr ~csa of 3.7 mg/l of Cd” in Salmo gairdneri by Calamari ef al. (1980); a 96 hr LC~~ of 23 m&/l of Cd’+ in Gasferosteus acdeafus (Pascoe and Mattey, 1977); a 96 hr LC~,, of 2. I3 mg/I of Cd’+ in Carassius awutus (McCarthy et ai., 1978). On the other hand, Vesteeg and Giesy (1986) discovered in Lepomis macrochirus an average lethal time of 63 days at a Cd2+ concen- tration of 12.7 mg/l, a finding which seems very close to our experimental results.

244 P. BATTAGLINI ei al.

Fig. 6. Gills of Carassius auratus, after 14 days of exposure to Cd I+ stained with Alcian blue at pH 2.5. Arrows indicate strongly coloured mucous cells at the bottom of the primary epithelium. x 800.

Fig. 7. Gills of Carassius auratus, after 14 days of exposure to Cd z+ treated with neuroaminidase and post-stained with Alcian blue at pH 2.5. Arrows indicate coloured mucous cells at the bottom of the

primary epithelium. x 800.

Fig. 8. Gills of Carussius aurattls, after 40 days of exposure to Cd” stained with Aician blue at pH 2.5. Note small coloured mucous cehs at the bottom and on the surface of the primary epithelium. x640.

Fig. 9. Gills of Carussius aura&s, after 40 days of exposure to Cd 2+ treated with neuroaminidase and post-stained with Alcian blue at pH 2.5. Note the lack of coioured cells at the base of the primary

epithelium. The same result is found in the controls. x 640.

Cadmium effect on goldfish gills 245

It is important to note that the cadmium toxic effect on fishes is due to several variables such as:

metal concentration, water chemical parameters, ex- perimental procedures (static or in flow), and of course reactivity to cadmium in the species tested (Sprague, 1987). In this regard the water hardness high value of our experimental conditions should be noted (520mg/l of CaCO,). As a matter of fact, this parameter might be considered one of the main factors affecting the cadmium toxic action on fresh-water fishes (Sprague, 1987; Calamari et al.,

1980). At the concentration of lOmg/l of Cd2+ that we

used (Table 2) the high alkaline value of the water and the slightly basic pH value cause a progressive

precipitation of this ion as cadmium carbonate according to the reaction:

Cd2+(aq) + 2HCO;(aq) + 2H,O

= CdCO,(s) + 2H,O+(aq).

This is important for a proper evaluation of the

results since it points out that fishes are subject to a toxicity level which evolves with time. In fact, it ranges from 10 mg/l of Cd2+ in solution to the complete disappearance already after 72 hr, together with the formation of CdCO, toxic deposits at the bottom. These deposits were ingested by the goldfish because of their habit of eating debris. Moreover, although they have a well developed sense of taste

Fig. 10. Gills of Carassius auratus, control animals. The cytochrome-oxidase reaction results strongly positive in rounded cells located in the apical part of the primary epithelium (arrows). x 640.

Fig. Il. Cytochrome-oxidase reaction in the gill of Curussius am-am after 48 hr of exposure to Cd2+. Note the decrease in the intensity of the reaction compared with controls. x 640.

Fig. 12. Cytochrome-oxidase reaction in the gill of Carassius nuratus after 96 hr of exposure to Cd*+. A sudden recovery of positivity is to be noted. x 800.

Fig. 13. Cytochrome-oxidase reaction in the gill of Curassius auratus after 7 days of exposure to CdZ+. Arrows indicate strongly clustered small positive cells. x 800.

246 P. BATTAGLINI et al.

(Misslin, 1971) which prevents them from ingesting harmful substances, they easily ingest the CdCO, which is tasteless.

However, the histological aspects previously de- scribed, indicate a reaction to toxic substances of the Carassius auratus also under these experimental con- ditions. This has already been reported by Oronsaye and Brafield (1984) in Gasterosteus aculeatus,

Karlsson-Norrgren et al. (1985) in zebrafish and rainbow trout, and by Versteeg and Giesy (1986) in Lepomis macrochirus. In any case, the histological data observed after 40 days reveals the possibility of functional recovery.

The histochemical data underline the initial void of the mucous cells during the first hours of the exper- iment followed by a strong recovery after 24 hr, and a subsequent void with its peak after 72 hr. Mucus secretion as a first response to the toxic action of cadmium (Enk and Mathis, 1977) and of other toxic substances found in the water environment (Labat et

al., 1974; Pequignot et al., 1975), might be the key to understanding the cadmium uptake values in the gill obtained with this experiment. These values, in fact, are found to be very high in the early experimental hours, and then to decrease to zero after 72 hr. In the following hours an increase in metal storage may be detected, although at considerably lower levels. These apparently anomalous results can be understood while considering both the CdC03 precipitation in the first 72 hr and the strong mucus secretion in these exposure periods to Cd’+. Hence, the great amount of cadmium found in the organ during the first 48 hr may be due to the CdC03 which adheres to the branchial surface covered with mucus rather than to the metal absorbed by the animal. Only after 72 hr of exposure to the toxic environment, when cadmium is only found as CdC03 at the bottom of the aquaria, can we ascribe cadmium concentration values to a real storage of this metal in the organ, indicating that in the second part of the experiment cadmium is ingested as carbonate. In similar experimental con- ditions data concerning the branchial storage of Cd’+ after 72 hr from the beginning of the experiment were substantially lower than those in the intestine (Battagiini et al., 1991).

It is interesting to note that in the mucopoly- saccharides turnover at longer intervals (7, 14 days) neuroaminidase-resistant and therefore sulphate acid mucins have been highlighted in contrast with what has been observed both in control animals and in specimens tested after 24 hr.

A change in mucins chemical properties and there- fore in the reactivity to A.B. was reported by Whitelaw (1975) in Carassius auratus after the in- traperitoneal administration of a high concentration of monovalent cations. On the other hand, in Salmo

gairdneri the change in the sialic acid is considered a sign of pollution stress (Arillo et al., 1979).

Histochemical data concerning the display of C.C.O. activity also suggest an initial phase charac-

terised by a massive reduction in the chloride cell positiveness as a reaction to environmental stress (Gargiulo et al., 1992), and a recovery during the following hours with its peak after 96 hr. In the second phase an alteration in the positive cells is detected (7 days) with a subsequent reduction in the number of these cells, and in reaction intensity over a long-term exposure (14,40 days). Our data disagree with the hypothesis of a reaction to cadmium toxicity determined by increase of chloride cells as proposed by Oronsaye and Brafield (1984). Changes similar to those observed for cadmium have been described in Carassius auratus when exposed to high doses of crude oil (Gargiulo et al., 1989).

Under the present experimental conditions, in spite of the total lack of mortality, histological and histo- chemical alterations were able to be observed. These could probably be due to the cadmium toxic action and be related to the respiratory activity of the chloride cells where the main branchial metabolic processes take place. Meanwhile, cadmium storage in the gills has also been detected following the same schedule of the experiment. Therefore, we hypoth- esise that in hard water, although cadmium disap- pears from the solution after the initial experimental hours, it acts as a toxic substance through the inges- tion of CdC03 and determines both morphological changes in the gills and signs of alterations of the respiratory activity.

Acknowledgements-This study was supported by a contri- bution of Italian MURST (40%). We wish to thank Mr M. Sammarco for technical support.

REFERENCES

Abel P. D. and Papoutsoglou S. E. (1987) Lethal toxicity of cadmium to Cyprinus carpio and Tilapia aurea. Bull. Envir. Contam. Toxic. 38, 382-386.

Arillo A., Margiocco C. and Melodia F. (1979) The gill sialic acid content as an index of environmental stress in rainbow trout, Salmo gairdneri, Richardson. J. Fish. Biol. 15, 405410.

Battaglini P., Ferrara L. and Soppelsa 0. (1991) Assorbi- mento di Cd*+ in vari organi di Carassius auratus in acque “dure” e “dolci”. Atti Sot. It. Sci. Vet. 45, (in press).

Brown M. W., Thomas D. G., Shurben D., Solbe J. F. de L. G., Kay J. and Cryer A. (1986) A comparison of the differential accumulation of cadmium in the tissues of three species of freshwater fish, Salmo gairdneri, Rutilus ruMus and Noemacheilus barbatulus. Comp. Biochem. Physiol. 84C, 213-217.

Bull K. (1975) Routine heavy metal analysis of biological samples. Science Chelsea J. Res. Chel. I, 23-35.

Calamari D., Marchetti R. and Vailati G. (1980) Influence of water hardness on cadmium toxicity to Salmo gairdneri Rich. Water Res. 14, 1421-1426.

Davalli P.. Serrazanetti G. P.. Caruent E. and Corti A. (1989) Responses of liver enzymes to cadmium adminis- tration in goldfish (Carassius auratus) at different times of the year. Comp. Biochem. Physiol. !MC, 177-181.

Enk M. D. and Mathis B. J. (1977) Distribution of cadmium and lead in a stream ecosystem. Hydrobiologia 52, (2-3), 153-158.

Cadmium effect on goldfish gills 247

Motais R. and Garcia-Romeu F. (1972) Transport mechan- ism in the teleostean gill and amphibian skin. Ann. Physiol. 34, 141-176.

Mowry R. W. and Winkler C. M. (1956) The coloration of acid carbohydrates of bacteria and fungi in tissue sections with special reference to capsules of Cryprococcus neofor- mans, Pneumococcus and Staphylococcus. Am. J. Pathol. 36, 628629.

Nachlas M. M., Crawford D. T., Goldstein T. P. and Seligman A. M. (1958) The histochemical demonstration of cytochrome-oxidase with a new reagent for the Nadi reaction. J. Histochem. Cytochem. 6, 445456.

Newstead .I. D. (1967) Fine structure of the respiratory

Garcia-Romeu F. and Masoni A. (1970) Sur la mise en evidence des cellules a chlorure de la branchie des pois- sons. Arch. Anat. Micr. Morph. Exp. 59, 289-294.

Gargiulo G., Andreozzi G., Antonucci R., Arcamone N., de Girolamo P., Battaglini P. (1989) I pesci come modello sperimentale nello studio degli inquinamenti letali e sub- letali di ccosistemi acquatici. Inter. Symp. Environ. Poll. Animal Pop., Pisa, 93-100.

Gargiulo G.,Andreozzi G., Antonucci R., Arcamone N., de Girolamo P. and Battaglini P. (1992) Behavioural and histochemical changes induced by osmotic stress in Poe- cilia reticulata Peters (Pisces, Cyprinodontidae). Comp. Biochem. Physiol. lOlA, 3342.

Gill T. S., Tewari H. and Pande J. (1991) In vivo and in vitro effects of cadmium on selected enzymes in different organs of the fish Barbus conchonius Ha. (rosy barb). Comp. Biochem. Physiol. lOOC, 501-505.

Hiltibran R. C. (1971) Effects of cadmium. zinc. manaanese and calcium on oxygen and phosphate metabolt<m of blue-gill liver mitocondria. J. Wat. Pollut. Control Fed. 43, 818-823.

Ishihara A. and Mugiya Y. (I 987) Ultrastructural evidence of calcium uptake by chloride cells in gills of goldfish, Carassius auratus. J. exp. Zool. 242, 121-129.

Jude D. J. (1973) Sublethal effects of ammonia and cad- mium on growth of green sunfish. Unpublished Ph.D. Thesis, Michigan State Univ., 193 pp. Reported in: Enk M. D. and Mathis B. J. (1977).

Karlsson-Norrgren L., Runn P., Haux C. and Forlin L. (1985) Cadmium-induced changes in gill morphology of zebrafish, Brachydanio rerio (Hamilton-Buchanan), and rainbow trout, Salmo gairdneri Richardson. J. Fish. Biol. 27, 81-95.

Kikuchi S. (1977) Mitochondria-rich (chloride) cells in the gill epithelia from four species of stenohaline fresh water teleosts. Cell Tiss. Res. 180, 87-98.

Kumada H., Kimura S. and Yokote M. (1980) Accumu- lation and biological effects of cadmium in Rainbow trout. Bull Jpn Sot. Sci. Fish 46, 977103.

Labat R., Penquignot J. and Chatelet A. (1974) Action toxique du cuivre sur les branchies de carpe (Cyprinus carpio). Annls Limnol. 10, 109-114.

Laurent P. and Dune1 S. (1980) Morphology of gill epithelia in fish. Am. J. Physiol. 238, R147-R159.

Masoni A. and Garcia-Romeu F. (1972) Accumulation et excretion de substances organiques par les cellules a clorure de la branchie d’Anguilla anguilla L. adaptte a l’eau de mer. Z. Zellforsh. 133, 389-398.

Matthiessen P. and Brafield A. E. (1973) The effects of dissolved zinc on the gills of the stickleback Gasterosteus aculeatus L. J. Fish. Biol. 5, 607613.

McCarthy L. S., Henry J. A. C. and Houston A. H. (1978) Toxicity of cadmium to goldfish, Carassius auratus, in hard and soft water. J. Fish Res. Board Can. 35, 3542.

Misslin R. (1971) Effets de l’iproniazide et de la reserpine, seules ou en association, sur une reaction conditionee d’evitement chez Carassius auratus. C. R. Lab. Sceanc. Acad. Sci. Paris 275, 974977.

lamellae of teleostean gills. Z. Zellforsh. 79, 398-428. _ Norev C. G.. Crver A. and Kav J. (1990) Cadmium uotake

and sequestraiion in the pike (Esox lucius). Comp. Bio- them. Physiol. 95C, 217-221.

Oronsaye J. A. 0. and Brafield A. E. (1984) The effect of dissolved cadmium on the chloride ceils of the gills of the stickleback, Gasterosteus aculeatus L. J. Fish. Biol. 25, 253-258.

Pascoe D. and Mattey D. L. (1977) Studies on the toxicity of cadmium to the three-spined stickleback Gasterosteus aculeatus L. J. Fish. Biol. 11, 207-215.

Pearse A. G. E. (1968) Histochemistry, Theoretical and Applied, 3rd edn, Vol. I. Churchill, London.

Pequignot J., Labat R. and Chatelet A. (1975) Action du sulfate de cuivre sur les cellules a mucus de l’alevin de truit (Salmo irideus). J. Europ. Toxicol. 8, 52-56.

Petrik P. and Bucher 0. (1969) A propos des ‘Chloride cells” dans l’epithelium des lamelles branchiales du poissons rouge. Z. ZeNforsch. 96, 6674.

Roberts K. S., Cryer A., Kay J., Solbe J. F. de L. G., Wharfe J. R. and Simpson W. R. (1979) The effects of exposure to sub-lethal concentration of cadmium on enzyme activi- ties and accumulation of the metal in tissue and organs of rainbow and brown trout (Salmo gairdneri Richardson and Salmo trutta Mario L.) Comv. Biochem Phvsiol. 62C. 135-140. ” ’ .

Spicer S. S. and Duvenci J. (1964) Histochemical character- istics of mucopolysaccharides in salivary and exorbital lacrimal glands. Anat. Rec. 149, 333-358.

Spicer S. S. and Meyer D. B. (1960) Histochemical differen- tiation of acid of mucopolysaccharides by means of combined aldehyde Fuchsin-Alcian Blue staining. Am. J. clin. Path. 33, 453460.

Sprague J. B. (1987) Effect of cadmium on freshwater fish. In: Cadmium in the Aquatic Environment (Edited by Nriagu J. 0. and Sprague J. B.), pp. 1399169. Wiley, New York.

Versteeg D. J. and Giesy J. P. (1986) The histological and biochemical effects of cadmium exposure in the Bluegill sunfish (Lepomis macrochirus). Ecotoxicol. Environ. Saf. 11, 3143.

Whitelaw D. A. (1975) Effect of monovalent cations on the nature of mucosubstance secreted by the gill epi- thelium in the goldfish Carassius auratus. S. Afr. J. Sci. 71, 25527.