Effect of γ-hexachlorocyclohexane (lindane) on carp (Cyprinus carpio)

ECOTOXICOLOCY

AND

ENVIRONMENTALSAFETY

13,339-345

(1987)

Effect of y-Hexachlorocyclohexane (Lindane) on Carp (Cyprinus carpio) I. Effect of Chronic Intoxication on Humoral Immunity in Relation to Tissue Pollutant Levels MURIEL

COSSARINI-DUNIER,* GILLES MONOD,* AND DANIEL LEPOT~

ANDR~E

DEMAEL,~

*Laboratoire d’Ecotoxicologie, INRA. Ecole Nationale V&rinaire de Lyon, B.P. 31, 6!?752 Charbonni&es Cedex, and tLaboratoire de Physiologie G&&-ale et Compake, UniversitC Claude Bernard, Lyon I, 69622 Villeurbanne Cedex, France Received April 3, I986 Lindane, the -y-isomer of hexachlorocyclohexane, one of the most widely used insecticides, was incorporated into carp food for 109 days. The effects of 10, 100, or 1000 ppm of lindane in food on hematocrits and on antibody production against Yersinia ruckeri bacterin in sera and mucus were followed for 79 days. No effect of lindane on these parameters and on spleen weight was observed. Lymphoid organs, spleen and kidney, were as highly contaminated at the beginning of the inoculation (Day 30) as at the end of the experiment. The tissue levels of lindane in liver, spleen, kidney, or whole body, determinated after 30 and 109 days of contamination, indicated that the tissue distribution was dependent on organ lipid contents. o 1987 Academic Press. Inc.

INTRODUCTION Lindane, the y-isomer of hexachlorocyclohexane (r-HCH) is one of the most widely used organochlorine insecticides for agricultural purposes. It is often implicated in acute aquatic environment pollution. Moreover, carps (Cyprinus carpio) are grown in ponds situated near cereal agricultural areas where lindane is commonly used. Chronic toxicity resulting from a long-term exposure is, however, rarely investigated. The consequence of chronic intoxication on the health of populations of freshwater fish, and particularly on their immune systems, could be an increased sensitivity to pathogenic agents from the environment (viruses, parasites, bacteria), leading to the progressive disappearance of fish populations. Few data are available concerning the effects of lindane on immune response of experimenta animals: 60 to 120 ppm injected into a rat inhibited the production of antibodies against human serum albumin (Rosival et al., 1974). Two administrations per OS of 12 ppm lindane to rabbits led to a significant reduction of agglutinating antibodies against Salmonella typhi (Desi, 1976). In fish the studies concerning effects of pesticides on the immune system comprised such aspects as spleen weight, serum protein level, leukocyte counting, but rarely antibody production (Zeeman and Brindley, 198 1). No studies were made on effects of lindane on the humoral response in fish. Our aim was to study the effect of long-term intoxication by lindane on the humoral response of carp against Yersiniu ruckeri bacterin. Hematocrits and the level 339

0147-6513187 Copy&t All n&s

$3.00

0 1987 by Academic Press, Inc. of reproduction in any form reserved.

340

COSSARINI-DUNIER

ET

AL

of antibody mucus were taken. The level of contamination was measured in whole fish and organs (kidney, spleen, liver, muscle) 1 month after the beginning of the contamination and at the end of the experiment (Day 109). MATERIALS

AND

METHODS

Animals. Carp (C. carpio) weighing 60-70 g were purchased from local fish farms (Meximieux, Ain, France). They were maintained in 150-liter tanks with a continuous flow of dechlorinated water, thermoregulated at 20 * 1“C. Fish were divided into four groups of 40 fish and fed daily with commercial carp pellets (Aqualim, Charente, France) at 1% of body weight. Analytical grade lindane (Rhone-Poulenc, Lyon, France) was diluted in acetone and incorporated into pellets at 10, 100, or 1000 mg/kg (ppm). Carps were subjected to a standard photoperiod of 12 hr of daylight. Antigen. After a short anesthesia of 0.3 ml of phenoxyethanol (Eastman Kodak) per liter of water, fish were injected (ip) with commercial Y. ruckeri vaccine (RhoneMerieux IFFA laboratory, Lyon, France), lo9 bacteria per fish in 0.2 ml of saline solution. Sampling procedure. Every 14 days, fish were bled (0.3 ml) from the dorsal aorta. One part blood was collected for hemato- and leukocrits in heparinized tubes which were immediately centrifuged. The remaining blood was allowed to clot for 2 hr at room temperature and then overnight at 4°C. After centrifugation sera were stored at -80°C before use. Fish were weighed four times during the 109 days of the experiment. Thirty days after the beginning of intoxication and on the final day (Day 109) 15 fish per group were sacrificed, 5 of which were weighed, ground in toto with an electric mixer, and then frozen at - 18°C until lindane extraction. Blood, mucus, spleen, kidney, liver, and muscle were collected from 10 other fish. Spleens were individually weighed, pooled, and then stored at - 18°C. Collection of trout mucus. The body surface of an anesthetized trout was rinsed with water. Mucus was gently scraped from the surface of the body, and care was taken not to include blood or other body fluids in the sample. Sterile saline was stirred into the viscous mucus samples (v/v), the mixtures were centrifuged, and the supematants were stored at -80°C until use. Antibodies were titrated in supematants by agglutination. Antibody titration. Anti-Y. ruckeri antibodies were titrated by agglutination in sera and mucus. On round-bottomed polystyrene microplates (Linbro-Titertek, Flow Laboratories, Maclean, VA) 25 ~1 of serum was diluted (v/v) with 0.05 Mphosphatebuffered saline (PBS), pH 7.5 (from 1: 1 to 1:4096), and then 50 ~1 of a suspension of Y. ruckeri (Formalin inactivated), containing about 7.5 X 10’ bacteria/ml (absorbance adjusted to 0.97 at 525 nm), was added and the microplates were shaken. Plates were incubated 2 hr at 37°C and then overnight at 4°C before reading on a mirror. A standard serum was tested for each serial. Variations of the standard never exceeded f one dilution. Lindane analysis. Lindane extraction was performed on about 10 g of whole carp homogenate and liver and 5 g of kidney and spleen. Samples were weighed and ground three times in 40 ml of a hexane/acetone (75/25) mixture using an UltraTurrax homogenizer. Hexane-acetonic phases were filtered on anhydrous sodium sulfate and poured into a distillation flask. The extract was evaporated and the flask was weighed to determine the extractable lipid content of the sample, and then 10 ml

CHRONIC Anti-Yersinia-ruckeri agglutinating (log 2)

LINDANE

TOXICIT\r’

IN CARP. I

341

titer A-41 A-A O-03 z-v

2 4

Control 10 m 100 ppm 1000 ppm

FIG. I. Kinetics of anti-Yrrsinia ruckeri antibodies after injection of lo9 cells on Day 0. Each point represents the mean of 30 to 40 titers per group tested by agglutination.

of hexane was added. Purification was performed on 5 ml of this hexanic extract by adding 2 ml of fuming sulfuric acid (7%). The mixture was vigorously shaken for 2 min and then centrifuged. The hexanic phase was kept at - 18°C until analysis. The gas chromatograph used was a Packard 437 A equipped with a 63Ni electron capture detector and an “on column” injector. A fused silica column (25 m, i.d. 0.22 mm) coated with SE-54 was used with Nz as carrier gas. Oven temperature programming was as follows: 50 to 170°C (30”C/min), 170 to 200°C (2”C/min). Diluting of the purified extracts were carried out in order to obtain lindane concentrations in the range of detector linearity. Lindane recovery was greater than 90%. Statistical analysis. An analysis of variance was done on the areas under the curves of kinetics. Each individual kinetic was integrated in a Wang minicomputer and the F values of Snedecor & Bartlett’s tests were used to determine whether the difference observed between groups was statistically significant. Figures 1 and 2 were drawn by the computer. Student’s t test was used to estimate the splenic index. RESULTS As shown in Table 1 contamination levels were in the same range on Day 30 and Day 109 for each lindane concentration in food. Nevertheless, lindane concentration in whole body and organs increased greatly with increasing contamination level of the food (10 to 1000 ppm). Thus, on Day 30, when the bacterin was injected, lymphoid organs (spleen and kidney) were significantly contaminated. Furthermore, for each group, spleen is less contaminated than kidney and liver if concentrations are expressed on a wet weight basis but not if concentrations are expressed on a lipid weight basis. This led to the conclusion that the tissue distribution of lindane depends significantly on the lipid content of organs because of the lipophilic character of this insecticide.

342

COSSARINI-DUNIER

ET

AL.

Hematocrit k-A’ 30

A-A

i

Control

2

10 wm 100 1000

O-03

F-V

coJ

4

I 0

7

ppm ppm

1

16

30

57

41

O$S

FIG. 2. Kinetics of percentage of hematocrits of 30 to 40 carps per group. of 30 to 40 hematocrits and the vertical bar the standard error.

Each point represents

the mean

Individual kinetics of anti-Y. ruckeri antibodies in sera were compared by statistical analysis. The mean of 30 to 40 kinetics per group for 79 days is presented in Fig. 1. The analysis of variance (F of Snedecor, 1.36; Bartlett’s test, x2 = 0.54) determined the absence of statistically significant differences between groups. Lindane has no immunodepressive effect on humoral response up to 1000 ppm under these experimental conditions. Hematocrits (Fig. 2) and leukocrits (data not shown) were individually taken during the same period. A noticeable variability was observed between groups but remains inferior to the intragroup variability. No conclusion about the

TABLE

1

LINDANE CONCENTRATIONS (ppm) IN ORGANS AND WHOLE BODIES OF CARP EXPOSED TO CONTAMINATED Foot Kidney Lindane in food hwm) lo* (9.4) r

30”

Spleen 109”

0.072 1.39

3.05 0.122

Liver

Whole

30

109

30

109

0.006 0.51

4.55 0.059

0.89 0.034

0.055 1.37

body

30

109

0.070 1.36kO.16 kO.025

0.103 2.76 +f 0.27 0.034

100(113) r lOOO(810) r

Note.

Analyses

18.4 1.12

16.4 0.62

10.6 0.12

14.0 0.20

11.2 0.55

11.5 0.35

1 .oo f-c 0.34 15.6 3.7

0.84 23. I f+ 0.30 8.5

98 5.02

1044.10

450.49

350.47

853.72

712.15

ll.O-c5.6 148k43

4.79 125 +f 27 1.99

of whole

body

were performed

on five fish; analysis

of organs

corresponds

to a pool

of 10 fish. Results

are expressed on a wet weight (ww) and a lipid weight (1~) basis from Day 0 to Day 109; the contamination group was always less than 0.005 ppm (w) and 0.5 ppm (1~) in whole body and organs. a Time of exposure in days. ’ Lindane administered. ’ Lindane found.

of the control

CHRONIC

LINDANE

TOXICITY

IN CARP,

343

I

TABLE 2 SPLENIC

OF CARP 30 AND OR 1000 ppm

INDEX

109 DAYS AFTER CONTAMINATION (mg/kg) OF LINDANE IN FOOD

BY

0, 10,100,

Lindane (ppm) Control

100

10

1000

Day 30

0.50 k 0.16”

0.72 -t 0.19*

0.65 + 0.20

0.63 f 0.22

Day 109

0.52 + 0.1 lb

0.53 * 0.14

0.65 + 0.22**

0.54 f 0.11

Note. Splenic index: a Mean b Mean

f standard k standard

spleen weight % total fish weight

&or, error,

n = IO. n = 20.

-

’

* Significantly increased from control for P < 0.02, Student’s f test. ** Significantly increased from control for P < 0.05, Student’s t test.

toxic effect of lindane on those parameters could be drawn. Evolution of spleen weight was compared between Day 30 and Day 109 (Table 2). According to Student’s t test, statistically significant increases were observed only on Day 30 for 10 ppm and Day 109 for 100 ppm, but no variation of the splenic index was obtained for longterm intoxication with 1000 ppm lindane. The titers of antibody in mucus were weak compared to those in sera (3 to 4 log2 inferior) for the four groups, and no differences were observed between them. The growth of those fish held at 20°C was equivalent during the 109-day period (15% increase). No symptoms of intoxication were observed. DISCUSSION With no results being available on the effects of lindane on the humoral response of fish, we chose Y. ruckeri bacterin as the antigenic model. The O-antigen of this bacteria was used with success as a model by Lamers and Pilarczyk (1982) on carp. This pathogenic agent of the enteric redmouth disease, recorded for the first time in France in 198 1 (Lesel et a/. , 1983), is not known as a pathogenic agent for carp. High titers of antibodies were obtained because fish were maintained in thermoregulated water (20°C); according to Avtalion (1969), carps are not able to synthesize antibodies below 15°C. Comparisons between groups were thus feasible. According to our results, the higher toxic pressures corresponding to each lindane concentration in food were attained at the time of antigen injection (Day 30) because of similar lindane concentrations on Day 30 and Day 109. Because of the lipophilic character of lindane, the insecticide distribution was related to organ lipid content. In spite of a significant level of contamination of lymphoid organs, lindane does not have an immunodepressive effect on either the humoral response or hematocrits of carps. The influence of seasonal fluctuations on these blood parameters is well known in trout (Chilmonczik, personal communication). Moreover, this technique did not seem precise enough for leukocrits and should be replaced by leukocyte counting. The splenic index presented only slight statistically significant differences; no effect of 1000 ppm lindane was observed. Our results are different from those of Springer

344

COSSARINI-DUNIER

ET AL.

et al. (1983) following oral intoxication of the mouse by 5, 10, or 15 ppm lindane for 30 days. They observed negative effects on lymphatic and hemopdietic tissues: reduction of blood hematocrit, decrease of the body weight, but no change in spleen weight. Concentrations of lindane reached in carp under our experimental conditions are much greater than those generally found in fish originating from the natural environment such as tidal Thames, where the concentrations of lindane found were less than 20 pg/kg (w/w) (Rickard and Dulley, 1983). In the roach (Rutilus rutilus) and perch (Perca jluviatilis) from Lahn River (West Germany), Schiiler et al. (1985) found around 1 pg lindane/kg (w/w). In the United States, the last National Pesticide Monitoring Program (1980-198 1) showed lindane concentrations of not more than 10 pg/kg (w/w) (Schmitt et al., 1985). In the French Inventory of Surface Water Quality (198 1-1982) the contamination levels in fish were in the same range except those caught downstream from lindane production factories. In this case concentrations were in the range of 1 mg/kg (w/w) (Monod, unpublished data). Thus, from these data and from the present experiment it would be possible to conclude that there are no immunotoxic effects of lindane under the conditions of natural contamination as far as antibody production is concerned. Nevertheless, it is necessary to take into account several factors. Wild fish are exposed simultaneously to many pollutants that can interact (Monod, 1985). Furthermore, immunotoxic effects of pollutants can be seen under ecological conditions (temperature, food supply, etc.) not optimal for living organisms (Porter et al., 1984). This study was conducted on carps and it is possible that other species are more sensitive to lindane than carp. In fact, Gluth and Hanke (1985) found that carp can survive 4 weeks in water containing 100 pg y-HCH/liter, while the L& (96 hr) for the gudgeon (Go&o gobio) was 75 pg y-HCH/liter (Marcelle and Thome, 1983). In fingerling rainbow trout (Sulmo guirdneri) the LCsO (96 hr) was 30 pugy-HCH/liter (Tooby and Durbin, 1975). In our experiment neither mortality nor symptoms of lindane poisoning were observed in any of the groups of carp. To complete this study of the effect of lindane on the immune response of carp, work is in progress to study the in vivo and in vitro effects on cellular immunity.

ACKNOWLEDGMENTS This work was supported by CNRS Grant ATP PIREN 82 No. 950 170. We are most grateful to RhonePoulenc. Lyon, France, for the gift of lindane and to Rhone-Merieux IFFA Laboratory for furnishing YERSIVAX vaccine. We thank Monique Gaulin for diligent typing.

REFERENCES AVTALION,

R. R. (1969).

Temperature effect on antibody production and immunological memory in Carp against bovine serum albumin (BSA). Immunology 17,927-93 1. DES, I. (1976). Lindane-toxicological studies. In Lindane C.I.E.L. Informations. p. 67. Proceedings of the Symposium on Lindane Lyon-Chazay. GLUTH, G., AND HANKE, W. (1985). A comparison of physiological changes in carp, Cyprinus carpio, induced by several pollutants at sublethal concentrations. I. The dependency on exposure time. Ecotoxicol. Environ. Saf: 9, 179-188. LAMER& M. J., AND PILARCZYK, A. (1982). Immune response and antigen localization in carp (Cyprinus carpio) after administration of Yersinia ruckeri O-antigen. Dev. Camp. Immunol. Suppl. 2, 107-l 14.

(Cyprinus carpio) immunized

CHRONIC

LINDANE

TOXICITY

IN CARP. I

345

LESEL. R., LIESEL,M., GAVINI. F.. AND VUILLAUME. A. (1983). Outbreak of enteric redmouth disease in Rainbow trout, Salmoguirdneri Richardson, in France. J. Fish Dis. 6(4), 385-387. MARCELLE, C., AND THOME, J. P. (1983). Acute toxicity and bioaccumulation of lindane in gudgeon, Gobio gobio (L.). Bull. Environ. Contam. Toxicol. 31,453-458. MONOD, G. (I 985). Egg mortality of Lake Geneva charr (Salvelinus alpinus L.) contaminated by PCB and DDT derivatives. Bull. Environ. Contam. Toxicol. 35,53 l-536. PORTER, W. P., HINSDILL, R.. FAIRBROTHER, A., OLSON, L. J., JAEGER, J., YUILL. T., BISGAARD, S., HUNTER, W. G., AND NOLAN. K. (1984). Toxicant-disease-environment interactions associated with suppressialn of immune system. growth and reproduction. Science 224, 10 14- 10 17. RICKARD, DI. G.. AND DULLEY, M. E. R. (1983). The levels of some heavy metals and chlorinated hydrocarbons in fish from the Tidal Thames. Environ Pollut. Ser. B 5, IO I-I 19. ROSIVAL, L., BARLOGUA, S., AND GRUNT, J. (1974). Effect of lindane on several immunological reactions of rats. Gig. Tr. ProjI Zabol. 6, 53. SCHMITT, C. J., ZAJICEK, J. L.. AND RIBICK, M. A. (1985). National pesticide monitoring program: Residues oforganochlorine chemicals in freshwater fish, 1980-8 1. Arch. Environ. Contam. Taxicol. 14,2X260. SCHOLER, W., BRUNN, H., AND MANZ. D. (1985). Pesticides and polychlorinated biphenyls in fish from the Lahn River. Bull. Environ. Contam. Tosicol. 34,608-6 16. SPRINGER, O., SKRINJAR, L., AND ORGIC. Z. (1983). The effect of pesticide lindane on haemopoietic tissues. Period. Biol. 85(3), 123-125. TOOBY. T. El.. AND DURBIN, F. J. (1975). Lindane residue accumulation and elimination in rainbow trout. (Salmogairdneri Richardson) and roach (Rutilus rutihs L.). Environ. Pollut. Ser. B 79-89. ZEEMAN, M. G., AND BRINDLEY, W. A. (1981). In Immunologic Consideration in To.xico1og.vVII (R. P. Sharma. Ed.), p. i-47. CRC Press, Boca Raton, FL.