Document RaEedrEwwnZ2XpKY1mQL64oez

PREDICTING THE. 8I0C01JCENTRATX0N POTENTIAL OP ORGANIC CHEMICALS , IN PISH : .. by ' .. . W. Brock Neely*, D. R. Branaon** and G. E. Blau* Tha Dow Chemical Company, Midland, Mich. *Ag-Organici Product Department **Haate Control Laboratory ***Computatione Reaearch Laboratory MONS 039633 ABSTRACT A regression equation was developed`to predict the bioconcen- -tration potential of chemicals in fish with the partition coefficient as the independent variable, using chemicals with a wide range in partition coefficient (carbon tetra chloride to hexachlorobenzene). The regression equation was then used to predict the bioconcentration factor of - Chlorpyrifos and Endrin in fish. The predicted values agreed 'with the experimental values reported in the.literature. The equation may be used to estimate the potential of new chemical to bioconcentrate in fish. INTRODUCTION The recent concern (1-3) over the wide spread residue levels in our environment of DDT and PCB (polychlorinated biphenyls) has raised questions regarding the distribution of other chlorinated hydrocarbons. There has been and still is a tendency in both the lay press and in some scientific- circles to generalize that other carbon-chlorine containing materials will have environmental characteristics similar to DDT. It is the purpose of this article to relate biocor.centration to the chemical and physical properties and to suggest a method for estimating the potential of a chemical to concen trate in fish. Knowledge of this potential is useful in making the business decision regarding the possible develop ment of new chemicals. . MQNs 039634 'i' i: . 2- ~ ' . '. .. ' . Xenega' pointed out the difficulties of defining the general term of bioconcentration. It is our belief that by confining our attention to fish in an aquatic environment we can circumvent these difficulties. Hence we shall define bioconcentration . factor as the ratio of the concentration of the chemical in ' . the fish at steady state with the concentration of the chemical in the exposure water under laboratory conditions5. When dealing with bioconcentration under these conditions and : with relatively stable molecules the equilibrium situation of the chemical between the fish and water should be related ' to a simple partitioning. In other words, the fish to a first approximation may be considered as a non polar solvent mixed with the polar solvent, water. If this is true, then the partition coefficient of the chemical should be correlated in a linear manner with the ratio of the steady state concen tration of the chemical in the fish and water. . ' This is the thesis that we will attempt to support by the experimental evidence to be presented. ,_ . HATSnihLS AWO METHODS ' ' '. '* ' Chemicals - The following chemicals were evaluated: 1) 1,1,2,2tetrachloro ethylene, 2) hexachlorobenzene, 3) 2, 21 ,4 ,<l '-tetrachlorobiphenyl, 4) 2-biphenylyl phenyl ether, 5) diphenyl ether, 6) carbon tetrachloride, and 7) p-dichlorobenzene. HONS 039635 Bioconcentration factor in fish - The method described by Branson et al.* was used to determine the bioconcentration factor in rainbow trout (Salmo gairdnsri Richardson). This method is based on determining the steady state concentration of the chemical in the muscle of 4-5" trout. Essentially there are three parts to the procedure; a) the rate of uptake of the chemical by fish is measured, b) the remaining fish are transferred to fresh water and the clearance rate is deter mined, c) a computer based technique is used to calculate the bioconcentration at steady state. | . Partition coefficient - The partition coefficient Of the' Chemical between n-octanol and water was either taken from the tabulation of Hansch et al.4 or calculated using the additivity principles7 as illustrated below. A partition coefficient between n-octanol and water for a chemical may be calculated using equation 1. .. log 1 substituent constants + lDg PH ' . il) where Px * Partition coefficient of chemical PH = Partition coefficient of parent structure the substituent constants were obtained from the listing of Hansch4. The various substituent constants used in this paper ere, shown in Table 1. The following is a typical example using 1,1,2,2-tetrachloro ethylene as the chemical. HONS 039636 4 log P^ +{4 x. .39) (four aliphatic Cl atoms) -(1 x .3) (douole bond) + (2 x .S) (2, -CH'~ groups) * 2.26 (Sum of substituent constants) RESULTS The results of measuring the bioconcentration factor in rainbow . trout end the partition coefficient of the various chemicals are shown in Table II. Using standard regression analysis' the line of best fit was drawn through the points of partition ' coefficient vs. bioconcentration factor and is shown in Figure i with the equation for the line being given in Equation 2. ' log (Biocone, factor) 0,56 log (Fart, coaff.) + 0.124 (2) It multiple coxreletion coefficient of 0.951 end e standard error of 0.369 was obtained for the regression. The 959 confidence level for these values are shown in Figure 1. The lines were calculated from the statistical equation given in Dreper a.smith*. .. ' * DISCUSSIOU ," ' Ac con be seen from Fig. 1 there is a high correlation between the partition coefficient of the chemical and its propensity mqns 039037 5 to bioconcentrate in fish. It ..would appear, in these examples, that chlorine is no more significant than any other group in making a contribution to lipophilicity. As more lipophilic groups are added the tendency to bioconcentrate is enhanced. The explanation for 2-biphenylyl phenyl ether falling outside the 95% confidence region in Figure 1 may be due to metabolism with the subsequent formation a more polar compound. Such a situation would yield a structure(s) with a partition coefficient lower than the parent compound. . ._ In using equation 2 for predicting the bioconcentration factor ih fish for a specific chemical it must be remembered that the confidence intervals about the regression line have to be expanded*. Only as the number of observations for a chemical are increased will the confidence level be as pictured in . Figure 1. .. With this in mind, it is instructive to see how good a predictor equation 2 is for determining the bioconcentration factor in fishi. . The data presented in Table III was taken from unpublished work at The Dow Chemical Company15-1 * as well as from the literature'7. In both cases the experimental procedure was slightly different and the fish species was mosquito fish (Gambusia affinia). It is rather striking, in view of these differences that such a close agreement between the experimental and calculated bioconcontration factor was observed. HUNS 039638 6- In interpreting tile bioconcentration factor for chlorpyrifos or any chemical it must be remembered that metabolism is an active process'*. Consequently the absolute amount of insecticide is constantly being reduced while the ratio will remain relatively fixed. The amount of fat deposits in the fish will also be an important factor in dictating the absolute amount of chemical found. ' ,,. .. It will be noted that as you move away from the middle of the regression line (Figure 1) the confidence ievel increases. This is illustrated with the large standard deviation that is associated with the calculated value for pyridinol in Table XXX. Consequently, less confidence must be placed on any predictions that fall outside the range of the line in Figure 2. Since the experimental bioconcentration factor value for DDT of 5.23" was outside the region established by this regression, no attempt was made to make a prediction. ' As more data is generated it will be important to verify and extend the confidence regions of this type of correlation. However, the present study does allow an investigator to begin rating the potential of new materials to concentrate. By matching this potential with the intended end use an early judgnent decision can be made as to the possible long term envir-'.-iTiental problems that may be faced. HONS 039639 ,. .. -7- . Finally,, it should be stressed" that in assessing the potential environmental hazard created by the presence of a foreign chemical the following considerations in addition to the par tition coefficient are important. . 1. Biological reactivity. If a molecule has a high metabolic rate the chances of accumulation are decreased. This is especially so since materials are usually converted to more polar chemicals and hence the affinity for lipids is decreased. ' '. . 2. How used. If the chemical is used in an open ended system i.e. has direct access to the environment the chances for having an environmental stress are improved. 3. Volume. If the annual production of the chemical is high (i.e. DDT production v;as in excess of 60 million lbs.) and all the other factors sucH as high partition coefficient, unreactive and wide distribution are present an environ . mental hazard will probably be present. MONS 039640 REFERENCED 1. Burnett, R., "DDT Residues: Distribution of Concentrations in Emerita analogs (Stimpson) along Coastal California." ` Science, 174, 606 (1971). . ' 2. "Persistent Pesticides and PCBs in the Environment.* Hews and Views, Nature, 240,' 219 (1972). 3. Gustafson, C. G., "PCBs - prevalent and persistent." Environmental Science and Technology, 4, 814 (1970). 4. Xenaga, E. E., "Guidelines for environmental study of pesticides: Determination of bioconcentration' potential." Residue Reviews, 44, 73 (1972). 5. Branson, D. R., Blau, G. E., Alexander, H. C., Thielen, D. R., ' and Neely, W. B., "A Bioconcentration test: Steady-State Concentrations of 2,21,4,41-tetrachlorobiphenyl in Trout." ' Submitted to Water Poll. Control Fed. 6. Leo, A., Hansch, C., and Elkins, D., "Partition Coefficients and Their Uses." Chem. Rev., 71, 525 (1971). . 7. Hansch, C., Leo, A., and Hikaitani, D., "On the Additivity - Constitutive Character of Partition Coefficients." J. Org. Chen., 37, 3090 (1972). 8. Draper, N. R., and Smith, H., "Applied Regression Analysis." John Viiley and Sons, Inc., New York, N. Y., (1966). 9. Brosier, J. S., "Bioconcentration of 1,1,2,2-tetrachlcroethyler . in Rainbow Trout." Unpublished data. The Dow Chemical Company 10. ' Brosier, J. S., Alexander, H. C., Thielen, D. R., and Elau, G. "Determination of the 3iocor,cantration Factor of Carbon Tetrachloride in Rainbow Trout, Salr.o qairdneri Richardson." Unpublished data, The Dow Chemical Company. 11. Brosier J. S. , "Bioconcentration of p-dichlorobonsene in' Rairbov Trout. " Unpublished data, The Dov/ Chemical Company 12. Branson, D. P.. , Litchfield, M., and Alexander, H. C., "Biocon-entrat ion of Diphenyl oxide in.Trout." Unpublished data, The Dow Chemical Company. 13. Branso:., D. Alexander, !!. C., and Litchfield, N. , "Bioconccntration of 2-biphenylyl phenyl ether in Trout. Unpublished data, The Uow Chemical Company. HONS 039641 14. Brunson, 0. Tc., Alexander1,--(I. C., and Litchfield, K., "Biocor.ccntration of Ilexachlorobcnzane in Trout. " Unpublished data. The Dov; Cherdcal Company. 15. Hedlund, R. T., "Bioconccntration of Chlorpyrifos by Mosquito Fish in a Flowing System." Unpublished data. The Dow Chemical Company. 16. Hedlund, R. T,, "Determination of the Bioconcentration Potential o 3,5,6-Trichloro-2-Pyridinol.* Unpublished data. The Dow Chemical Company. 17. 18. ` . Ferguson, D. E., Ludke, J. L., and Murphy, G. G., "Dynamics of Endrin Uptake and Release by Resistant and Susceptible Strains of Mosouito Fish." Trans. Am. Fisheries Soc., 95, 335 (1966).* 'j . ' 1 Smith, G. N., Watson, B. Si, and Fisher, F. S., "The Metabolism of .['"C] 0,0-Diethyl 0-(3,5,6-trichloro-2- pyridyl) phosphorothioate in Fish." J. Econ. Entomol., - 59, 1464 (1966). .` . 1.9. Hamelink, J. L., VJaybrant, R. C., and Ball, R. C., "A Proposal: Exchange Equilibria Control the Degree Chlorinated Hydrocarbons are Biologically Magnified in Lentic Environments Trans. Amer. Fisheries Soc., 100, 207 (1971). - 20. Box, G. E. P., Tiao, G. C., "Bayesian Inference in Statistical Analysis." Addison-Wesley, Hew York, SKY. (1973). HUNS 039642 Relationship Between Partition Coefficient & . .. The Potential Of Chemicals To Accumulate In Trout Muscle HONS 0 3 9 6 4 3  . ' :. . : TABLE I . . List of Substituent Constants Used for Calculating Partition Coefficients (see Ref. 6 for complete listing.) Group . 6ubstitutent Constant Aromatic - Cl oh benzene Cl. on phenol ' . o ' '' ' P Benzene Phenol 0.7 0.73 1.04 0.98 2.13 1.46 Aliohatic Cl . double bond -CH2CHC1, 0.39 -0.30 0.50 1.97 HONS 039644  *** TABLE II Bioconcentration factor in fish and the partition coefficients of the chemicals studied. .\ . Chemical Partition Coeff.1 Bioconc. Factor1 Ref ^ 1. 1,1,2,2-tetrachlor ethylene 2.26** 2. CCl' 2.36* 3. p-dichlorobenzene 3.38* 4. diphenyl ether 4.20* 5. 2-biphenylyl phenyl ether 5.66* 6. hexachlorobenzene 6.22* 7. 2,2*,4,4'-tetrachloro diphenyl ether . 6.72* 1.59(1.4-1.74)' I.APL/ IC -/.3o\ *.57(1.-43 1.0 57' 2.36(2.32-2.39) 2.30(2.23-2.34) 9 10 11 12 2.74(2.64-2.81) 3.93(3.80-4.07) 13 14 4.09(4.00-4.16) 5. *A11 values are logarithms to the base 10. . .' Calculated value. ; 'Experimental value (Ref. 6). " . 'Figures in parenthesis are the nonsymnetrical 95% .confidence , limits calculated by a Bayesian estimation procedure (Ref. 20) HONS 039645 iauLE in The use of regression equation 2 for predicting the biocon- centrstion factor. .. Chemical log (Part, coeff.) log (Bioconc. factor) Endrin Chlorpyrifos* 3,5,6-trichloro . pyridinol '' 5.6b 4.82 '' . Calculated Experimental 3.47 + .9B9d ' 3.17*/ 2.87 + 0.963d | 2.67* ^ 1.35b . ' . . ' 0.88 + Hl39a 0.49" a0,0-diethyl 0-(3,5,6-trichloro-2-pyridyl) phosphorothioate ^Calculated . :. ' cxperimental dStanderd deviation calculated from Draper & Smith (Ref. 20}. MONS 039646