Evaluation of growth and energy storage as biological markers of DDT exposure in sailfin mollies

ECOTOXICOLOGY

AND ENVIRONMENTAL

SAFETY

29, I- I2 ( 1994)

Evaluation of Growth and Energy Storage as Biological Markers of DDT Exposure in Sailfin Mollies MICHAEL

J. BENTON, **I ALISON

*Environmenral tDepartrnent

Toxicology Research c$ Pharmacology.

C. NIMROD,~

Program, University

Received

AND WILLIAM

H. BENSON*.~

Research Insrirule of Pharmacerrrical ojMississippi. Universiry. Mississippi

Julv

Sciences 386 77

and

13. I993

Direct and indirect measures of growth and energy storage were evaluated as indicators of subchronic I, I, I -trichloro-2-(o-chlorophenyl)-2-( p-chlorophenyl)ethane (o,p’-DDT) exposure in juvenile sailfin mollies (Poecilia lalipinna). Three-day-old mollies were exposed to 0, I, IO. 25, 50, 75, and 100 &liter o,p’-DDT for 2 I days. Tissue residues, percentage weight gain, RNA and DNA content, RNA:DNA ratio. percentage total lipid, percentage triglyceride. percentage total protein, and triglyceride:total lipid ratio were measured following exposure. Mortality was concentration and time dependent, with 100% mortality at 75 and 100 &liter. Among controls and remaining treatments, tissue residues (0.50 to 363 ng/mg dry wt). percentage weight gain (I I6 to 596%). percentage total lipid (2.84 to 4.33%). and percentage triglyceride (I .Ol to 3.22%) were significantly different. Tissue residues were positively correlated with concentration, while percentage weight gain, percentage lipid, percentage triglyceride, and triglyceride:total lipid ratio were negatively correlated with concentration. Direct measures are likely to remain the method of choice for evaluating effects of toxicants on growth in laboratory exposures because of their relative simplicity and reliability. However, indirect measures ofenergy storage, such as triglyceride: total lipid ratio. rather than direct measures of various lipid fractions may be more reliable indicators of the energetic cost of toxicant stress. 0 1994 Academac Press. Inc.

INTRODUCTION Effective environmental monitoring and water quality management requires the development of meaningful biological markers as indicators of toxic exposure and/or predictors of contaminant effects. A great number of biomarkers are now in use or being evaluated, as evidenced by several recent compendia (e.g., Adams, 1990; McCarthy and Shugart, 1990; Huggett et al.. 1992; Peakall, 1992). However, while factors such as metallothionein and heat-shock protein induction may be reliable indicators of exposure to stress, their ecological ramifications are largely unknown. Other measures, such as changes in growth and energy storage, which relate directly to patterns of energy allocation, may be more easily interpreted in terms of ecological consequences. Measures of growth and energy storage may be direct (e.g., weight or body length changes) or indirect. Because DNA content per cell remains relatively constant, while RNA content varies with rate of protein synthesis, the ratio of RNA to DNA may serve as an indirect indicator of recent growth and condition. RNA:DNA ratio in fish was first used as an indicator of growth rate; the ratio decreased when fish were deprived of food and increased when feeding was resumed (Bulow, 1970; Wilder and Stanley, r To whom correspondence should be addressed at current address: Department of Environmental Health, P.O. Box 70682. East Tennessee State University, Johnson City. TN 37614. I

0 147-65 13194 $6.00 Copyright 0 1994 by Academic Press. Inc. All rights of reproduction in any form reserved.

2

BENTON,

NIMROD,

AND

BENSON

1983). There has also been some success in developing RNA:DNA ratio as an indicator of toxicant exposure in fishes (Kearns and Atchison, 1979; Barron and Adelman, 1984). However, the majority of studies with fishes and aquatic invertebrates have shown RNA:DNA ratio to be less sensitive an indicator of growth than more direct measures (Wilder and Stanley, 1983; Cleveland ef al., 1986; McKee and Knowles, 1986; Knowles and McKee, 1987; McKee et al., 1989). Lipids represent a readily available source of energy for fish (Glass et al., 1976; Miller et al., 1975). As many as six lipid classes are present, with phospholipids and triglycerides (triacylglycerols) the predominating forms. Phospholipids serve primarily a structural role in cell and subcellular membranes and serum lipoproteins, while triglycerides are the primary energy storage lipid form (Sheridan, 1983). Because stress responses are energy-draining processes (Shreck, 1990) it is the triglyceride fraction of total lipids that is likely to be reduced under such conditions, although oxidative deterioration of cell membrane lipids may also occur (Thomas, 1990). Under contaminated conditions, therefore, total lipid content, percentage triglyceride, and the ratio of triglycerides to total lipids may be altered. Decreases in total lipid content of fish during exposure to endosulfan, methyl parathion, and crude petroleum have been documented (Murty and Devi, 1982; Dey et al., 1983; Rao and Rao, 1984) although increases in total lipids in response to PCB and DDT exposure have also been reported (Buhler et al.. 1969; Benson et al., 1993a). In addition, storage lipid fractions have been shown to decrease relative to structural lipid fractions in fish and microbe populations occupying contaminated habitats (Adams ei al., 1990; Guckert et al., 1992). Although glycogen and lipids are generally utilized first as energy sources, some organisms also catabolize proteins under conditions of stress. Contaminant-induced protein catabolism has been noted in several invertebrate species (Riley and Mix. 198 I ; Bhagyalakshmi ef al.. 1983; Graney and Giesy, 1986) but apparently has not been documented in fish. In this study, both direct and indirect measures of growth (weight gain. RNA and DNA content, RNA:DNA ratio) and energy storage (percentage lipids. triglycerides, and protein, triglyceride:lipid ratio) were employed to determine the effects of 2 I days exposure of an estuarine, live-bearing fish (sailfin molly) to a chlorinated pesticide. The model contaminant selected was 1, I, I-trichloro-2-(o-chlorophenyl)-2-( p-chlorophenyl)ethane (o,p’-DDT). DDT and its metabolites are known to have a variety of physiological and behavioral effects on fish (Ogilvie and Anderson, 1965: Hickey ef al., 1966; Elson, 1967; Anderson, 1968; Augin and Johansen, 1969; Anderson and Peterson, 1969; Anderson and Prins, 1970) and are a common contaminant of soils and sediments, particularly in areas where this compound has been heavily used as an agricultural pesticide (Cooper et al.. 1987; Wiemeyer et al., 1988; Benson ei al.. 1993b). DDT and its metabolites are also designated as U.S. Environmental Protection Agency priority pollutants (Keith and Telliard, 1979). MATERIALS

AND

METHODS

Model Organism The study organism was the sailfin molly, Poecilia lufipinnu, a live-bearer native to estuarine ecosystems throughout North America and an important forage fish in aquatic food webs. Sailfin mollies originally taken from St. Mark’s National Wildlife Refuge, Wakulla County, Florida, have been cultured in our aquatic facility for approximately

GROWTH

AND

ENERGY

STORAGE

AS

BIOLOGICAL

MARKERS

3

2 years. General culturing conditions are: temperature, 30 + 1°C; salinity, 3 O/00; and photoperiod, 16 hr light:8 hr dark. Breeding tanks were checked twice daily for young, which were removed immediately and placed into 2-liter glass chambers containing aerated 3 O/O0 salinity water. Young were kept in these chambers for 3 days, during which time they were fed finely ground Tetramin Flake Food ad libilurn. After 3 days, young were transferred to exposure vessels. Model Compound For this experiment,

o,p’-DDT was selected as a model contaminant.

E.vpo.wre Procedure Anhydrous o,p’-DDT (99.0% purity; Chem Service, West Chester, PA) was initially dissolved in acetone and stored in injection vials in the following concentrations: 10, 100,250, 500. 750, and IO00 pg/ml. These stock solutions were mixed in 3 O/O0 water to create the following concentrations used as exposure treatments: 1, 10, 25, 50, 75, and 100 pg/liter. Controls (water only) and solvent controls (water plus 100 &liter acetone; hereafter 0 pg/liter o,p’-DDT) were also prepared. Individual exposure vesselsconsisted of 2-liter glass jars containing I liter treatment water from previously prepared Erlenmeyer flasks. Five replicates of each treatment were maintained, and solutions were changed every 24 hr. Water was gently aerated, and vessels were covered with inverted petri dishes. Exposure vessels and Erlenmeyer flasks were kept on a wet table which maintained exposure water temperature at 30°C. Light cycle was 16 hr light:8 hr dark. Mean dissolved oxygen of controls was 6.12 + 0.33 mg 02/liter; mean conductivity of controls was 5.32 + 0.59 pmhos. Initially, I2 fish were weighed and introduced into each vessel. Water samples ( 100 ml) for DDT analysis were taken from all treatments and replicates immediately before (Hour 24) and immediately after (Hour 0) water replacement on Days I, 7. 14. and 2 1, and refrigerated (4°C) until analyzed. Fish were fed once daily. Based on a preliminary study, it was determined that the optimal feeding rate (i.e., growth maximization without excessive water turbidity) for 3 day-old sailfin mollies was 5 mg Tetramin Flake Food/fish/day during the first 7 days of exposure. with an increase of 2 mg/fish/day on Day 7 and again on Day 14. Duration of exposure was 2 I days. On Day 2 I, fish were reweighed and prepared for chemical analyses by first placing them on dry ice, then storing them in an ultracold freezer (-60°C). Analyses consisted of tissue DDT concentration and total lipid, triglyceride, RNA, DNA, and total protein per unit weight. For RNA, DNA, and total protein analyses only tails were used. All other analyses were on whole bodies.

Nucleic acid and ~OINIprotein awl~w’s. The tail of each fish to be analyzed for nucleic acids was removed with a scalpel just caudal to the abdominal cavity. The procedure for nucleic acid analysis was a modified version of Karsten and Wollenberger ( 1972). RNA and DNA were determined by measuring fluorescence of ethidium bromide binding to nucleic acids. Fluorescence was measured with a Perkin-Elmer LS5B Luminescence Spectrometer. Fluorescence values were compared to standard linear curves: RNA (Sigma Chemical Co., St. Louis, MO) and DNA (Sigma Chemical Co.).

BENTON.

NIMROD.

AND

BENSON

Tissue homogenate from nucleic acid analyses was assayed for total protein content using the method of Bradford (1976). Absorbance was measured at 630 nm with a Bio-Tek Instruments EL 3 12e Microplate Bio-Kinetics Reader. Concentrations were calculated by means of a standard curve of bovine serum albumin (Sigma Chemical Co.) in phosphate-buffered saline. Lipid and frigllwride analyses. Frozen whole-body samples were lyophilized overnight, then cut into pieces to facilitate homogenization. The extraction procedure was a modification of Ways and Hanahan (1964) as described by Christie (1987). Lipid content was then determined gravimetrically. Glycerol content of the extracted lipids was determined using the procedure of Bucolo and David ( 1973). Glycerol concentration was expressed as milligrams triolein. the most common triglyceride in fish tissue. Chenvical Analysis Esposwe water and tissue analyses. Briefly, repeated petroleum ether extracts of water samples were exchanged with hexane and concentrated to 1 ml. Quantification of DDT was by gas chromatography as described below. Each tissue sample to be analyzed for o,p’-DDT residues consisted of five whole fish from the same replicate treatment. The sample was first homogenized manually with a Teflon pestle. Repeated petroleum ether extracts of 100 to 400 mg of sample were reduced to 0.5 to I ml with a nitrogen stream. then transferred to a hexane-washed florosil column to remove fats and other biogenic materials. Analytes were eluted with ethyl ether in hexane, and the eluent was concentrated to I ml and analyzed by gas chromatography as described below. Water extracts were analyzed for o,p’-DDT, and tissue extracts were analyzed for o,p’-DDT and metabolites (p,p’-DDT. o-p’-DDD. p,p’-DDD, o,p’-DDE. and p.p’-DDE). All analyses were performed with a Hewlett-Packard (H-P) Model 5890 Series II capillary CC equipped with dual autosamplers and dual H-P standard 63-Ni electron capture detectors and associated electronics. Chromatographic data were collected on an H-P Vectra 386/25 data station using H-P Chemstation software. For tissue samples the residue was reported as ng/mg dry wt (wet wt is 76% water). Samples from the control treatment served as quality assurance/quality control samples for tissue. laboratory, and reagent control. A sample from each treatment was duplicated.

Differences among treatments in tissue residues, percentage weight gain, percentage protein, percentage lipid, percentage triglyceride, triglyceride:lipid. pg RNA/mg dry wt, pug DNA/mg dry wt, and RNA:DNA ratio were tested by ANOVA. StudentNewman-Keuls tests were employed to identify differences in mean responses among treatments. Linearity of the relationships of percentage weight gain. percentage lipid. percentage triglyceride, and triglyceride:lipid values to treatment level was tested using linear regression analysis. Cumulative percentage mortality data were used to estimate median survival times (LT50) and their 95% confidence limits, employing the method of Litchfield (1949).

GROWTH

AND

ENERGY

STORAGE

AS

BIOLOGICAL

5

MARKERS

RESULTS

Exposure to o,p’-DDT resulted in a concentration-dependent increase in cumulative mortality. Over 2 1 days there was no mortality in either the control or the 0 pg DDT/ liter treatment, while mortality for I, IO, 25. 50, 75, and 100 pg/liter was 2, 2, 5, 78, 100, and 100% respectively. LTso values were also concentration dependent: LTsO’s and 95% confidence intervals were 16.5 days (14.5 to 18.8 days), 9.8 days (9. I to 10.5 days), and 8.4 days (7.8 to 9.1 days) for 50, 75, and 100 pg/liter, respectively.

Actual concentrations of o,#-DDT in the control and treatments were representative of nominal values. Only the IO pg/liter treatment deviated by more than 20% from nominal (see Table I). Actual concentrations were used in all statistical analyses and figures, while nominal concentrations were used in tables and text for the sake of clarity. No DDT analyses were performed on fish from 75 or 100 pg/liter treatment due to mortality (see above). Tissue o,p’-DDT residues differed significantly among treatments, with the control and each treatment representing a statistically distinct group (ANOVA, F = 79.47, P < 0.000 I : Student-Newman-Keuls test, P < 0.05) (Table 1). Tissue o,p’-DDT residues were also highly correlated with treatment level (linear regression, 1’ = 0.94, P < 0.0001) (Fig. 1). Tissue residues of p,p’-DDT and two DDT metabolites (p,p’-DDE, o,p’-DDD) showed a concentration-dependent response to actual treatment values (Table I). For two other DDT metabolites, tissue residues were too low (o,p’-DDE) or data were insufficient (p,p’-DDD) to judge their responses (see Table I).

TABLE

I

CONCENTRATIONSOF DDT AND METABOLITES IN EXPOSURE WATER AND SAILFIN MOLLY TISSUE SAMPLES FOLLOWING 2 I-DAY EXPOSURETO o.p’-DDT“ Nominal treatment (&liter

Water (&liter) o.p’-DDT Tissue residue (w/g) o.p’-DDT p.p’-DDT o.p’-DDE p.p’-DDE o.p’-DDD /A/I’-DDD

o.p’-DDT)

Control

0

I

10

0.00

0.00

I .23 ? 0.60

5. IO f 2.08

20.02 + 15.64

45.81

o.oo* 0.00 0.00 0.00 0.00 0.00

O.OOh 15.48 + 7.00’ 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.70 k 0.65 0.00 0.00

f 24.90d I.16 2 0.53 0.00 0.18 f 0.10 5.70 L 2.68 0.28 f 0.07

196.7

+ + + + + +

362.8

75.04

25

3.04 0.02 0.14 14.32 0.56

LIValues are means -t I standard deviation. b-r Values with different superscripts differ significantly: Student-Newman-Keuls

50

44.49"' 1.20 0.04 0.17 6.04 0.53

+

33.31

k 62.32' + 1.95 0.00 1.55 + 0.25 16.90 + 4.20 0.00 5.05

test, P < 0.05.

BENTON.

NIMROD,

AND BENSON

500 F 2 -- 400 Ool 25 5300 -

p<0.0001

0.’ 0

I

I

30

40

50

ACTUAL TREATMENT (ug o,p’-DDT/liter) FIG. I. Regression of tissue residues (rig/g o,p’-DDT) on treatment (rg o,p’-DDT/liter) following 2 I -day exposure.

for sailfin mollies

Growth and Energy Storage Responses Results of the 2 1-day exposure for all measured responses are reflected in Table 2. No response data for 75 or 100 pg/liter were recorded due to 100% mortality in those treatments (see above). ANOVAs revealed significant differences among treatments in percentage weight gain, percentage lipid, percentage triglyceride, and triglyceride:lipid ratio (F = 29.44, TABLE MEASURE~OFGROWTHAND FOLLOWING

2

ENERGY STORAGEFORSAILFIN 2 I -DAY EXPOSURE TO o.p’-DDT”

Nominal treatment(&liter

o.p'-DDT)

Control

0

% Wt. gain

456.6 + 53.3'

595.8 + 174.6b

480.4 k 5S.Zb

529.1 + 94.0b

439.6 + 49.9'

115.9 * 21.5'

~g RNA/w

5.834 2 0.819

5.000 + 0.727

5.007 2 I.091

4.486 + 0.515

4.781 + 0.590

5.100 c I.150

I% DNA/w

0.916 k 0.089

0.934 + 0.180

1.003 2 0.228

0.964 k 0.122

0.879 + 0.104

0.939 + 0.005

RNA:DNA

6.41

+ I.31

5.45

+ 0.85

5.40

2 0.61

4.79

2 0.37

5.41

+ 0.57

5.43

W Protein

6.59

2 0.90

6.51

+ 0.95

7.99

? 2.32

7.39

f I.21

6.48

+ 1.77

5.52

+ 0.08

W Lipid

4.33

k 1.24'.'

3.93

k 0.456,'

5.33

+ 0.506

3.81

+ 0.58"'

4.28

+ 0.48"'

2.84

+ 1.30'

'STriglyceride

2.50

+ 0.84b

2.44

+ 0.47'

3.22

+ 0.32'

2.03

+ 0.45'

2.23

+ 0.47'

I.01

+ 0.53'

Triglyceride:lipid

0.568 + 0.582b

0.617 2 0.060'

I

MOLLIES

0.613 + 0.100"

IO

0.530 + 0.063'

25

0.518 -r- 0.069b

’ Values are means f I standard deviation. b.cValues with different superscripts differ significantly: Student-Newman-Keuls

50

f I.25

0.364 + 0.072'

test, P < 0.05.

GROWTH

AND

ENERGY

STORAGE TABLE

ANALYSIS OF VARIANCE ~I-DAY EXPOSURE

AS

FOR RESULTS TO o,p’-DDT

79.47 29.44 1.17 0.33 1.50 1.12 2.92 6.28 5.39

Tissue residue % Weight Gain” a RNA/mg KS DNA/w RNA: DNA ratio” % Protein % Lipid” % Triglyceride” Triglyceride: Lipid”

OF

P

I;

4

variable

7

MARKERS

3

AMONG TREATMENT OF SAILF~N MOLLIES

Response

’ Dependent

BIOLOGICAL


1 I

0.35 0.89 0.23 0.38 0.0362 0.0009 0.0022

log transformed.

2.92, 6.28, 5.39, respectively: P = 0.000 I, 0.362, 0.0009,0.0022, respectively) (Table 3), but separation of means tests for all these measures revealed that only the 50 pg/ liter treatment differed significantly from other treatments (Student-Newman-Keuls test, P < 0.05) (Tables 2 and 3). However, all these responses were negatively correlated with treatment level (linear regression, ? = 0.84, 0.28, 0.53, 0.52, respectively; P = 0.000 I, 0.0379, 0.0006, 0.0006, respectively) (Figs. 2-5, respectively). There were no significant differences among treatments for RNA content, DNA content, RNA:DNA ratio, or percentage protein (ANOVAs, F = 1.17,0.33, 1.50, 1.12, respectively; P = 0.35, 0.89, 0.23, 0.38, respectively) (Tables 2 and 3). Neither were there significant correlations between any ofthese parameters and actual treatment

0

20

10

ACTUAL FIG. 2. Regression of log-transformed mollies following 2 1-day exposure.

30

TREATMENT percentage

weight

40

50

@g o,p’-DDTlIiter) gain on treatment

(rg o,p’-DDT/liter)

for sailfin

BENTON.

NIMROD,

AND

BENSON

r*=0.28 p=O.O379

I 10

11 0

ACTUAL FIG. 3. Regression of log-transformed following 21-day exposure.

I 20

TREATMENT percent

(linear regressions, 3 = 0.05,0.03, respectively).

/ 30

I 50

(CIg o,p’-DDT/Iiter)

lipid on treatment

0.04,0.12,

I 40

(rg o,p’-DDT/liter)

for sailfin

respectively: P = 0.58,0.76,0.69,

mollies

0.27.

DISCUSSION Effects of DDT exposure on growth were directly measurable by weighing fish immediately before and after exposure. However, changes in indirect measures of

I

I

I

I

1

J

0

10

20

30

40

50

ACTUAL FIG. 4. Regression of log-transformed mollies following 2 l-day exposure.

TREATMENT percent

triglyceride

@g o,p’-DDT/Iiter) on treatment

(pg o.p’-DDT/liter)

for sailfin

GROWTH

AND

ENERGY

STORAGE

AS

BIOLOGICAL

13 3. . -0.65

-

-1.05

t

r2=0.52 p=O.O006

0

10

ACTUAL FIG. mollies

9

MARKERS

5. Regression of log-transformed following 2 I -day exposure.

20

TREATMENT triglyceride:lipid

30

/

I

40

50

@g o,p’-DDThter) ratio

on treatment

(pg o,d-DDT/liter)

for sailfin

growth-percentage RNA and RNA:DNA ratio-were not detected. Research to date has been inconclusive as to the value of RNA:DNA ratio in fish as a biomarker of contaminant stress. Several laboratory exposures of aquatic invertebrates and some salmonids have shown that shifts in RNA concentration are often less sensitive an indicator ofcontaminant stress than are more direct measures (Cleveland, 1986; McKee and Knowles, 1986; Knowles and McKee, 1987; McKee et al., 1989). However, laboratory exposures of larval fish (Barron and Adelman, 1984) and in sift studies of yellow perch (Kearns and Atchison, 1979) salmonids (Wilder and Stanley, 1983) and redbreast sunfish (Adams et al.. 1990) have found RNA:DNA ratios to be negatively related to metal and organic contaminant levels. Direct measures are likely to remain the method of choice for evaluating the effects of toxicants on growth in laboratory exposures. Length and weight are more quickly, more reliably, and less expensively measured than RNA:DNA ratio. However, growth can be difficult to measure in silu, and RNA concentration may be a useful biomarker of growth rate under field conditions (see Mayer et al., 1992). It should be noted that biotic and abiotic factors, such as temperature and nutrition, may also affect RNA concentration (Bulow et al., 1981; Goolish et al.. 1984), and care must be taken when interpreting data from wild populations.

Energy Storage Eflecu In the present study, total proteins were not a good indicator of o,p’-DDT exposure. Only a slight, statistically nonsignificant decrease in percentage protein was observed at the highest treatment level (Table 1). Whole-body proteins are largely structural in nature and do not generally serve as energy reserves (Mayer et al., 1992). Their usefulness as indicators of stress in aquatic invertebrates notwithstanding (Riley and Mix, 198 I ; Bhagyalakshmi ef al.. 1983; Graney and Giesy, 1986) our results support Mayer

10

BENTON,

NIMROD,

AND BENSON

el al.‘s (1992) contention that proteins are catabolized only as a last resort by many organisms and have little utility as a biomarker. Two other direct measures of energy storage-percentage total lipid and percentage triglyceride-were significantly affected by DDT exposure. Both parameters as well as triglyceride:total lipid ratio were significantly correlated with treatment, although the strength of the correlation was decidedly weaker for total lipid (Figs. 3-5). This suggests that while total lipid stores may not respond predictably to varying levels of contamination in sailfin mollies, certain lipid fractions and the relative amounts of stored energy may. These results are consistent with those of Adams et al. ( 1990), who showed that triglyceride:total lipid ratio was markedly decreased in redbreast sunfish occupying PCB- and mercury-contaminated waters. Other studies have also shown that total lipids generally decrease in response to toxicant exposure, although various lipid fractions may not change (Murty and Devi, 1982: Dey et al.. 1983; Rao and Rao, 1984). Because triglycerides represent the primary stored-energy form in fish, reduced percentage triglyceride and triglyceride:total lipid ratios in our study were most likely due to increased metabolism. Anderson (197 1) showed that standard oxygen consumption of Atlantic salmon at 20°C was reduced to approximately 95% of baseline values at 10 pg/liter DDT, and increased to approximately 122 and 140% over baseline at 25 and 50 pg/liter DDT, respectively. However, possible effects of DDT on fish activity, appetite, nutrient assimilation, and lipid production cannot be discounted (see Rice, 1990). Because of the important energetic role of the triglyceride fraction of lipid stores, triglyceride:total lipid ratio is likely to be an important indicator of the energetic cost of acute stress. The number of physiological and behavioral factors that can affect lipid stores, however, means that mechanistic interpretations of this ratio must be carefully considered, especially in wild populations. CONCLUSIONS Because of their relative simplicity and reliability, direct measures (e.g., weight, body length) are likely to remain the method of choice for evaluating the effects of toxicants on growth in fishes. An indirect measure-RNA:DNA ratio-apparently was not affected by contaminant exposure in a predictable manner in many organisms, and so does not appear to be a reliable indicator of recent growth in sailfin mollies. In contrast, an indirect measure of energy storage-triglyceride:lipid ratio-corrects for the statistical variability in measurements of total lipids or lipid fractions alone, and therefore may be a more sensitive indicator of the energetic effects of contaminant exposure. ACKNOWLEDGMENTS The authors thank Leslie Rutherford, James Allgood, Dr. James M. CYNeal. Michelle Malott. and Mildred Ridgway for their able technical assistance.The activities on which this research is based were supported by the U.S. Environmental Protection Agency (R 8 17280-O I-O).

REFERENCES ADAMS, S. M. (1990). Biokogid Indicafors oj.S~ess in Fish. American Fisheries Society. Bethesda. MD. ADAMS, S. M., SHUGART,L. R.. SOUTHWORTH,G. R., AND HINTON, D. E. (I 990). Application ofbioindicators in assessingthe health of fish populations experiencing contaminant stress.In Biomarkers q/Environmmfa/ Confaminafion (J. F. McCarthy and L. R. Shugart. Eds.), pp 333-353. Lewis Publishers, Boca Raton, FL.

GROWTH

AND

ENERGY

STORAGE

AS

BIOLOGICAL

MARKERS

II

ANDERSON, J. M. (1968). ElTect of sublethal DDT on the lateral line of brook trout. Su/ve/inlrs.funrina/is. J. Fish. Res. Bourd Cm 25, 2677-2682. ANDERSON,J. M. (197 I ). Sublethal effectsand changes in ecosystems. Assessment of the effectsof pollutants on physiology and behaviour. Proc. R. Sot. London B Biol. Sci. 1777, 307-320. ANDERSON, J. M.. AND PETERSON,M. R. (1969). DDT: Sublethal effects on nervous system function in brook trout (Suhe/inlrs.fonrinalis). Science 164, 440-44 I. ANDERSON,J. M., AND PRINS, H. B. (1970). Effects of sublethal DDT on a simple reflex in brook trout. J. Fish Res. Board Can. 21, 33 l-334. AUBIN, A. E.. AND JOHANSEN. P. H. (1969). The effects of an acute DDT exposure on the spontaneous electrical activity of goldfish cerebellum. Can. J. Zoo/. 47, 163-166. BARRON, M. G., AND ADELMAN, I. R. (1984). Nucleic acid, protein content and growth of larval fish sublethally exposed to various toxicants. Cult. J. Fish. Aqfru/. Sci. 41, I4 I - 150. BENSON, W. H., DILLON, T., AND SUEDEL, B. (1993a) Evaluation of reproductive effects of anthropogenic chemicals in fish. In Fish Physiology. Takologv. and W’urer Qm/i/~~ Munagemenl: Proceedings of an Internurionul S~wqmsiwn. November 3-5. 1992. Nunjing CJniversiy, Nunjing. Jiungm Province. PRC. EPA. in press. BENSON.W. H., O’NEAL, J. M.. ALLGOOD, J. C.. ELSOHLY, M. A., AND SUMMERS,J. K. (1993b). Evaluation of tissue residues for the Environmental Monitoring and Assessment Program Near Coastal-Louisianian Demonstration. In Proceeding ef the Mississippi M’urer Resowces Cor$erence. April 6-7. 1993. Jackson MS. BHAGYALAKSHMI. A.. REDDY. P. S.. AND RAMAMURTHI. R. (1983). Muscle nitrogen metabolism offreshwater crab, &io(e/~lnrsu sent’s .sene.u,Fabricuis, during acute and chronic sumithion intoxication. Tosicol. Let!. 17. 89-93. BRADFORD,M. M. (1976). A rapid and sensitive method for the quantitation of protein utilizing the principle of protein-dye binding. Am/. Biochern. 72, 248-254. BUCOLO,G.. AND DAVID, H. (1973). Quantitative determination ofserum triglycerides by the use ofenzymes. Clin. Chem. 19, 476-482. BUHLER, D. R.. RASMUSSEN,M. E.. AND SHANKS, W. E. (1969). Chronic oral DDT toxicity in juvenile coho and chinook salmon. Toxicol. Appl. Phummol. 14, 535-555. BULOW, F. J. (1970). RNA-DNA ratios as indicators ofgrowth in fish: A review. In .4ge and Groltlh q/Fish (R. Summerfelt and G. Hall, Eds.), pp 45-64. Iowa State Univ. Press. Ames. IA. BULOW, F. J.. ZEMAN, M. E.. WINNINGHAM, J. R., AND HUDSON, W. F. (198 I). Seasonal variation in RNA-DNA ratios and in indicators of feeding, reproduction, energy storage, and condition in a population of bluegill. Leopomis rmchrochirrrs Ralinesque. J. Fish. Biol. 18, 237-244. CHRISTIE,W. W. ( 1987). High-Pe$:li,rmance Liylrid Chrornulograph~~ and Lipids. Pergamon Press, Oxford, UK. CLEVELAND, L.. LITTLE, E. E.. HAMILTON, S. J., BUCKLER, D. R.. AND HUNN, J. B. (1986). Interactive toxicity of aluminum and acidity to early life stages of brook trout. Trans. .dm. Fish. Sot. 115, 6 10-620. COOPER,C. M., DENDY, F. E.. MCHENRY, J. R., AND RITCHIE.J. C. (1987). Residual pesticide concentrations in Bear Creek, Mississippi. 1976 to 1979. J. Emiron. Qauli/.~~16, 69-72 DEY, A. C.. KICENIUK, J. W.. WILLIAMS, U. P., KHAN, R. A., AND PAYNE, J. F. (1983). Long term exposure of marine fish to crude petroleum. I. Studies on liver lipids and fatty acids in cod (Gadas morhrra) and winter llounder (Psetrdoplewonecves anwricanm). Camp. Biochem. Physiol 75, 93-10 I. ELSON. P. F. (1967). Effects on wild young salmon of spraying DDT over New Brunswick forests. J. Fish. Res. Board Can. 24, 73 I-767. GLASS, R. L.. KRICK, T. P., SAND, D. M., RAHN, C. H.. AND SCHLENK. H. (1975). Ruranoid fatty acids from fish lipids. Lipids 10, 695-702. GOOLISH, E. M.. BARRON, M. G.. AND ADELMAN. I. R. (1984). Thermoacclimatory response of nucleic acids and protein content of carp muscle tissue: Influence ofgrowth rate and relationship to glycine uptake by scales. Cm. J. Zool. 62, 2 164-2 170. GRANEY, R. L.. AND GIESY, J. P. (1986). Effects of long-term exposure to pentachlorophenol on the free amino acid pool and energy reserves of the freshwater amphipod Gummaras pseadolinmaetrs Bouslield (Crustacea, Amphipod). Ecoto.vicol. Environ. Su/: 12. 233-25 I. GUCKERT, J. B.. NOLD, S. C., BOSTON,H. L., AND WHITE, D. C. (I 992). Periphyton response in an industrial receiving stream: Lipid-based physiological stressanalysis and pattern recognition of microbial community structure. Cm. J. Fish. rl91ru1. Sci. 49. 2579-2587.

12

BENTON.

NIMROD.

AND

BENSON

HICKEY, J. J., KEITH, J. A., AND COON, F. B. (1966). An exploration of pesticides in a Lake Michigan ecosystem. J. Appl. Ecol. 3, I4 I - 154. HUGGETT, R. J.. K~MERLE, R. A.. MEHRLE, P. M.. AND BERGMAN, H. L. (I 992). Bio,narkcrs: Biochemica/, Physiological. and Hisrological Markers of Anthropogenic Svess. Lewis Publishers. Boca Raton. FL. KARSTEN, U., AND WOLLENBERGER, A. (1972). Determination of DNA and RNA in homogenized cells and tissues by surface tluorometry. Anal. Biochan. 77, 463-470. KEARNS, P. K., AND ATCHISON, G. J. (1979). Effects of trace metals on growth of yellow perch (Perca j7avescens) as measured by RNA-DNA ratios. Environ. Biol. Fishes 4, 383-387. KNOWLES, C. 0.. AND MCKEE, M. J. (1987). Protein and nucleic acid content in Daphnia magna during chronic exposure to cadmium. Ecorosicol. Environ. Sal: 13, 290-300. LICHTF~ELD,J. T., JR. ( 1949). A method for rapid graphic solution of time-percent effect curves. J. Phamacol. Exp.

Ther.

94, 399-408.

MAYER, F. L., VERSTEEG, D. J., MCKEE, M. J., FOLMAR, L. C., GRANEY. R. L.. MCCUME, D. C., AND RATTNER, B. A. (I 992). Physiological and nonspecific biomarkers. In Biomarkers: Biochemical, Physiological, and Hislological Markers qfilnrhropogenic Srress (R. J. Huggett, R. A. Kimerle. P. M. Mehrle. Jr., and H. L. Bergman, Eds.), pp. S-85. Lewis Publishers. Boca Raton. FL. MCCARTHY, J. F.. AND SHUGART, L. R. ( 1990). Biomarkers qfEn~~ironmen/al Confamination. Lewis Publishers, Boca Raton, FL. MCKEE, M. J.. AND KNOWLES, C. 0. (1986). Effects of fenvalerate on biochemical parameters, survival, and reproduction of Daphnia magna. Ecalosicol. Environ. SaJ: 12, 70-84. MCKEE, M. J.. KNOWLES, C. 0.. AND BUCKLER, D. R. (1989). Effects of aluminum on the biochemical composition of Atlantic salmon. .4rch. Environ. Conlam. To.uicol. 18, 243-248. MILLER, N. G. A., HILL, M. W.. AND SMITH, M. W. (I 976). Positional and species analyses of membrane phospholipids extracted from goldfish adapted to different environmental temperatures. Biochem. Bioph!s. Acra

455,

644-654.

MURTY, A. S., AND DEVI, A. P. (1982). The effect of endosulfan and its isomers in tissue protein. glycogen. and lipids in the fish (Channa puncratus). Peslic. Biochem. Ph!:Gol. 17, 280-286. OCILVIE, D. M., AND ANDERSON, J. M. (1965). Effect of DDT on temperature selection by young Atlantic salmon, Salmo sob-. J. Fish. Res. Board Can. 22, 503-5 12. PEAKALL, D. (1992). Animal Biomarkers as Pollution 1ndicaror.s. Chapman & Hall, London. RAO, K. S. P.. AND RAO, K. V. R. ( 1984). Changes in the tissue lipid profiles of fish (Oreochromis mossambicxs) during methyl parathion toxicity-A time course study. Tosirol. Lerr. 21, 147-I 53. RICE, J. A. (1990). Bioenergetics modeling approaches to evaluation ofstress in fishes..4rtr. Fish. Sot S!v~p. 8, 80-92. RILEY, R. T.. AND MIX, M. C. (198 I). The effects of naphthalene on glucose metabolism in the European flat oyster Osrrea edulis. Comp. B&hem. Ph]‘siol. C 70, 13-20. SHERIDAN, M. A. (1989). Alterations in lipid metabolism accompanying smoltification and seawater adaptation of salmonid fish. Aquacuhlrre 82, I9 l-203. SHREW C. B. ( 1990). Ph.wiological, Behanioral. and Pe$ormance Indicarors qf.S/ress Am. Fish. Sot. .S!,mp. 8, 29-37. SMITH, R. K.. AND COLE, C. F. (1973). Effects of egg concentrations of DDT and dieldrin on development in winter flounder (Psnrdoplerrronec!es americanus). J. Fish. Res. Board Can. 30, 1894. THOMAS, P. (1990). Molecular and biochemical responses of fish to stressors and their potential use in environmental monitoring. Am. Fish. Sot. Symp. 8, 9-28. WALKER, W. W., MALTOX, W. G., AND RISEBROUGH,R. W. (1973). Pollutant and shell thickness determinations of peregrine eggs from West Greenland. ,4rric 26, 255-256. WAYS, P., AND HANAHAN, D. J. (1964). Characterization and quatification of red cell lipids in normal man. J. Lipid Res. 5, 3 18-328. WIEMEYER, S. M., BUNCK, C. M., AND KRYNITSKY, A. J. (1988). Organochlorine pesticides. plychlorinated biphenyls, and mercury in Osprey eggs( 1970-79) and their relationship to shell thinning and productivity. Arch.

Environ.

Cotuam.

Tosicol.

17, 767-787.

WILDER, I. B., AND STANLEY, J. G. (1983). RNA-DNA ratios as an index to growth in salmonid fishes in the laboratory and in streams contaminated by carbaryl. J. Fish. Biol. 22, I65- 172.