Ecotoxicology and Environmental Safety 47, 306}313 (2000)

Environmental Research, Section B

doi:10.1006/eesa.2000.1990, available online at http://www.idealibrary.com on

Uptake of Cadmium Adsorbed on Particulates by Gills of Goldfish(Carassius auratus)

Shu Tao,1 Changfa Liu, R. Dawson, Aiming Long, and Fuliu XuDepartment of Urban and Environmental Sciences, Peking University, Beijing 100871

Received March 6, 2000

Gold5sh (Carassius auratus) were exposed to mixtures ofconstant dissolved cadmium (0.01 mg/L) and cadmium adsorbedon gibbsite particles at concentrations of 0, 0.025, 0.050, 0.075,and 0.100 mg/L. The gills of the 5sh were excised after a 5-dayexposure experiment and both cadmium and aluminum in thegills were measured. The gills were also examined with a lightmicroscope for surface adherence of the particles after the expo-sure. The evidence collected demonstrated that the concentrationof cadmium in the gills increased with increased concentration ofparticulate cadmium during the exposure while the dissolvedcadmium remained constant. The ratio of cadmium to aluminumin the gills was apparently higher than that in the solution,indicating the stripping and translocation of cadmium from theparticles. A multistep uptake process is proposed and thebioavailability of particulate cadmium to 5sh gills is discussed.( 2000 Academic Press

Key Words: 5sh; gills; cadmium; particulate metal; speci-ation; bioavailability.

INTRODUCTION

Much of the recent literature in environmental bio-geochemistry indicates that the speciation of trace metals inan aquatic system is more meaningful than its total expo-sure concentration. Chemical speciation in an aquatic sys-tem is of central importance to bioavailability. Although theconcepts of bioavailability and speciation have been widelyaccepted, the knowledge of the processes governing thebehavior of various metal species to aquatic organisms islimited and is critical to a better understanding of metalbioavailability and toxicity.

Fish are often exposed to naturally suspended solids ofvarious concentrations, especially during heavy rainfall orspring freshets. Owing to direct contact with suspendedsolids, "sh gills are most a!ected by the particles. The directe!ect of suspended solids on "sh over a wide range of

1To whom correspondence should be addressed.

306

0147-6513/00 $35.00Copyright ( 2000 by Academic PressAll rights of reproduction in any form reserved.

concentrations has been the subject of numerous investiga-tions (Martens and Servizi, 1993). As one of the key com-partments of "sh, gills are also the "rst and the most impor-tant target of waterborne toxicants (Roy and Campbell,1995).

Considering the bioavailability of trace metals to "sh, it isgenerally recognized that the hydrate ions and hydroxocomplexes are the most bioavailable forms of metals for "shuptake via the gills among all of the species that exist in theaquatic environment (Erickson et al., 1994). Other speciesidenti"ed as bioavailable forms are hydroxo and other inor-ganic complexes (Simkiss, 1983; Leland and Kuwabara,1985).

Adsorption on suspended solids is one of the major pro-cesses controlling the metal speciation in the aquatic system.Aquatic solids are composed of a mixture of material fromdi!erent sources. The "nely divided suspended material hasan extremely large surface area. As a result, these particleshave a large capacity for physical adsorption and absorp-tion of ionic species and serve as an important sink formetals in natural water systems. The adsorption of metalsonto suspended particles, in many cases, is the main cause ofthe low content of dissolved metal species in water systems(Allen, 1996).

The bioavailability of metals in suspended materials hasreceived less attention than deserved. Most studies dealingwith particulate metal availability have thus far focused on"lter-feeding organisms or uptake through ingestion(Benson et al., 1994). As indicated by Playle (1998) the gillmodeling approach should be expanded to include metalsbound to particulate matter, for example, in turbid riverinesystems. It is unwise to dismiss particulate-bound metals ascompletely unavailable to aquatic organisms. In the de-velopment of a new test system for exposing "sh to resusp-ended sediments, Cope and colleagues (1996) found that1.8% of the cadmium associated with the sediment wasdetectable in juvenile bluegills (¸epomis macrochirus) afterexposure for 28 days. The results of both light microscopicand electron microscopic examination demonstrated that athigh concentrations, suspended particles in ambient water

UPTAKE OF Cd ADSORBED ON PARTICULATES BY GOLDFISH 307

could be phagocytosed by the epithelial cells. It wasspeculated that phagocytic uptake of contaminated par-ticles may provide an entry pathway for toxic substances(Martens and Servizi, 1993).

Recently, Tao and colleagues evaluated the bioavailabil-ity of copper to "sh by exposing neon tetras (Paracheirodoninnesi) to various species of copper. It was observed thatalthough kaolin clay could reduce "sh uptake of copper, theresults of stoichiometric analysis indicated that a fraction ofthe kaolin clay-adsorbed copper might be available for "shuptake via gills (Tao et al., 1999a, Liang et al., 1999). Similarphenomena on particulate lead uptake by gold"sh (Caras-sius auratus) was reported (Tao et al., 1999b).

Fish gills are physiologically complex and are also animportant target of toxic metal in surrounding water (Reidand McDonald, 1991). In a 250-g adult rainbow trout,for example, the gills make up 60% of the total externalsurface area and are exposed, by ventilation, to more than48 L of water per hour (Reid, 1989). The water layer near thesurface of gills is a very important microenvironment thatmay be markedly di!erent from the surrounding water. ThepH of the gill microenvironment could be altered by move-ment of NH

3, NH`

4, CO

2, and HCO~

3, across the gill

surface (Newman and Jagoe, 1994; Wright and Wood,1985). Metal speciation as well as mobility and bioavailabil-ity in the bulk solution could be signi"cantly modi"edwithin this microenvironment (Playle and Wood, 1989). Theresults of a series of chemical equilibrium calculations usingMINTEQA2 for "sh gills demonstrated a signi"cant changein metal speciation with respect to bioavailability (Wenet al., 2000).

Another important feature of the "sh gill microenviron-ment is the existence of mucus which may play an importantrole in metal uptake. Gill epithelia are usually covered withan extracellular polyanionic matrix functioning as an ionexchange system (Part and Lock, 1983). With varied a$n-ities for various metals, mucus likely concentrates metalswithin the gill microenvironment (Kirchner, 1987). It mayalso serve to slough o! potentially toxic ions (Varanasi andMurkey, 1978).

The objective of this study was to investigate possibleuptake of particulate cadmium by "sh gills and to explorethe mechanism of such an uptake.

EXPERIMENTAL

Water and Fish

The exposure experiments were conducted in recon-stituted water synthesized by adding reagent-grade saltsto deionized water. The ionic composition of the majorrivers of China was adopted for the synthetic waterused in this study. The concentrations of major cationsand anions in the water were [Na`] 31 mg/L, [Ca2`]65 mg/L, [K`] 14 mg/L, [Mg2`] 16 mg/L, [Cl~] 115 mg/L,

[CO2~3

] 103 mg/L, [NO~3] 5 mg/L, and [SO2~

4] 58 mg/L.

The pH and alkalinity of the water were measured as8.9 and 0.359 mmol/L, respectively, after the exposureexperiment.

Gold"sh (Carassius auratus) with an average weight of9.8 g (range, 6.9}12.7 g) were obtained from the Tongzhou"sh market in Beijing. On arrival in the laboratory, the"shes were placed in tap water for storage during which thewater was replaced every 5}6 days. The "sh were fed everyday until 5 days before the exposure experiment. The "shwere acclimated in the synthetic water for 24 h immediatelybefore the exposure experiment.

Reagents

Commercial gibbsite (Fangshan Kaolin factory, Beijing)was ground in an agate mortar. Particles less than 34 lm(400 mesh) in size were collected by sieving. A suspension ofthe gibbsite was prepared for the experiment.

The reagents used in the experiment were of analyticalgrade or better and deionized water was used throughoutfor metal determination. All glassware was soaked in 5%nitric acid (v/v) for 24 hours and rinsed with de-ionizedwater before use.

Characterization of the Adsorption of Cadmium on Gibbsite

Before the design of the exposure experiment, the gibbsiteparticles were characterized by establishing a cadmium ad-sorption isotherm. Fifty milliliters of gibbsite suspension(15 mg/L) was titrated with cadmium and the solution wascontinuously shaken for 12 h. The supernatant was mea-sured for cadmium using a PAR polarograph Model 384 ina di!erential pulse anodic stripping mode in conjunctionwith a Model 303 hanging mercury drop electrode aftercentrifugation. It was found that a Langmuir isotherm can"t the measured results very well:

Cd(p)"7.7211]13.9057]Cd(a)

1#13.9057]Cd(a)(r2"0.989, n"10),

where Cd(p) (mg/g) is the concentration of the particulatecadmium adsorbed on gibbsite and Cd(a) (mg/L) representsdissolved cadmium in the water. The equation was used tocalculate the amounts of gibbsite and cadmium to be addedduring the exposure experiment.

Exposure Experiments

To set up a reference for comparison, the "sh wereexposed to dissolved cadmium at varied concentrationsof 0, 0.05, 0.10, 0.30, and 0.50 mg/L. The accumulation

308 TAO ET AL.

of cadmium in the "sh gills was measured after exposurefor 10 days. The experiment was carried out in glass tanksholding 10 L of synthetic water with four "sh in each tank(treatment). The average weight of the "sh was 9.77$2.72 g.The measured pH of the water was 8.84$0.05. The temper-ature of the water varied within the range 23}253C. Part ofthe gills of each "sh was excited and one composed samplefrom four "sh in each treatment was prepared for analysis.A quantitative relationship between exposure concentrationand accumulation quantity was established, based on whichthe availability of particulate cadmium to "sh gills couldbe evaluated. The results of cadmium accumulation in gillsof "sh exposed to dissolved cadmium at various concentra-tions were compared with those of "sh gills exposed toparticulate cadmium.

The basic idea for examining the possible uptake of cad-mium adsorbed on particles is to conduct an exposureexperiment during which "sh are exposed to adsorbed cad-mium at varied concentrations while the concentration ofdissolved cadmium in the system is maintained at a constantlevel.

According to the Langmuir isotherm derived by titration,the concentrations of total cadmium and gibbsite particleswere calculated to maintain a constant level of dissolvedcadmium and varied concentrations of particulate cadmium(Table 1).

Exposures were started by transferring three "sh to eachglass tank holding 800 mL of synthetic water spiked withcadmium and gibbsite as listed in Table 1. The averageweight of the "sh was 4.85$0.79 g. Nonexposed "sh fromthe holding tank served as the control group (TreatmentNo. 0). Fish were not fed during the exposure experimentsso as to avoid the potential for a signi"cant change in metalspeciation. The solution in the tanks was stirred occa-sionally to keep the particles in suspension during the ex-periment. The solution pH was measured as 7.19$0.16;water temperature was between 19 and 213C. Two duplicatetanks were set up for each treatment. After exposure for5 days, the gill arches were excised from both sides of the "shgills. Samples were weighed and digested before measure-ments for cadmium and aluminum.

TABLE 1Experiment Design for the Particulate Cadmium Exposure

Treatment No. 0 1 2 3 4 5

Concentration of dissolvedcadmium (mg/L) 0 0.010 0.010 0.010 0.010 0.010

Concentration of particulatecadmium (mg/L) 0 0.000 0.025 0.050 0.075 0.100

Total concentration ofcadmium (mg/L) 0 0.010 0.035 0.060 0.085 0.110

Concentration of gibbsite (g/L) 0 0.000 0.025 0.050 0.075 0.100

Microscopy Examination

The gill arches were excised from the "sh exposed tosuspended particles in the solution. The specimens weresquashed on glass slides and covered with coverglass whichwere mounted with gum. The prepared samples were exam-ined and photographically enlarged (400]) using lightmicroscopy to observe the particles adhering to the surfaceof the gills.

Sample Analysis

After drying at 1053C and weighing, the gill tissue(ca. 0.5 g) excised from the "sh was digested using a micro-wave oven (Model MDS-2000). Concentrated nitric acid(5.0 mL) and 30% hydroperoxide (2.0 mL) were used forthe digestion in a stepwise mode. To dissolve gibbsiteparticles, the solution were transferred into a 50-mLbeaker and 3 to 5 drops of concentrated H

2SO

4were added

before digestion on a hot plate at 2003C. The resultingresidue was then totally dissolved in deionized water withno solid residue left. The solution was neutralized beforealuminum and cadmium determination using 1.0 mol/LNaOH.

The cadmium concentration in the samples was measuredusing a PAR polarograph Model 384 in a di!erential pulseanodic stripping mode in conjunction with a Model 303hanging mercury drop electrode. During the polarographicdetermination, the sample solution was bu!ered with1.0 mol/L NaAc}HAc (1.0 mL sample and 9.0 mL bu!ersolution) and was purged for 10 min. A deposition time of120 s, equilibrium time of 30 s, scan range of !0.8 to!0.2V, voltage pulse height, of 50 mV, and scan rate of4 mV/s were adopted for the measurement.

The aluminum content of the gill samples was determinedbased on Dougan and Wilson's (1974) absorptiometric pro-cedure. Twenty milliliters of digested sample was thoroughlymixed with 0.4 mL HCl (5.0 mol/L) before being spiked with1.0 mL of 1,10-phenanthroline (0.1%). Then 1.0 mL of cat-echol violet (0.0375%) and 5.0 mL of 6-hexamine were ad-ded in sequence. The absorbance of the solution at 585 nmwas recorded 15 to 30 min later using a Unicam UV4-100 uv/vis spectrometer.

RESULTS

Uptake of Dissolved Cadmium by Fish Gills

During a 10-day experiment, the "sh were exposed tovarious levels of dissolved cadmium in aqueous solution.The concentrations of dissolved cadmium varied from 0 to3.281 lmol/L in seven treatments including a control. Themeasured contents of cadmium that accumulated in the gillsof the "sh are plotted against total dissolved cadmiumconcentrations in Fig. 1.

FIG. 1. Accumulation of cadmium in the gills of gold"sh exposed tovarious concentrations of dissolved cadmium.

FIG. 2. Cadmium concentrations in the gills of the "sh exposed tovarious concentrations of particulate cadmium (0}0.10 mg/L) and a con-stant concentration of dissolved cadmium at 0.01 mg/L.

UPTAKE OF Cd ADSORBED ON PARTICULATES BY GOLDFISH 309

Without humic substances or suspended solids as organicligands or adsorbents, many species of inorganic complexesexist in aqueous solution. Five inorganic ligands of primaryimportance in natural waters are F~, Cl~, SO2~

4, OH~, and

HCO2~3

. For cationic metals, the &&free'' or hydrated metalion (often six water molecules in the "rst hydration shell) hasbeen suggested to be the most bioavailable form for "shuptake (Morel, 1984). Kuimr studied the speciation pro"leof cadmium and its bioconcentration in an aquatic ecosys-tem. It was found that the major inorganic metallic speciesare present at picomolar concentrations. Cadmium accu-mulated in the teleost and quantity of accumulation corre-lated well with Cd2` in the exposure medium, exemplifyingthat the ionic species of cadmium represent the principalbioavailable forms to "sh (Kuimr et al., 1996). In addition toaquo ions, some inorganic complexes are also bioavailableto varied extents and their bioavailability depends on thelipid solubility of the complexes formed (Erickson et al.,1994; Leland and Kuwabara, 1985). Simkiss (1983) suggeststhat inorganic ligands can facilitate metal penetration ofmembranes by forming relatively uncharged complexes. Forexample, Cd in seawater can enter as chloro complexes.Similarly, Bingham et al. (1984) correlated the e!ect of Cl~on Cd bioavailability to its in#uence on Cd2` activity.

Chemical equilibrium is often required to characterize thedistribution of various metal species in the system. For thisstudy, since the composition, pH, and total dissolved cad-mium concentration of the water remained constant duringthe exposure experiment with or without the suspendedsolids, it was reasonable to assume that the fraction of allbioavailable species was proportional to the concentrationof dissolved cadmium. Dissolved cadmium, instead ofbioavailable cadmium, was applied in this study.

As presented in Fig. 1, for the "sh gills, a general increas-ing trend in cadmium accumulation with increase in dis-solved cadmium concentration in the bulk solution can berecognized. Many researchers suggested that the trend ofmetal uptake by "sh gills can be depicted by a Langmuirisotherm (Playle, 1998). The results of such a "t are provided

in Fig. 1, which illustrates a more or less linear increase inaccumulation at low concentration and a trend of slowingdown at high concentration.

Uptake of Particulate Cadmium by Fish Gills

Due to chemical equilibrium, there is simply no way tomaintain an elevated level of particulate cadmium in thesystem without dissolved free metal. Therefore, the partic-ulate cadmium exposure experiment was designed in sucha way that a constant level of dissolved cadmium wasmaintained at 0.010 mg/L, while the concentrations of par-ticulate cadmium varied from 0 to 0.100 mg/L (Table 1).Such a design was realized by changing the amounts of totalcadmium and gibbsite particles spiked based on the identi-"ed isotherm. The measured cadmium contents of the gillsof the experimental "sh after the exposure treatments arepresented in Fig. 2.

As can be seen in Fig. 2, even though the amountsdissolved cadmium in all treatments were identical, theaccumulation of cadmium in the gills of the "sh exposed toparticulate cadmium at varied concentrations increasedalong with the exposure level, demonstrating an elevatedaccumulation trend. Brief examination of the data in Fig. 2indicates that cadmium accumulation increased almost lin-early with the increase in particulate cadmium concentra-tion. Such results indicate that particulate cadmium wasavailable to "sh through the gill respiration process. Thisresult agrees with the results of previous studies using cop-per and lead. It was reported that after exposure to copperadsorbed on kaolin clay (0}182 mg/L), the copper accumu-lated in the gills of neon tetras (Paracheirodon innesi) (Taoet al., 1999a). Similarly, elevated lead content was observedin the gills of gold"sh (Carassius auratus) exposed to variedconcentrations of particulate lead (0, 0.4, 0.8, 1.2, 2.0, and3.0 mg/L) adsorbed on gibbsite particles (Tao et al., 1999b).For both copper and lead, the accumulation in the "sh gills

FIG. 3. Translocation of particulate cadmium into the "sh gills.

FIG. 4. Gibbsite particles attached to the surface of the gill epithelialcells (squashed specimen, light microscopy, 400]).

310 TAO ET AL.

increased almost linearly with the increase in particulatecopper or lead content in the water.

The uptake of particulate cadmium by the "sh gills wasmore or less expected after the clear evidence that partic-ulate copper and lead are bioavailable to "sh gills. However,it was surprising to see that the uptake e$ciency of partic-ulate cadmium exceeded that of dissolved cadmium in thesame system by the same "sh. According to the results inFig. 2 ("rst data point on the left), around 0.088 mg/kgcadmium was detected in the gills of "sh exposed only todissolved cadmium of 0.01 mg/L (without particulate cad-mium) for 5 days. Compared with this, a concentration of0.879 mg/kg cadmium in the "sh gills was detected afterexposure to 0.50 mg/L particulate cadmium plus 0.01 mg/Ldissolved cadmium for 5 days. The rate of uptake of partic-ulate cadmium by "sh gills was almost twice as much as thatof dissolved cadmium. Further study is necessary to fullyunderstand this phenomenon.

DISCUSSION

Mechanism for Particulate Cadmium Uptake by Fish Gills

A mechanism for uptake of particulate copper and leadvia "sh gills was proposed previously (Tao et al., 1999a, b). Itwas suggested that the key to particulate metal uptake by"sh gills is gill mucus, which is renewed continuously, andthe special microenvironment very near the surface of thegills. Fish gills constitute a sievelike structure formed by gillarches, "laments, and lamellae with the wall of the lamellae.Water #ows through the narrow channels between theselamellae. Gill epithelia are usually covered with an extracel-lular matrix. In "sh, mucus contains a variety of glycopro-teins. The mucus layer is a polyanionic matrix functioningas an ion exchange system, with di!erent a$nities for di!er-ent metals (Part and Lock, 1983). It may also serve toaccumulate and slough o! potentially toxic ions (Varanasiand Murkey, 1978). In addition, the microenvironmentwithin the narrow channels between lamellae is di!erentfrom that of the bulk solution. With the transfer of a numberof chemicals, including NH

3, NH`

4, CO

2, and HCO~

3, the

pH near the gills must be altered. It was reported that thepH in the gill microenvironment of rainbow trout (Salmogairdneri) di!ers from that of bulk solution when the pH ofthe bulk solution is either above or below 5.3 (Randall et al.,1991). The balance point depends on the chemical composi-tion of the water. For the exact same water, it was foundthat the balance pH of "sh gills is 6.9 (Tao et al., 2000). Thismeans that with the pH of the bulk solution around 7.2, thepH of the "sh gill microenvironment is slightly lower. Themicrolayer chemistry was, therefore, possibly favorable formetal desorption from those particles that adhere to themucus layer of the gills.

It is therefore proposed that the gibbsite particles passingthrough the narrow channels could adhere to the surface of

the mucus layer. The rest of the particles on the gill surfaceallowed time for cadmium desorption within the relativeacidic conditions of the gill microenvironment. Since mucuslikely concentrates metals near gill surfaces in close proxim-ity to membrane transport sites (Kirchner, 1987), the desor-bed cadmium could readily translocate into the mucus layerand consequently into epithelial cells. The gibbsite particlesdid not stay on the gill surface and left the gill together withthe sloughed mucus. The processes of particulate cadmiumuptake by gills are illustrated in Fig. 3.

Adherence of Particles to the Gills

The "rst step in the proposed translocation of particulatecadmium is adherence of the gibbsite particles to the mucuscovering the gill surface. To con"rm this assumption,photos were taken of squashed "sh gills specimens under400]microscopy and one of the photos is provided inFig. 4.

The gibbsite particles were easily identi"able by compar-ing the photo with those of a suspended solution of pure

FIG. 5. Aluminum accumulated in the gills of "sh exposed to variousconcentrations of particulate cadmium in the bulk solution.

FIG. 6. Cd/Al ratios in the gills of "sh exposed to various concentra-tions of particulate cadmium in water with constant concentration ofdissolved cadmium at 0.01 mg/L.

UPTAKE OF Cd ADSORBED ON PARTICULATES BY GOLDFISH 311

gibbsite particles serving as a control. It can be clearly seenthat the crystals of gibbsite rested on the surface of the gill"laments which are usually covered by a layer of mucus. Ina similar study on the bioavailability of particulate lead to"sh gills, Tao and colleagues (1999b) observed adherence ofparticles to "sh gills using thin pieces of gill section. Sincethe process of cutting involves a series of solution soakingand washing procedures during which some of the particlesadhering to the gills might be washed away along with somemucus, particles could not be seen on every specimen. Byusing the squashed specimen technique, most particles onthe gill surface remained where they were as indicated inFig. 4. Observations on both gill sections and squashedspecimens demonstrated that some particles that passedthrough the gill channels during aspiration adhered to thesurface of gill epithelia.

The results of a study using Sacramento squaw"sh (Hes-peroleucus symmetricus) exposed to suspended styrenemicrospheres (31-9-m) also revealed that the microspherescould be retained on the mucus-covered buccopharyngealsurface (Sanderson et al., 1998).

Aluminum Contents of Gills

All the evidence presented so far illustrates only the accu-mulation of cadmium in "sh gills and the adherence ofparticles to gills. In fact, the adsorbed cadmium in the watercan be transported into the gill tissues in di!erent ways:desorption and translocation as described previously orphagocytosis together with the particles. The questionremaining is: Which is the dominant process? Martens andServizi (1993) studied suspended sediment particles insidethe gills of Oncorhynchus kisutch, O. gorbuscha, O. nerka, andO. tshawytscha. Through use of light microscopy, electronmicroscopy, and X-ray di!raction they found that theparticles (0.27}0.54 lm) could be phagocytosed by gillepithelial cells. However, the quantity of particlesphagocytosed was very limited. The Phagocytosis of gib-bsite by the gills is a possible cause of cadmium accumula-tion in gill tissue.

If phagocytosis of the particles is a signi"cant mechanismof particulate cadmium uptake, elevated aluminum in thegills would be detected as well. Aluminum content wasmeasured in this study together with cadmium. The resultsare given in Fig. 5. It can be seen that with the presenceof gibbsite in the bulk solution (the last four data points inthe "gure), the aluminum content of gills was signi"cantlyhigher than that of gills exposed only to dissolved cadmium(the "rst data point). However, no increased trend of alumi-num accumulation was observed with an increase in gib-bsite content in the bulk solution.

Rinsing the gill samples in distilled water after excisiondid not guarantee that all particles adhering to the gillsurface, especially those within the "ne structure of the gills,

were removed. Even though gibbsite is generally insolublein water under experimental conditions, some dissolution oftrace amounts of aluminum from the particles could not beentirely ruled out. As such, it is impossible to tell whetherthe aluminum detected in the gill samples originated fromdissolved species in the gill tissue or particles stuck on thesurface of the gills or particles phagocytosed by the gills.

To distinguish the origins of the elevated aluminum in thegills, the aluminum-to-cadmium ratios in the gills werecalculated and compared with the ratio in the bulk solution.The results of this calculation are plotted in Fig. 6 againstthe exposed particulate cadmium in the water, which isproportional to the gibbsite content in the water.

If the accumulation of aluminum in the "sh gills was duemainly to either phagocytosis of gibbsite or adherence of theparticles to the gill surface during sampling and analysis, thecadmium-to-aluminum ratio in the gills would be more orless the same as that in the water. It can easily be calculatedbased on amounts of gibbsite and cadmium spiked (Table 1)that the cadmium-to-aluminum concentration ratio in thewater ranged from 0.0031 to 0.0040 (w/w) in the particulate

FIG. 7. Comparison of Cd/Al ratios in water and gills.

312 TAO ET AL.

cadmium exposure experiment. On the other hand, themeasured cadmium-to-aluminum ratio in the gills variedfrom 0.0369 to 0.212 (w/w). The latter is at least one order ofmagnitude higher than the former depending on the partic-ulate cadmium concentration in the water. This signi"cantdi!erence is graphically indicated in Fig. 7.

The more favorable cadmium accumulation in the gillswhen compared with aluminum indicates that neitherphagocytosis nor adherence of the cadmium adsorbed par-ticles was the primary mechanism causing cadmium accu-mulation in the "sh gills. The only possible mechanism thatcannot be ruled out is the stripping, or desorption, of cad-mium from the gibbsite particles that temporarily adheredto the gill surface mucus, as illustrated in Fig. 3.

CONCLUSION

After exposure to particulate cadmium adsorbed on gibb-site, cadmium accumulation in "sh gills increased with theincrease in the exposure concentration of the particulatecadmium. Evidence collected demonstrated that the pri-mary mechanism of particulate cadmium uptake by "sh gillsis a four-step process: (1) adherence of the particles to thesurface of the gills covered by a layer of mucus; (2) desorp-tion of cadmium from the particles under favorable condi-tions in the gill microenvironment; (3) translocation of thecadmium into gill epithelia; and (4) sloughing o! of theparticles along with the mucus.

ACKNOWLEDGMENTS

Funding was provided by The National Excellent Young Scientist Fundof China (49525102) and International Copper Association, Ltd.(TPT0604).

REFERENCES

Allen, R. J. (1996). ¹he Role of Particulate Matter in the Fate of Con-taminants in Aquatic Ecosystems, Sci. Ser. No. 142. Inland Water Direc-torate, National Water Res. Inst., Burlington, Ontario.

Benson, W. H., Alberts, J. J., Allen, H. E., and Hunt, C. D. (1994). Synopsisof discussion session on the bioavailability of inorganic contaminants. InBioavailability: Physical, Chemical, and Biological Interactions (J. L.Hamelink, P. F. Landrum, H. L. Bergaman, and W. H. Benson, Eds.),pp. 63}71. Lewis, Boca Raton, FL.

Bingham, F. T., Sposito, G., and Strong, J. E. (1984). The e!ect of chlorideon the availability of cadmium. J. Environ. Qual. 13, 71}74.

Cope, W. G., Wiener, J. G., and Steingraeber, M. T. (1996). Test system forexposing "sh to resuspended contaminated sediment. Environ. Pollut.91, 177}182.

Dougan, W. K., and Wilson, A. L. (1974). The absorptiometric deter-mination of aluminum in water, a comparison of some chromogenicreagents and the development of an improved method. Analyst 99,413}430.

Erickson, R. J., Bills, T. D., Clark, J. R., Hansen, D. J., Knezovich, J. P.,Mayer, F. L., Jr., and McElroy, A. E. (1994). Synopsis of discussionsession on physicochemical factors a!ecting toxicity. In Bioavailability:Physical, Chemical, and Biological Interactions (J. L. Hamelink, P. F.Landrum, H. L. Bergman, and W.H. Benson, Eds.), pp. 31}38. Lewis,Boca Raton, FL.

Kirchner, L. B. (1987). External changed layer and Na` regulation. InOsmotic and <olume Regulation (C. Karker-Jorgensen, E. Skadhauge,and J. Hess-Thaysen, Eds.), pp. 310}332. Academic Press, New York.

Kuimr, A., Kaur, I., and Mathur, R. P. (1996). Speciation pro"le of cad-mium and its bioconcentration in an aquatic system. Environ. ¹echnol.17, 629}635.

Leland, H. V., and Kuwabara, J. S. (1985). Trace metals. In Fundamentals ofAquatic ¹oxicology (G. M. Rand, and S. R. Petrocelli, Eds.), pp. 374}415.Hemisphere, New York.

Liang, T., Tao, S., and Lam, K. C. (1999). Accumulation of copper in "sh(gill removing) and "sh gill. Acta Ecol. Sin. 19(99), 763}766.

Martens, D. W., and Servizi, J. A. (1993). Suspended sediment particlesinside gills and spleens of juvenile paci"c salmon (Oncorhynchus spp.).Can J. Fish Aquat. Sci. 50, 586}590.

Morel, F. (1984). Principles of Aquatic Chemistry. Wiley, New York.

Newman, M. C., and Jagoe, C. H. (1994). Ligands and the bioavailability ofmetals in aquatic environments. In Bioavailability: Physical, Chemical,and Biological Interactions (J. L. Hamelink, P. F. Landrum, H. L.Bergman, and W. H. Benson, Eds.), pp. 39}62. Lewis Boca Raton, FL.

Part, P., and Lock, R. A. C. (1983). Di!usion of calcium, cadmium andmercury in a mucuous solution from rainbow trout. Comp. Biochem.Physiol. C 75, 259}265.

Playle, R. C. (1998). Modelling metal interactions at "sh gills. Sci. ¹otalEnviron. 219, 147}163.

Playle, R. C., and Wood, C. M. (1989). Water pH and aluminum chemistryin the micro-environment of rainbow trout during acid and aluminumexposures. J. Comp. Physiol. B 159, 539}550.

Randall, D., Lin, H., and Wright, P. A. (1991). Gill water #ow and thechemistry of the boundary layer. Physiol Zool. 64, 26}38.

Reid, S. D. (1989). Metal}Gill Surface Interactions in Rainbow ¹rout,Oncorhynchus mykiss. Ph.D. thesis, McMaster University, Hamilton,Ontario.

Reid, S. D., and McDonald, D. G. (1991). Metal binding activity of the gillsof rainbow trout (Oncorhynchus mykiss). Can. J. Fish Aquat. Sci. 48,1061}1068.

Roy, R., and Campbell, P. G. C. (1995). Survival time modeling of exposureof juvenile Atlantic salmon (Salmo salar) to mixtures of aluminumand zinc in soft water at low pH. Aquat. ¹oxicol. (Amsterdam) 33(2),155}176.

UPTAKE OF Cd ADSORBED ON PARTICULATES BY GOLDFISH 313

Sanderson, S. L., Mort, M. E., and Cech, J. J., Jr. (1998). Particle retentionby non-suspension-feeding cyprinid "shes. Can. J. Fish Aquat. Sci. 55,861}868.

Simkiss, K. (1983). Lipid solubility of heavy metals in saline solutions.J. Mar. Biol. Assoc. ;K 63, 1}7.

Tao, S., Liang, T., Liu, C. F., and Xu, S. P. (1999a). Uptake of copper byneon tetras (Paracheirodon innesi) in the presence and absence of partic-ulate and humic matter. Ecotoxicology 8, 269}275.

Tao, S., Liu, C. F., Dawson, R., Cao, J., and Li, B. G. (1999b). Uptake ofparticulate lead via the gills of "sh (Carassius auratus). Arch Environ.Contam. ¹oxicol. 37, 352}357.

Tao, S., Liang, A. M., Liu, C. F., and Dawson, R. (2000). Copper speciationin the gill microenvironment of carp (Cyprinus carpio) at various pH.In press.

Varanasi, U., and Murkey, D. (1978). Uptake and release of lead andcadmium in skin and mucus of coho salmon (Onchorhynchus kisutch).Comp. Biochem. Physiol. C 60, 187}194.

Wen, Y., Tao, S., Long, A. M., Liu, C. F., and Xu, F. L. (2000). Speciation ofcopper in the "sh microenvironment. Environ. Chem. in press.

Wright, P. A., and Wood, C. M. (1985). An analysis of branchial ammoniaexcretion in the freshwater rainbow trout: E!ects of environmental pHchange and sodium uptake blockade. J. Exp. Biol. 114, 329}353.