• 沒有找到結果。

Mode of Action and Growth Toxicity of Arsenic to Tilapia Oreochromis mossambicus Can Be Determined Bioenergetically

N/A
N/A
Protected

Academic year: 2021

Share "Mode of Action and Growth Toxicity of Arsenic to Tilapia Oreochromis mossambicus Can Be Determined Bioenergetically"

Copied!
9
0
0

加載中.... (立即查看全文)

全文

(1)

Mode of Action and Growth Toxicity of Arsenic to Tilapia

Oreochromis mossambicus Can Be Determined Bioenergetically

J. W. Tsai,1C.-M. Liao1 1

Ecotoxicological Modeling Center, Department of Bioenvironmental Systems Engineering, National Taiwan University, Taipei, Taiwan 10617 Republic of China

Received: 26 April 2005/Accepted: 26 June 2005

Abstract.We present a bioenergetics-based approach to ana-lyze the chronic effects and growth toxicity mode of action in tilapia Oreochromis mossambicus exposed to waterborne As and to predict fish growth under different exposure scenarios. 7-dayexposure bioassays showed that tilapia accumulate As when exposed to waterborne As. We conducted growth bio-assays to assess chronic As toxicity to tilapia. We incorporated a universal ontogenetic growth model with the DEBtoxtheory to explore the mode of action of As toxicity. Our results show that the specific growth rates of exposed tilapia are inversely proportional to As concentrations and are calculated as 0.76% d)1in 0 lg mL)1, 0.57% d)1in 1 lg mL)1, 0.2% d)1in 2 lg mL)1, and 0.04% d–1 in 4 lg mL)1 As, respectively. We showed that the internal threshold concentration did not change significantlywith time, demonstrating that the critical bodyresidue approach is applicable for As toxicityassess-ment. We distinguished between three modes of action of As, including direct effects on growth and indirect effects byway of maintenance and food consumption. Our results support that decreased feeding accounts for the growth decrease in the case of feeding ad libitum. The feeding decrease model also illus-trates the growth trajectories of tilapia during the entire whole life span, suggesting that the maximum biomass of tilapia are 1038.75 g in uncontaminated water and 872.97 g in 1 lg mL)1, 403.06 g in 2 lg mL)1, and 336.65 g in 4 lg mL)1As, respectively. We suggest that considering modes of action in ecotoxicologynot onlyimproves our understanding of the toxicities of chemicals, it is also useful in setting up models and avoiding pitfalls in species- and site-specific environ-mental risk assessment. This proposed framework for tilapia gives preliminaryinformation relevant to aquacultural and ecologic management.

Long-term ingestion of the groundwater contaminated by inorganic As has been found to induce blackfoot disease (BFD) in residents of the southwestern coastal area of Taiwan (Chen

et al. 2001). Currently, most of the people living in these areas do not drink water from artesian wells because tap water has been made available in this area. However, artesian well water is still used for aquaculture. Farming tilapia (Orechromis mossambicus) is one of the most promising aquatic endeavors in the BFD area because of its high market value. Liao et al. (2003) pointed out that the As concentration in BFD-area pond water ranged from 8.1 to 251.7 lg L–1. Arsenic contents in several farming ponds exceed the water-qualitycriteria for total As in freshwater ecosystems (150 lg L–1) as documented by the Criterion Continuous Concentration (United States Envi-ronmental Protection Agency[USEPA] 2002). If As levels in pond water increase, severe effects mayoccur to the health of farmed fish and mayeven pose a potential risk to the people who consume tilapia farmed in the BFD area.

The use of assimilated energyhas been extensivelyem-ployed by physiology and ecosystem scientists in recent years to determine the growth of organisms and the productivityof ecosystems (Kooijman and Bedaux 1996; Beyers et al. 1999; Sherwood et al. 2000). Fish constantlyconsume energyto maintain life and offset the effects of multiple stressors such as dailyfluctuations in water temperature, availabilityof food, and pollutants in the environment (Wedemeyer et al. 1984). Therefore, assessing the impact of chronic exposure to chemicals byusing energymetabolism as a performance re-sponse could be a rigorous physiologic and ecologic approach to toxicityassessment.

Organisms acquire energybyingesting food from their environment. The assimilated energyis stored in reserve be-fore biologic use. Pery et al. (2003) pointed out that exposure to a toxic chemical maydecrease resource acquisition from the environment and cause a decrease in reproduction rate. Beyers et al. (1999) indicated that organisms must compensate for these chemical stresses with detoxicification mechanisms, which require energy, and their effect can be evaluated using a bioenergetic model. Because maintenance (including detoxi-cification) cost has priorityover growth in fish bioenergetic theory, maintenance cost competes with growth investment for the allocation of energythat is used from the reserves, and a decrease in assimilation translates into a decrease in the amount of energythat is used from the reserves (Beyers et al.

(2)

1999; Congdon et al. 2001). Therefore, an increase in the energycost for life maintenance could leads to a decrease in growth rate (Congdon et al. 2001).

A mode of action is defined as a common set of physiologic and behavioral signs that characterize a type of adverse bio-logic response (Landis and Yu 1999). Escher and Hermens (2002) indicated that elucidating the detailed chemical-specific modes of a metalÕs toxic action could enhance the prediction power of models byproviding a mechanistic explanation for chemical risk assessment in ecotoxicology. Barata and Baird (2000) further suggested that the ecotoxicologic modes of action of different chemicals can be determined bioenergeti-callybystudying sublethal effects on food acquisition and hence growth and reproduction rates.

In the present study, specific efforts were paid to quanti-tativelyrelate As concentrations in tilapia to extent of growth inhibition. We conducted bioassays to determine if growth decrease occurs in chronic-exposure conditions, including a 7-daybioaccumulation test to determine the toxicokinetic process of As and a chronic bioassayto observe the organ-ismÕs growth trajectories in different exposure scenarios. We further developed bioenergetics-based mechanistic models to elucidate and predict growth effects of chronic As exposure. The objectives of this studywere threefold: (1) to quantita-tivelydetermine the relations between As exposure and growth inhibition; (2) to identifythe mode of action domi-nating As growth inhibition; and (3) to develop a residue-based mechanistic growth model to predict individual growth in different exposure scenarios. We believe a comprehensive understanding of the mode of action of As toxicityto tilapia will be of great benefit to aquaculture management.

Materials and Methods

Test Fish and Experimental Protocol

Male tilapia Orechromis mossambicus, age 8 to 9 months (mean body length 12.9 € 1.54 cm (mean € SD) and mean weight = 10.58 € 1.52 g wet weight), were supplied byTaiwan Fisheries Research Institute (Tainan, Taiwan), where theyare hatched in the laboratoryand con-sidered uncontaminated byAs. Tilapia were visiblyfree of any deformities, lesions, or diseases. Fish were kept on ice during trans-port from Tainan to the Ecotoxicological Modeling Center, Depart-ment of BioenvironDepart-mental Systems Engineering, National Taiwan University, Taipei, Taiwan. On arrival in our laboratory, the fish were allowed to acclimate in tap water at 27.7C € 0.24C during a light-to-dark cycle of 12:12 for at least 14 days before the initiation of exposure tests. Fish were fed dailytwice with artificial food, and water pH ranged from 7.6 to 7.8. Mortalitywas <5% of the population during acclimatization, and no weight losses were observed.

The experiments employed an aqueous exposure route, so all test media were prepared using deionized water. Chemical stock solutions were prepared bydissolving a calculated amount of reagent-grade sodium arsenite (NaAsO2) in deionized water, and the new stock solutions were prepared as needed during the toxicitytests. All experiments were carried out in 54-L indoor rectangular fiberglass aquaria such that the dissolved oxygen in each tank was maintained at close to saturation byaeration (7.21 € 0.1 lg mL–1). The temperature in each aquarium was maintained at 26.7C € 0.24C using sub-merged heaters. Water pH was maintained at 7.75 € 0.02. The

pho-toperiod was 16 hours of light to 8 hours of dark with a light intensity of 1400 € 100 lux. All of the experiments were assigned to two replicate tanks. We replaced 40% to 60% As solution every1 to 2 days to avoid the regression of ambient water quality. To keep the As concentration constant, the entire As solution was replaced weeklyin each tank.

Three series of semistatic tests were conducted in this study. In the first series, we conducted a range-finding test to determine the As contamination level in the As bioaccumulation and chronic toxicity bioassays by exposing tilapia to differing As concentrations of 0.25, 0.5, 1, 2, 4, and 6 lg mL–1. The results of the preliminarytest re-vealed that the median lethal tolerance (LT50) of tilapia at £ 4 lg mL–1As was >28 days. In the second series, a bioaccumulation assay was conducted to investigate the time course of uptake and depuration of chemicals in tilapia. We conducted an uptake experiment in As concentration of 1 lg mL–1for 7 days based on the suggestion by Suhendrayatna et al. (2002). The measured As concentration was 0.89 € 0.06 lg mL–1As in the bioassay. The As concentrations used in this experiment were 20 to 50 times higher than that in the field environment conditions needed to produce high As levels in tilapia.

Uptake Experiment

The fish were fed a commercial fish food once a day, 7 days a week, at a low rate of 0.5% fish biomass to avoid As contamination of the feed remaining in the aquaria. Uneaten food and faces were siphoned from the bottom of the aquaria every day. To conduct analysis of As accumulation kinetics, five fish were sequentiallyharvested from solutions after 0, 1, 2, 4, and 7 days of exposure. The fish were rinsed with deionized water and then anesthetized in pH-neutralized tricaine methane sulfonate (MS-222) (Sigma Chemical, St. Louis, MO) solution. Fish samples were freeze-dried overnight and then ground to fine powder in a grinder (Tai-Hsiang S36-89, Taiwan). A 500-mg portion of the powder was digested in 10 mL concentrated HNO3 (65% weight) overnight at room temperature. The resulting solution was evaporated and the residues redissolved in 0.1 N HCl.

Chronic Toxicity Bioassays

We conducted a 4-week chronic toxicitybioassayto determine the toxic effects on the tilapia growth response when exposed to water-borne As concentrations. The nominal As concentrations for the chronic test were 1, 2, 4, and 0 lg mL–1, and the corresponding measured As concentrations were 0.87 € 0.35, 1.77 € 0.34, and 3.56 € 0.69 lg mL–1, respectively. All of the chronic tests were re-peated 2 times, and each concentration was assigned to 2 replicate tanks for 28 days. For each dose of As, 10 tilapia were exposed. Fish were fed 2 times/d with commercial fish food at a rate of 4% fish biomass. Uneaten food was siphoned from the aquaria 30 minutes after feeding. We replaced 50% to 60% As solution every1 to 2 days to avoid the regression of ambient water qualityand As concentration. The entire As solution was replaced weeklyin each tank. Mortality was monitored at 0, 6, and 12 hours through the first dayof exposure, then twice dailyuntil the end of the test. Everyweek, the mean body weights of each exposed group were recorded, and we calculated the growth rates in different As concentrations.

Chemical Analysis

A Perkin-Elmer Model 5100PC atomic absorption spectrometer (Per-kin-Elmer, Shelton, CT) equipped with an HGA-300 graphite furnace atomizer was used to analyze As. Analytic quality control was

(3)

achieved by digesting and analyzing identical amounts of rehydrated (90% H2O) standard reference material (dogfish muscle, DORM-2; NRC-CNRC, Canada). Recoveryrate was 94.6% € 3.6%, and levels of detection were 0.62 lg As L–1for water samples and 0.05 lg As g–1 for tissue samples. We detected As concentrations in each test media, and exposure water characteristics during the test were measured three times weeklyin one selected replicated aquarium for analysis of As. The 10-mL water samples were acidified (pH < 1) with 5 mL 1 N HNO3and then stored at –4C in the dark until theywere analyzed.

Data Analysis

The specific dailygrowth rate (kg,% d–1) of tilapia was estimated as (Sherwood et al. 2000), kg¼ ln Wt W0   =dt  100 ð1Þ

where Wtand W0is the bodyweight of tilapia at time t and the initial of experiment, respectively. Determination of toxicokinetic parame-ters was done byfitting concentration data to the integrated form of the kinetic equation for constant water exposure using iterative non-linear regression (Reinfelder et al. 1998; Clason et al. 2003),

CfðtÞ ¼ Cfð0Þeðk2þkgÞtþ k1 k2þ kg Cw 1 eðk2þkgÞt   ; ð2Þ

where Cfis the time-dependent As concentration in tilapia (lg g–1), k1 is the tilapia uptake rate constant (mL g–1d–1), k

2is the depuration rate (d–1) constant, and t is the time in days. The bioconcentration factor (BCF) can be calculated as: BCF = k1/(k2+kg), representing the net accumulation abilitythat is the result of the competition between uptake and depuration associated with growth dilution, and Cwis the mean measure waterborne As concentration (lg mL–1). Equation 2 provides a toxicokinetics-based approach to predict the accumulative As profile in constant-exposure scenarios.

We expressed the growth of tilapia as growth coefficient (mean bodyweight after 1, 2, 3, and 4 weeks/mean bodyweight at the start of the experiment) (Gomot 1997) for each As concentration every week with respect to initial bodyweight before exposure to As. The values of the growth coefficient for each concentration were plotted with arithmetic coordinates with corresponding regression equations. The curves obtained give an estimation of the external effect con-centrations for 10% growth inhibition (EC10). The USEPA (2000) recommended that the internal effect concentration causing 10% re-sponse (IEC10) could be used as a surrogate threshold of regulatory end point in ecologic risk assessment and that the IEC10 can be estimated from the critical bodyresidue (CBR) model as IE-C10= EC50· BCF (McCartyand Mackay1993).

Models

We attempted to construct a bioenergetics-based model that reflects the mode of action to simulate the growth of tilapia under different exposure scenarios. The DEBtox theory(Kooijman et al. 1996) de-scribes the modes of action of chemical toxicitybased on the emphasis of resource allocation, a bioenergetics-based viewpoint that differs from the general aspect describing changes in physiology or behavior inhibition. The basic assumption of the DEBtoxtheoryis that an organism must take up chemicals before theycan exert an effect. Second, once the chemical is inside the target tissues, it increases the

probabilityof an adverse response and affects a parameter of the general ontogenetic growth model (e.g., the assimilation rate). DEBtox indicates that chemical effects act bywayof three types of mode of action including direct effects on growth and indirect effects on maintenance and food assimilation and that onlyone of these effects occurs at a time in the lower effect range of the chemical (Kooijman et al. 1996).

West et al. (2001) developed a mechanistic model, referred to as the West growth model, to describe ontogenetic growth trajectories of organisms instead of using the conventional growth model based on a statistical approach. The West growth model is a general quantitative model based on fundamental principles for the use of the consumed energybetween maintenance of existing tissue versus the reproduction of new biomass, and it has described the growth of manydiverse species successfully(West and Brown 2004). This model character-izes the slowing of growth as bodysize increases as being related to limitations on the capacityto supplysufficient resources to support further increase in bodymass. We adapted the West growth model as the growth model without toxicity(West et al. 2001):

WðtÞ ¼ Wmax 0 1 1  W0 Wmax 0  1=4 " # eA0t=4W1=4max 0 ( )4 ; ð3Þ

where Wmax0and W0are maximum bodyweight (g) in uncontami-nated water and mass at birth (g), respectively. A0is a species-specific growth coefficient (g1/4 d–1) in that A0 ” B0mcEc0–1, where B0is a taxon-specific constant (W), mcis the mass of a cell (g), and Ec0is the metabolic energyrequired to create a new cell (J). A0can be estimated byoptimal fitting Equation 3 to the body-growth profile in control exposure conditions.

We distinguished three modes of toxic action on As growth inhi-bition in tilapia: (1) increased cost of growth, (2) increased cost of maintenance, and (3) decrease feeding. McCartyand Mackay(1993), Kooijman and Bedaux (1996), and Pery et al. (2003) suggested that we mayrelate the extent of adverse effects proportional to the dif-ference between accumulated chemical concentration (Cf(t)) and IEC10 in that IEC10 is adapted as the effect threshold for chronic growth inhibition. We introduced a stress function (S(t)) to describe the extent of adverse effect as (Kooijman and Bedaux 1996; Pery et al. 2003):

SðtÞ ¼ b½CfðtÞ  IEC10ðtÞ; ð4Þ

where b accounts for the level of toxicity(g g–1) once Cf exceeds IEC10. We predicted Cfin various exposure scenarios byincorporat-ing experimental-derived toxicokinetic parameters into Equation 2.

In the case of increase growth energycost, i.e., Cfexceeds IEC10, we assume that the metabolic energy(Ec) required to create a new cell are multiplied by[1+S(t)] and expressed as Ec= Ec0[1+S(t)], where Ec0is the growth energycost in control condition. We have S(t) = 0 and Ec= Ec0when Cf £ IEC10. We substituted the effect function into the West growth model, obtaining the mode of action on growth cost:

WðtÞ ¼ 1130 1  1  0:05 1130  1=4 " # eAt=411301=4 ( )4 ; ð5Þ

where A = B0mc(Ec0[1+S(t)])–1= A0[1+S(t)]–1and the two constants of 1130 and 0.05 represent the maximum biomass and the mass at

(4)

birth (g) of tilapia, respectively, in uncontaminated water (www.fishbase.org/home.htm).

When maintenance energycost increases, chemicals are likelyto increase maintenance costs to compensating for the effects of expo-sure (Beyers et al. 1999). Because maintenance cost has priorityover growth, such an increase leads to decreased growth rate. We multi-plied bodyweight by[1+S(t)] to account for an increase in the maintenance costs, resulting in the decreased time-dependent body weight (Kooijman et al. 1996):

WðtÞ½1 þ SðtÞ ¼ 1130 1  1  0:05 1130  1=4 " # ( eA0t=411301=4 o1=4 : ð6Þ

In contrast, when feeding decreases, a growth decrease acts by decreasing incoming energy. The maximum assimilation rate does not appear in the West growth model, yet it can be captured by maximum weight (Wmax) (Kooijman and Bedaux 1996). Here, maximum weight is defined analogouslyto the definition of maintenance cost to account for the growth decrease effect on assimilation as Wmax = Wmax0 [1-S(t)], and substituting that relation into the West growth model (Equation 3) leads to:

WðtÞ ¼ Wmax 1 1  0:05 Wmax  1=4 " # eA0t=4W1=4max ( )4 : ð7Þ

Equations 5, 6, and 7 describe the modes of action that lead to effects on either coefficient of growth cost (A), time-dependent body weight (W(t)), or ultimate bodyweight (Wmax). Parameters of A0, S(t), and Wmaxare estimated byfitting the West growth model (Equation 3) and three effect models (Equations 5, 6, and 7) to

concentration-specific growth data using iterative nonlinear regression. A standard analysis of variance test (ANOVA; ScheffeÕs t test) was employed to determine the significance of differences between model values and mean actual data on bodyweight in different groups. In addition, goodness-of-fit was evaluated using the sum of squares (SSs) between the description and data, computed from SS =P

N i¼1

ðxi XiÞ2, where N denotes the number of measurements, xiis the predicted data, and Xiis the measured result corresponding to data point i.

We employed the nonlinear option of Statistica software (StatSoft, Tulsa, OK) to perform all curve fittings in this study. Statistica soft-ware was also used to calculate the coefficient of determination (r2) and statistical analyses (ANOVA and Student t test). Statistical sig-nificance was judged at p < 0.05.

Results

Toxicokinetics

The 7-daywater-exposure bioassayof As in tilapia was sig-nificantlycorrelated with nonlinear regression profiles (r2 = 0.97, p < 0.05) resulting from the best fit of the first-order bioaccumulation model (Fig. 1A). The estimated uptake rate constant (k1), depuration rate constant (k2), and BCF were 0.39 mL g–1d–1, 0.075 d–1, and 4.70 mL g–1, respectively. The BCF was >1, showing that the tilapia accumulated waterborne As. The toxicokinetic parameters not onlydescribed As kinetics in tilapia but also could be applied to predict As residue in tilapia. We assumed that toxicokinetic parameters were independent of As concentration in chronic-exposure conditions. We em-ployed Equation 2, cooperating with experimental-derived toxicokinetic parameters, to predict the profiles of As kinetics when tilapia were exposed to 1, 2, and 4 lg mL–1waterborne 0 5 10 15 20 25 30 35 0 10 20 30 40 50 60 70 80 90 100

Time (d)

A

s

co

n

ce

n

tr

a

tio

n

in

tila

p

ia

(

g

g

-1

dr

y w

t.

)

B 4 µ 1 2µg mL-1 0 0.5 1 1.5 2 2.5 0 1 2 3 4 5 6 7 r2 = 0.97, p < 0.05 A

Fig. 1. (A) Bioassays of tilapia exposed to 1 lg mL)1waterborne As during a 7-dayuptake. Symbols represent mean € 1 SE (n = 5). The solid line is the best-fit regression curve from the one-compartment bioaccumulation model of tilapia. (B) The predicted As concentration in tilapia when exposed to 1, 2, and 4 lg mL)1As during a 100-dayuptake

(5)

As, respectively(Fig. 1B). The corresponding steady-state As concentrations in the tilapia were 5.2 lg g–1 in 110 days, 10.4 lg g–1in 85 days, and, 20.8 lg g–1in 77 days, respectively (Fig. 1B).

Chronic Toxicity

Our bioassays revealed that all of the exposure concentrations (1, 2, and 4 lg mL–1) affected the growth of tilapia (Fig. 2). In the control groups, the tilapia grew progressivelyfrom 12.89 to 15.95 g with the extension of duration, and the specific growth rate (SGR) was calculated as 0.76% d–1(Fig. 2). The SGR of tilapia exhibited high variation in different experimental set-tings. The SGR of our control group fell within the reported values, which range from 0.35% to 1.8% d–1 (Balasubrama-nian and Bai 1996; Uchida et al. 2003).

At the intermediate concentrations, between 1 and 4 lg mL–1 As, a clear growth inhibition was observed from the second week. Biomass loss even occurred in the second week in 2 lg mL–1As and in the second and third weeks in 4 lg mL–1 As. The SGRs of exposure tilapia were negatively correlated with As concentrations and calculated as 0.57 % d–1in 1 lg mL–1, 0.25% d–1in 2 lg mL–1, and 0.04 % d–1 in 4 lg mL–1As, respectively. The tilapia almost tended to stop growing in 4 lg mL–1 As, the SRG approximately19 times lower than that in the control, showing that the fish were suffering sublethal effects rather than chronic growth inhibitions (Fig. 2). Liao et al. (2004) indicated maximum mortalityup to 70% when the tilapia were exposed to 4 lg mL–1waterborne As.

We used the concentration-specific growth coefficients to establish the regression equations (Table 1 and Fig. 3). We calculated EC10values and then estimated IEC10values using the CBR approach. The IEC10s were used as an internal threshold concentration. The EC10s did not change signifi-cantlywith time and ranged from 0.38 to 0.41 lg mL–1

(Fig. 3). The constant EC10value demonstrated that the CBR approach is applicable to estimate the corresponding internal threshold concentration (i.e., IEC10) in chronic As toxicity assessment.

Mode of Action

Figure 4 shows the optimal fits of the feeding-decrease, growth-cost, and maintenance-cost models to the experimental data. Table 2 lists the estimates of model parameters. In the growth-cost model, the species-specific growth coefficient (A) seems not to depend on the exposure As concentration, revealing that the metabolic energyrequired to create a cell does not change significantlyin different exposure conditions. Our results indicated that the growth-cost model could not discriminate the mode of action of As toxicitywell (Fig. 4). In the maintenance-cost model, the values of toxic stress (S(t)) increased slightlywith increasing As concentrations, revealing that this model can describe the decrease in fish bodyweight with increases in As concentration. In the feeding-decrease model, the estimated maximum bodyweight (Wmax) was negativelycorrelation with As concentrations, which means that the As has direct effects on maximum bodyweight by decreasing the appetite.

Statistical analyses indicated that no significant differences (p > 0.05 with Student t tests) were observed between the means of data and the descriptions of the three models in As con-centrations of 0 and 1 lg mL–1, resulting in difficulties of assessing As toxicityin lower exposures (£ 1 lg mL–1). However, significant differences existed between descriptions of the maintenance-cost and growth-cost models and the measured data in the 2- and 4-lg mL–1experiments. We further employed SSs of the differences between model predictions and mean actual data to assess the performance between three effect models (Table 3). Obviously, the fits of the feeding-decrease model were more accurate than the others. The values of SSs increased with the gradient of the concentrations, e.g., the feeding-decrease model, from 0.009 to 0.245 g2. Similar results also occurred in the fitting of the other two models. Kooijman and Bedaux (1996) indicated that there might not be onlyone mode of action accounting for the toxic effect at higher concentrations, that several other physiologic processes might be responsible, indicating that the proposed DEBtox– based toxicitymodels maybe restricted to describe the growth

Table 1. Estimated toxic effects of regressive equations of As on growth for tilapia O. mossambicus after 1 to 4 weeks

Time (wk) Regression equation r2 EC10 (lg mL-1) IEC10 (lg g)1)a 1 Y =1.051–0.016X 0.75 0.38 1.79 2 Y = 1.084–0.033X 0.71 0.43 2.02 3 Y = 1.159–0.041X 0.84 0.40 1.88 4 Y = 1.221–0.055X 0.95 0.41 1.93 aIEC 10= EC10· BCF, where BCF is 4.70 mL)1g)1. BCF = Bioconcentration factor.

EC10= Effect concentration for 10% growth inhibition. IEC10= Internal effect concentration choosing 10% response.

wt .) t we t ( ig h we dy Bo 12 12.5 13 13.5 14 14.5 15 15.5 16 16.5 Time (week) g 0 4 3 2 1 0 4 1µg mL-1 2

Fig. 2. Growth curves of tilapia O. mossambicus exposed to water-borne As concentrations during 4-week bioassays. Symbols (open squares, closed triangles, open circles, and closed diamonds) are the measured data.

(6)

trajectories of organisms when theyare suffering sublethal or acute toxic effects. Our results support that the mode of action of As growth inhibition acts through decreased feeding, i.e., the growth decrease acts through decreasing incoming energy.

Prediction of Tilapia Growth

We employed the feeding decrease model to illustrate the growth trajectories of tilapia from birth to natural death in different exposure scenarios (Fig. 5). The reported life span of tilapia O. mossambicus is 11 years (approximately 4000 days) (www.fishbase.org/home.htm). The maximum bodyweight (Wmax) of the control tilapia and the 1 lg mL–1As group were 1038.75 and 872.97 g, respectively, whereas for the groups exposed to 2 and 4 lg mL–1As, the corresponding Wmaxs were 403.06 and 336.65 g, respectively(Fig. 5). Tilapia potentially grow to maximum bodyweight until their end of life. This is comprehensible because in uncontaminated conditions, when

fish are feeding ad libitum, individuals store surplus metabolic energyin reserve, which causes an increase in biomass even theyhave alreadyreached mature bodysize. In contrast, when fish are consistentlyexposed to higher concentrations during a longer duration, fish translate large amount of assimilated energyfrom growth or maintenance to compensate for the stress of toxicants, thus inducing growth cessation or inhibition (Beyers et al. 1999; Sherwood et al. 2000).

Discussion

Chronic Toxicity and Toxicokinetics

Because of the scarcityof chronic toxicitydata, acute-to-chronic ratios often are traditionallyemployed to derive qualitystandards for prolonged exposure to toxicants; how-ever, changes in toxicitywith long-term exposure might be attributed to a change in the mode of action and the induction of physiologic acclimation or genetic adaptation to local contaminant regimes (Forrester et al. 2003). This would cause limitations in assessing the long-term chemical effects with acute toxicitydata because it seems plausible that organisms might somehow become weakened after enduring long-term chemical loading, and, nonspecifically, initially sublethal ef-fects might worsen with time.

We assumed that chronic toxicityis initiated when the accumulated chemical exceeds the internal threshold concen-tration, represented byIEC10. The magnitude-of-toxicityeffect can be expressed as being proportional to the difference be-tween accumulated chemicals and IEC10 and can be formu-lated as a stress function as shown in Equation 4. IEC10can be accuratelyderived from the chronic bioassays data using

sta-0.98 1 1.02 1.04 1.06 1.08 01 23 0.95 1 1.05 1.1 1.15 1.2 0 1 2 3 4 0.92 0.96 1 1.04 1.08 1.12 1.16 0 1 2 3 4 Week 1 Week 2 4 1 1.05 1.1 1.15 1.2 1.25 1.3 0 1 2 3 4 Data Fitted model Gr owth coefficient As concentration in water (µg mL-1) Week 4 Week 3

Fig. 3. Growth coefficients of tilapia versus nominal As concentrations during different exposure periods

Table 2. The estimated effect parameters of the growth-cost, main-tenance cost, and feeding-decrease modelsa

Treatment (lg mL)1) Growth cost model, A (g1/4d)1)b Maintenance cost model, S(t)c Feeding decrease model, Wmax(g)d Control 0.024 € 0.006 0.00 € 0.00 1100.82 € 21.49 1 0.023 € 0.006 0.03 € 0.02 924.00 € 69.79 2 0.023 € 0.005 0.11 € 0.03 421.54 € 49.23 4 0.022 € 0.005 0.16 € 0.05 352.13 € 52.74 Data are expressed as mean € SD.

aEstimated from Eq. (5). bEstimated from Eq. (6). cEstimated from Eq. (7).

(7)

tistical techniques. Thus, the extent of toxicityis strongly determined bypredicted As residue. Our simulations eluci-dated that As residue in tilapia was proportional to waterborne concentrations. The first-order bioaccumulation model has been extensivelyapplied to describe and predict chemical kinetics in aquatic organisms (Reinfelder et al. 1998). McGeer et al. (2003) pointed out that the first-order BCF-based bio-accumulation model for metals is onlyapplicable for residue predictions in the lower range of exposures, in which the up-take process does not limit the rate of upup-take. Suhendrayatna et al. (2002) indicated that the higher concentrations (>10 lg mL–1) of As (III) are toxic to tilapia, thus affecting accumu-lation of As bytilapia, and the total As accumulated in tilapia is proportional to external concentrations <5 lg mL–1. We confirmed that our hypothesis that As residues in tilapia under

chronic-exposures conditions (£ 4 lg mL–1) would be almost completelycaptured byour proposed model.

Mode of Action of Growth Inhibition

Our studyrevealed that As toxicityacts bycausing a decrease in feeding. Although the mechanism accounting for growth decrease is statisticallysignificant, the biologic meaning of this result remains unclear. Rankin and Dxion (1994) pointed out that an immediate decrease in feeding in response to both waterborne and dietaryAs exposure has been observed in freshwater fish species. Health (1995) pointed out that de-creased food consumption frequentlyoccurs with chemical exposure, especiallyduring the earlydays of exposure. When organisms are exposed to chemical toxicants, the effects of chemical exposure disturb the homeostasis of the organism. As the organismÕs physiologic systems adjusts to compensate for specific effects from the mode of action of the chemical, a number of nonspecific homestatic mechanisms are also in-duced to re-establish equilibrium. This stage maybe associated with a loss of feeding, loss of equilibrium, and behavioral changes (Beyers et al. 1999). Beyers et al. (1999) pointed out that the mechanism for the suppression of feeding is unknown, but it maybe related to physiologic effects of the general adaptation syndrome. Physiologic changes that induce repair mechanisms maydecrease abilityor desire to process food (Health 1995). Pedlar and Klaverkamp (2002) revealed that the impairments of chemoreception mayhave been a mechanism for food refusal.

10 11 12 13 14 15 12 14 16 18 20 0 1 2 3 4 10 12 14 16 18 0 1 2 3 4 B A 11 12 13 14 15 16 4 3 2 0 1 3 4µg mL-1 D C 2µg mL-1 4 2 1 0 0µg mL-1 1µg mL-1 Data

Feeding decreasing model Growth cost model Maintenance cost model

Body weight (g wet wt.)

Time (week)

Fig. 4. Optimal fittings of the feeding-decrease, growth-cost, and maintenance-cost models. Error bar represents mean € 1 SD

Table 3. The SSs of the difference between measured growth data and estimations of the growth-cost, feeding-decrease, and maintenance-cost models As concentrate Growth-cost model Maintenance-cost model Feeding-decrease model 0 lg mL)1 0.057 0.056 0.009 1 lg mL)1 0.887 0.223 0.165 2 lg mL)1 3.122 0.927 0.205 4 lg mL)1 5.069 1.919 0.245

aThe feeding decrease model describes the data better than the other two models.

(8)

The loss of appetite and decrease of growth suggest that the homeostatic mechanism of exposed tilapia are overwhelmed, resulting in damages and activation of repair and homeostatic mechanisms to re-establish equilibrium. The increased main-tenance cost fails to account for the growth decrease, indi-cating that these tilapia have yet to compensate for the As stressor. No tilapia died during the bioassays, showing that the measured growth data derived from the experimental protocol is suitable for chronic mode of action identification.

Metabolic rate is a good measure of energybeing expended for compensation because it integrates all physiologic process. Fits of three effect models revealed that apparent growth de-creases occurred because of decreased feeding. Despite this, the modified West growth model employed in this study is appli-cable to the description and prediction of As toxicity. Never-theless, to better assist accurate risk assessment posed by metals in aquatic ecosystems, more studies and experimental data are needed to validate applications of the proposed models.

Application of DEBtoxand West Growth Model in Ecotoxicology

The West growth model has never been employed in an eco-toxicologic study. Our study provided a novel assessment framework to analyze the mode of action of metal toxicity to aquatic organisms bylinking the West growth model and the DEBtoxtheoryin a bioenergetics-based approach. The DEBtox theorydistinguishes three types of effects on growth, including direct effects and indirect effects bywayof maintenance and assimilation. The inherent assumption is that onlyone of these effects occurs at the same time in the lower effect range of the chemical (Kooijman and Bedaux, 1996). Our bioenergetics-based toxicitymodel well describes the trend of growth in lower concentrations (i.e., 1 lg mL–1), yet the bias between the model description and the measured data increases with the gradient of exposure concentration. We inferred that multiple effects might work together to induce growth toxicityin higher concentra-tions. Our single-mode-of-action–based effect models maynot be reliable in higher concentration (e.g., sublethal exposure

conditions). Sherwood et al. (2000) indicated that the growth inhibition of yellow perch in a heavy-metal (Cd, Cu, and Zn)– polluted lake was attributable to decreased conversion effi-ciencyof the fish and not simplyjust to decreased food intake.

Individual development is fuelled bymetabolism and occurs primarilybycell division. Incoming energyand material from the environment are transformed into metabolic energyand consequentlytransported through hierarchical branching net-work systems for life-sustaining activities and production of new tissue (West et al. 2001). The West growth model de-scribes the universal properties of individual growth based on the first principles of the basic of the conservation of metabolic energy, the allometric scaling of metabolic rate, and the energetic cost of producing and maintaining biomass. The capabilityof this model has been validated for quantitatively predicting growth curves from birth to mature bodysize for all multicellular organisms. This universal growth model provides a basis for understanding the general and fundamental features governing organism growth. Although some criticisms indicate that the conceptual foundation of this model is not applicable to the growth of birds and their life-historyproperties (Ricklefs 2003). West et al. (2004) indicated that this model does not intend to account for all of the observed variation in growth rate and life histories, but it indeed provide a baseline for developing more detailed treatments of ontogenetic growth.

The species-specific growth coefficient (A0) relates the rate of energyallocation to produce a new cell to the rate of the whole organismÕs metabolic rate, which fuels this biosynthesis in terms of normalization (West et al. 2004). Our studyshows that the values of A0 do not change significantlyin different exposure concentrations (Table 2), demonstrating that water-borne As exposures do not disturb the energytranslations between life-sustenance activities and new biomass produc-tion. The growth inhibition byAs exposure is not induced by increasing the energycost to propagate new bodytissues. The concentration-effect tilapia growth trajectories could be well described bydecreasing the values of maximum biomass (Wmax) in the West growth model (Table 2), i.e., the feeding-decrease model. Several studies have shown that in many organisms, from fruit flies to humans, severe restriction of food supplyduring development can prolong time to maturity and result in smaller adult size (Davidovitz et al. 2003; West et al. 2004), which corresponds with the basic description of the feeding-decrease model in the DEBtoxtheory.

In conclusion, the proposed bioenergetics-based growth-ef-fect model allows us to make a comprehensive surveyof growth effects during the entire life cycle of an organism when stressed bychemicals. Different modes of action can have similar effects, but verydifferent consequences, at the indi-vidual level when the data are integrated at the population, community, or ecosystem levels (Barata and Baird 2000). We believe that a mechanistic-based studyto understand the mode of action would improve anyattempt to create predictive models for ecotoxicologic assessment.

References

Balasubramanian PR, Bai RK (1996) Biogas plant-effluent as an or-ganic fertilizer in monosex, monoculture of fish (Oreochromis mossambicus). Bioresource Technol 55:119–124

0 100 200 300 400 500 600 700 800 900 1000 1100 1200 0 500 1000 1500 2000 2500 3000 3500 4000 Time (d) Bo d y we ig h t (g we t wt .) 0µg mL-1 1µg mL-1 2µg mL-1 4µg mL-1

Feeding decrease model

Fig. 5. Predictions of the growth of tilapia O. mossambicus during their entire life span in different waterborne As concentrations using the feeding-decrease model

(9)

Barata C, Baird DJ (2000) Determining the ecotoxicological mode of action of chemicals from measurements made on individuals: Results from instar-based tests with Daphnia magna Straus. Aquat Toxicol 48:195–209

Beyers DW, Rice JA, Clements WH, Henry CJ (1999) Estimating physiological cost of chemical exposure: Integrating energetics and stress to quantifytoxic effects in fish. Can J Fish Aquat Sci 56:814–822

Chen CM, Yu SC, Liu MC (2001) Use of Japanese medaka (Oryzia latipes) and tilapia (Oreochromis mossambicus) in toxicitytests on different industrial effluents in Taiwan. Arch Environ Contam Toxicol 40:363–370

Clason B, Duquesne S, Liess M, Schulz R, Zauke GP (2003) Bioac-cumulation of trace metals in the Antarctic amphipod Paramoera walkeri (Stebbing, 1906): Comparison of two compartment and hyperbolic toxicokinetic models. Aquat Toxicol 65:117–140 Congdon JD, Dunham AE, Hopkins WA, Rowe CL, Hinton TG

(2001) Resource allocation-based life histories: A conceptual basis for studies of ecological toxicology. Environ Toxicol Chem 20:1698–1703

Davidovitz G, DÕAmico LJ, Nijhout HF (2003) Critical weight in the development of insect bodysize. Evol Dev 5:188–197

Escher BI, Hermens JLM (2002) Mode of action in ecotoxicology: Their role in bodyburdens, species sensitivity, QSARS, and mixture effects. Environ Sci Technol 36:4201–4217

Forrester GE, Fredericks BI, Gerdeman D, Evans B, Steele MA, Zayed K, et al. (2003) Growth of estuarine fish is associated with the combined concentration of sediment contaminants and shows no adaptation or acclimation to past conditions. Mar Environ Res 56:423–442

Gomat A (1997) Dose-dependent effects of cadmium on the growth of snails in toxicitybioassays. Arch Environ Contam Toxicol 33:209–216

Health AG (1995) Water pollution and fish physiology, 2nd ed. Lewis, New York, NY

Kooijman SALM, Bedaux JJM (1996) The analysis of aquatic toxicity data. VU Universitypress, Amsterdam, The Netherlands Landis WG, Yu MH (1999) Introduction to environmental toxicology:

Impacts of chemicals upon ecological systems. Lewis, FL Liao CM, Chen BC, Singh S, Lin MC, Liu CW, Han BC (2003) Acute

toxicityand bioaccumulation of arsenic in tilapia (Oreochromis mossambicus) from a blackfoot disease area in Taiwan. Environ Toxicol 18:252–259

Liao CM, Tsai JW, Ling MP, Liang HM, Chou YH, Yang PT (2004) Organ-specific toxicokinetics and dose-response of arsenic in tilapia Oreochromis mossambicus. Arch Environ Contam Toxicol 47:502–510

McCartyLS, MackayD (1993) Enhancing ecotoxicological modeling and assessment. Environ Sci Technol 9:1719–1728

McGeer JC, Brix KV, Skeaff JM, DeForest DK, Brigham SI, Adams WJ, et al. (2003) Inverse relationship between bioconcentration factor and exposure concentration for metals: Implications for hazard assessment of metals in the aquatic environment. Environ Toxicol Chem 22:1017–1037

Pedlar RM, Klaverkamp JF (2002) Accumulation and distribution of dietaryarsenic in lake whitefish (Coregonus clupeaformis). Aquat Toxicol 57:153–166

PeryARR, Ducrot V, Mons R, Garric J (2003) Modelling toxicityand mode of action of chemicals to analyse growth and emergence tests with the midge Chironomus riparius. Aquat Toxicol 65:281–292 Rankin MG, Dixon DG (1994) Acute and chronic toxicityof

water-borne arsenite to rainbow trout (Oncorhynchus mykiss). Can J Fish Aquat Sci 51:372–380

Reinfelder JR, Fisher NS, Luoma SN, Nichols JW, Wang WX (1998) Trace element trophic transfer in aquatic organisms: A critique of the kinetic model approach. Sci Total Environ 219:117–135 Ricklefs RE (2003) Is rate of ontogenetic growth constrained

byre-source supplyor tissue growth potential? A comment on West et al.Õs model. Funct Ecol 17:384–393

Sherwood GD, Rasmussen JB, Rowan DJ, Brodeur J, Hontela A (2000) Bioenergetic costs of heavymetal exposure in yellow perch (Perca flavescens): In situ estimates with a radiotracer (137Cs) technique. Can J Fish Aquat Sci 57:441–450

Suhendrayatna , Ohki A, Nakajima T, Maeda S (2002) Studies on the accumulation and transformation of arsenic in fresh organisms. II. Accumulation and transformation of arsenic compounds by Tilapia mossambica. Chemosphere 46:325–331

Uchida K, Kajimura S, RileyLG, Hirano T, Aida K, Grau EG (2003) Effects of fasting on growth hormone/insulin-like growth factor I axis in the tilapia, Oreochromis mossambicus. Comp Biochem Physiol A 134:429–439

United States Environmental Protection Agency(2000) Technical progress report of the implementation plan for probabilistic eco-logical assessments: Aquatic systems. Meeting scheduled for April 6–7. United States Environmental Protection Agency, Washington, DC

United States Environmental Protection Agency(2002) National recommended water qualitycriteria: 2002. EPA-822-R-02-047. Available at: http://www.epa.gov/ost/pc/revcom.pdf.

Wedemeyer GA, McLeay DJ, Goodyear CP (1984) Assessing the tolerance of fish and fish populations to environmental stress: The problems and methods of monitoring. In: Cairns VW, Hodson PV, Nriagu JO, (eds) Contaminant effects on fisheries. Wiley, New York, NY, pp 163–278

West GB, Brown JH (2004) LifeÕs universal scaling laws. Physics Today57:36–42

West GB, Brown JH, Enquist BJ (2001) A general model for onto-genetic growth. Nature 413:628–631

數據

Fig. 1. (A) Bioassays of tilapia exposed to 1 lg mL )1 waterborne As during a 7-dayuptake
Figure 4 shows the optimal fits of the feeding-decrease, growth-cost, and maintenance-cost models to the experimental data
Fig. 3. Growth coefficients of tilapia versus nominal As concentrations during different exposure periods
Fig. 4. Optimal fittings of the feeding-decrease, growth-cost, and maintenance-cost models
+2

參考文獻

相關文件

Xianggang zaji (miscellaneous notes on Hong Kong) was written by an English and translated into Chinese by a local Chinese literati.. Doubts can therefore be cast as to whether

Now, nearly all of the current flows through wire S since it has a much lower resistance than the light bulb. The light bulb does not glow because the current flowing through it

In order to facilitate the schools using integrated or mixed mode of curriculum organization to adopt the modules of Life and Society (S1-3) for improving their

Using this formalism we derive an exact differential equation for the partition function of two-dimensional gravity as a function of the string coupling constant that governs the

a) Visitor arrivals is growing at a compound annual growth rate. The number of visitors fluctuates from 2012 to 2018 and does not increase in compound growth rate in reality.

y Define  clearly  the  concept  of  economic  growth  and  development  (Economic  growth  can  simply  be  defined  as  a  rise  in  GDP  or  GDP  per 

• To consider the purpose of the task-based approach and the inductive approach in the learning and teaching of grammar at the secondary level.. • To take part in demonstrations

ii.) On main menu, click on Action and go to Action In-Tray. iii.) Inside Action In-Tray, click on Subject Draft ER Request application and go to ER Request detail. You can