COMPARISON
OF M E T H O D S
OF P R E S E R V I N G
PESTICIDE WILLIAM
H. S T I C K E L ,
LUCILLE
TISSUES
FOR
ANALYSIS
F. S T I C K E L ,
RUSSELL
A. D Y R L A N D
Patuxent Wildlife Research Center, U. S. Fish and WtTdlife Service, Laurel, MD 20708. U.S.A. and DONALD
L. H U G H E S
Hazleton-Raltech, Inc., Madison, WI 53707, U.S.A.
(Received May 27, 1983) Abstract. Formalin preservation, freezing, spoiling followed by freezing, and phenoxyethanol were compared in terms of concentrations of DDT, DDD, DDE, endrin, and heptachlor epoxide measured in brain, liver and carcass of birds fed dietary dosages of pesticides and in spiked egg homogenate. Phenoxyethanol proved to be an unsatisfactory preservative; the amount of'extractable lipid' was excessive, and measurements of concentrations in replicates were erratic. Concentrations of residues in formalin-preserved and frozen samples did not differ significantly in any tissue. Percentage lipid in brains and eggs, however, were significantly lower in formalin-preserved samples. Samples of muscle and liver that had been spoiled before freezing yielded less DDD, and muscle samples yielded more DDT than formalin-preserved samples. We conclude that formalin preservation is a satisfactory method for preservation of field samples and that the warming and spoiling of samples that may occur unavoidably in the field will not result in misleading analytical results.
1. Introduction
Freezing remains the customary way of storing samples for residue analysis, but it often is impossible for collectors in the field. Even in the laboratory, freezing creates serious problems of freezer space and presents constant worries about freezer failure and thawing of samples in shipment. Also, it is known that some microbial action continues in frozen samples, and this might cause a gradual degradation of residues. For these reasons, samples from several of our studies were preserved in a 4 percent formaldehyde solution. Many workers have used this technique in the field and several favorable laboratory tests have been reported (Cox, 1970; French et al., 1971; Deubert et al., 1973). The percentage of lipid extracted from brains, however, proved to be consistently lower after formalin preservation than after freezing. The possibilities of 2-phenoxyethanol as described by Nakanishi et al. (1969) also seemed promising. A test therefore was planned to compare four methods of preserving and storing tissues: freezing, formalin, phenoxyethanol, and spoilage followed by freezing (to learn what happens when tissues remain unfrozen for considerable periods). The chemicals selected were: p,p'-DDT, which metabolizes into DDE and DDD; endrin, which is said to degrade rapidly in living animals; and heptachlor, which changes rapidly to heptachlor epoxide in animal tissues. Dosage levels were selected with the goal EnvironmentalMonitoring and Assessment 4 (1984) 113-118. 9 1984 by D. Reidel Publishing Company.
0167-6369/84/0042-0113500.90.
114
W. H, STICKEL ET AL.
of having realistic but readily measurable residues of each chemical. Although these chemicals are no longer used extensively in the United States, they are used in other countries, including wintering areas of birds that breed in the United States. Organochlorine residues continue to occur in environmental samples and the need for monitoring continues (Fleming et al., 1983).
2. Experimental Methods Dietary dosages of heptachlor, dieldrin, and endrin were fed to adult male grackles
(Quiscalus quiscula), resulting in natural incorporation of chemicals into the tissues and producing samples comparable to those from the field. Birds were fed heptachlor and p , p ' - D D T at 5 ppm (dry weight) in the diet for 9 days, June 15-24, 1971. For the last 3 days of this period, endrin was added to the diet at 4.36 ppm. Chemicals were dissolved in Wesson oil and blended into diet of turkey starter crumbles. Birds were sacrificed with chloroform vapor on June 24 and dissected at once. They appeared healthy and were in good flesh and fat. Upon dissection, each brain was bisected longitudinally and each half was randomly assigned to a different preservation method, with the restriction that the two halves be assigned to the same two methods only twice. The same random assignment procedure and restrictions were applied to liver and breast muscle samples, except that these tissues were cut into strips, then into chunks and alternate chunks were assigned to two preservatives. Breast muscle also was trimmed of all visible fat before subdivision. All samples were weighed to 0.1 mg before preservation. Residue readings were based on these weights. Brain samples weighed 1.0 to 1.5 g, liver samples 1.0 to 1.9 g, and muscle samples 3.6 to 6.3 g. The jars used had a capacity of 34 ml and were nearly filled when fluid was used. A 4~o formaldehyde solution was made by diluting chemically pure formalin with distilled water. One part of 2-phenoxyethanol was used with 50 parts of distilled water. The freezer operated at - 17 to - 14 ~ The samples to be spoiled were left open in the laboratory for about 2 hr and were then kept loosely capped in a refrigerator at 7 ~C for 21 days, when mold appeared on tissues;jars then were sealed and frozen. Egg samples, in contrast to tissue samples, were spiked directly, providing a comparison of preservation methods, but very likely resulting in a higher percentage recovery than would occur with naturally incorporated chemicals. Contents of commercial hens' eggs were stirred and put through a 3-ram-mesh sieve to remove the thick parts, chiefly chalazae. Of the fluid egg, 250 g was weighed into a 1000 ml beaker. One ml of hexane containing DDT, endrin, and heptachlor was added to provide spiking at the rate of 4 ppm (wet weight) of each chemical. The mixture was stirred mechanically to assure distribution of chemical. Approximately 11 g of the mixture was poured into each of 12 small, tared jars and weighed to 0.1 mg. Three of the jars were then filled with formalin and three with phenoxyethanol. Three were promptly frozen. Three that were designated for spoiling were left open in the laboratory for about 2 hr and were then placed in a refrigerator at 7 ~C, on August 4, 1971. The bottles were kept loosely capped
COMPARISON OF METHODS OF PRESERVING TISSUES
ll5
and were inspected frequently, but no apparent change appeared until 41 days later. At that time, a brownish darkening was clearly visible, but there was no odor of decay and no macroscopically visible mold. Bottles were then sealed and frozen. All samples were sent to WARF Institute (now Hazleton-Raltech, Inc.) late in 1971 and were analyzed early in 1972. Frozen samples, including those previously spoiled, were kept frozen in shipment and in storage until analysis. Fluid-preserved samples were kept at ambient temperature.
3. Chemical Analytical Methods Frozen samples, including those previously spoiled, were ground with anhydrous sodium sulfate and allowed to air-dry for 48 to 72 hr. They were then placed in 33 x 94 mm Whatman extraction thimbles and extracted by Soxhlet for 8 hr using 70 ml ethyl ether and 170 ml petroleum ether. The fluid was concentrated to 5 to 10 ml by steam bath and made to 50 ml with petroleum ether. Tissues in formalin or phenoxyethanol were treated like the frozen tissues. In addition, the preservative fluid from each sample was shaken three times with the Soxhlet solution. The two extracts for each sample were combined, concentrated to 5 to 10 ml and made to 50 ml as stated above. Egg in formalin or phenoxyethanol was placed in a l-liter separatory funnel, preservative and all, with 100 ml of H 2 0 and shaken for 1 min. Then 100 ml ofisopropyl alcohol was added and the funnel was shaken for another minute. The mixture was extracted with 150 ml of a 50:50 mixture of ethyl ether and petroleum ether. The lower layer (water) was drained into a second separatory funnel and extracted with 50 ml of isopropyl alcohol and 100 ml of a 50 : 50 mixture of ethyl ether and petroleum ether. The layers were allowed to separate, and the lower layer was discarded. The two solvent layers were combined and washed twice with water to remove the alcohol. The extract was transferred to an Erlenmeyer flask, concentrated by steam bath, and made to 50 ml with petroleum ether. For cleanup, an aliquot of the extract was placed on standardized florisil (Pesticide Analytical Manual, Vol. 1, Section 211.15, 1/1/68). A typical elution was 150 ml of 5 ~o ethyl ether in petroleum ether followed by 240 ml of 15 ~o ethyl ether in petroleum ether. Eluates were concentrated by steam bath to 10 to 15 ml and made to 25 ml with hexane. For analysis, 10 gl or less were injected into a Barber Colman Model 500 gas chromatograph. Instrument conditions were: column temperature, 200 ~ injector, 225 ~ detector, 245 ~ column, 1.2 m x 3 mm glass, packed with 5~o DC-200 on Gas Chrom Q 80/100 mesh; nitrogen cartier gas with flow rate of 80 ml rain - 1. Quantification was by peak height in comparison with analytical standards. Limit of detection was 0.05 ppm. Lipids were determined by placing an aliquot of extract in a tared 50 ml beaker and reducing it to dryness on a steam bath followed by 2 to 4 hr in an oven at 40 ~C. The weight of content remaining was taken to be the amount of lipid in the aliquot. The percentage of lipid was calculated on the basis of the fresh wet weight of the sample.
116
w . H . STICKEL ET AL.
All residues are expressed as ppm based on the fresh wet weights taken before preservation. The ppm readings would have been a little higher if based on preserved wet weights, for samples lost weight in storage. Of six muscle samples used for duplicate analyses, one in formalin lost 0.8~o of fresh wet weight, one frozen lost 5.0~o, one spoiled and then frozen lost 8.3~, and three in phenoxyethanol varied from a gain of 0.7~ to a loss of 12.9~/o (average loss, 7.2~o). 4. Results and Discussion
Phenoxyethanol proved to be unsatisfactory as a preservative (Table I): the amount of 'extractable lipid' was excessive; detectable levels of DDT were obtained from only one of six samples of muscle, and quantities of both endrin and heptachlor epoxide were extremely low in comparison with other preservatives; detectable levels of endrin were obtained from only three of six liver samples, and DDT was not obtained from any of the six. The other three methods of preservation differed little from each other in the concentrations of chemicals measured in brain tissues, but the percentage of lipid determined TABLE I C o n c e n t r a t i o n s of pesticides in tissue s a m p l e s p r e s e r v e d by different m e t h o d s a Preservation b
Lipid percent
DDT ppm
DDD ppm
DDE ppm
endrin ppm
HE ppm
Brain tissue Fr Fo SFr Ph
5.0 3.1 5.2 43.0
-+ 0.08 + 0.13 + 0.14 _+ 3.13
b a b c
0.54 0.46 0.57 0.47
a a a a
0.11 0.15 0.13 0.15
+ 0.024 -+ 0.013 + 0.034 + 0.022
a a a a
0.31 0.25 0.28 0.25
+ 0.036 _+ 0.020 + 0.049 _+ 0.028
a a a a
0.31 0.29 0.24 0.24
+ 0.030 + 0.014 _+ 0.019 + 0.032
a a a a
0.83+0.094 0.77 + 0.105 0.67 _+ 0.089 0.77 -+ 0.09~
M u s c l e tissue Fr Fo SFr Ph
1.0 1.1 1.0 11.2
+_ 0.14 + 0.11 + 0.07 _+ 1.18
a a a b
0.54 + 0.085 a 0.33 + 0.043 a 0.85 + 0.047 b -
0.28 0.34 0.18 0.38
-+ 0.021 + 0.060 -+ 0.038 + 0.049
ab b a b
0.20 0.29 0.25 0.25
_+ 0.017 + 0.049 _+ 0.022 + 0.047
a a a a
0.30 0.28 0.29 0.08
+ 0.025 -+ 0.044 + 0.033 _+ 0.023
b b b a
0.83 0.78 0.84 0.49
+ 0.07( + 0.07; _+ 0.07( + 0.02.'
2.2 2.1 2.6 41.0
+ 0.20 +_ 0.15 + 0.20 + 3.19
a a a b
0.96 + 0.184 a 1.09 + 0.184 a 0.67 + 0.137 a -
0.63 0.54 0.92 0.63
+ + + +
0.079 0.040 0.126 0.065
ab a b ab
0.48 0.54 0.48 0.59
+ + + +
0.063 0.007 0.054 0.069
a a a a
0.58 + 0.070 a 0.57 + 0.072 a 0.53 + 0.061 a -
1.91 1.43 1.54 1.09
+ 0.40~ _+ 0.08; + 0.18, + 0.29'
12.5 11.1 t2.4 15.0
+ 0.07 _+ 0.16 + 0.03 _+ 0.16
b a b c
4.30 3.05 3.90 3.08
0.19 0.30 0.24 0.15
+ + + +
0.012 0.073 0.049 0.006
a a a a
0.10 0.06 0.11 0.07
+ + + +
0.003 0.009 0.017 0.007
a a a a
3.42 3.16 3.48 2.99
4.67 + 0.51 3.83 + 0.62 4.08_+0.34 3.68 -+ 0.20
Liver tissue Fr Fo SFr Ph Egg contents Fr Fo SFr Ph
-+ 0.074 + 0.033 -+ 0.062 + 0.072
+ 0.309 + 0.352 _+ 0.338 + 0.066
a a a a
+ 0.292 + 0.190 + 0.209 -+ 0.155
a a a a
C o m p a r i s o n s by analysis of v a r i a n c e and t-tests (P < 0.05). Values followed by the same letter do not differ significa~ from each other. b F r = frozen, - 17 to - 14 ~ F o = 4~o f o r m a l d e h y d e solution; SFr = left open in l a b o r a t o r y for 2 hr, in refriger: at 7 ~ for 21 days, t h e n frozen, - 17 to - 14 ~ Ph = 1 p a r t p h e n o x y e t h a n o l : 50 p a r t s distilled water. a
COMPARISON OF METHODS OF PRESERVING TISSUES
| 17
for samples preserved in formalin, 3.1 ~o, was significantly lower than from samples that were frozen, 5.0~0, or spoiled and then frozen, 5.2~o (Table I). Concentrations of DDE, endrin, and heptachlor epoxide measured in muscle tissue did not differ among the three methods; the concentration of DDT, however, was significantly higher from spoiled-frozen samples than from either frozen or formalinpreserved samples; that of DDD was significantly lower than from formalin-preserved samples (although formalin-preserved and frozen samples did not differ from each other). Percentages of lipid did not differ from each other. Concentrations of DDE, endrin, and heptachlor epoxide measured in liver tissue also did not differ among the three methods. Concentrations of DDD, however, were significantly higher among spoiled samples than among formalin-preserved samples; that of DDT was correspondingly lower, although not significantly so. Percentages of lipid extracted did not differ. As with brain tissue, percentage of lipid extracted from eggs preserved in formalin, 11.1 ~o, was significantly lower than from samples that were frozen, 12.5 ~o, or spoiled and then frozen, 12.4~o. Concentrations of residues, however, did not differ. Because chemicals in egg samples resulted from spiking, traditional recovery data can be calculated from the quantitative data in Table I: recoveries of total DDT (DDT + DDD + DDE) were 115 ~o from frozen samples, 85 ~o from formalin samples, 106~o from spoiled-frozen and 83~o from phenoxyethanol. They were 86, 79, 87, and 75~o for endrin and 117, 96, 102, and 92~o for heptachlor epoxide, listed in the same
TABLE II Variability of concentrations of pesticides in tissue samples preserved by different methods Tissue
Preservation method ~ Fr CV b
Fo CV b
SFr CV u
Ph CV b
Parts per million residue Brain 33.3 Muscle 24.0 Liver 38.5 Egg 12.4
21.0 37.1 26.9 29.4
37.0 30.3 33.8 25.3
32.6 12.8
Percentage lipid Brain Muscle Liver Egg
10.2 22.9 17.1 2.5
6.7 18.8 18.9 0.4
17.8 25.7 19.0 1.9
3.9 34.2 22.5 1.0
a Abbreviations as in Table 1. b Combined average coefficients of variation for DDT, D D D , DDE, endrin, and heptachlor epoxide. CV = (standard deviation/ mean) x 100.
118
w.H. STICKEL ET AL.
order. These values may help to explain a tendency for residues in other tissues (brain, muscle, liver) to be slightly higher in frozen than in formalin-preserved samples. Variability of concentrations did not differ greatly among samples preserved by different methods (Table II). Tests were not made of ethanol or of sodium chloride as preservatives because work by Dr Richard Prouty (personal communication) at Patuxent demonstrated that ethanol is not superior to formalin and that salt yields poor results. Isopropyl alcohol also has proven to be a poor preservative for samples to be analyzed for mirex (Carlson et aL, 1973). Both 4~o formaldehyde solution and Bouin's solution proved satisfactory for DDT and dieldrin analysis in fish samples (Deubert et al., 1973); 3~o formaldehyde solution gave satisfactory results in DDT analysis of marine phytoplankton (Cox, 1970). A 10% solution of phenol provided satisfactory preservation of liver samples for dieldrin and DDT analysis, without proportional changes of DDT and DDD with time; it was believed, however, to possibly produce analytical column difficulties (French et aL, 1971). In the same study, a 4% formaldehyde solution was satisfactory for dieldrin and DDE. DDT, however, decreased with storage time and DDD increased, although the combined total of the two did not change significantly. The authors suggested that a higher percentage of formaldehyde would slow any such change. We conclude that preservation in 4 ~o formaldehyde solution is a satisfactory method for field samples and that the warming and spoiling of samples that may occur unavoidably in the field will not result in misleading analytical results. References Carlson, D. A., Banks, W. A.0 and Wojcik, D. P.: 1973, 'Distortion of Mirex Residues in Insects Owing to Use of Isopropyl Alcohol as a Collection Solvent', Bull. Environ. Contain. Toxicol. 9, 365-369. Cox, J. L.: 1970, 'DDT Residues in Marine Phytoplankton: Increase from 1955 to 1969', Science 170, 71-73. Deubert, K. H., Timmerman, J. S., and McCloskey, L. R.: 1973, 'Effects of Fixation on the Extraction of Dieldrin and p,p'-DDT from Muscle Tissue', Bull. Environ. Contam. Toxicol. 9, 54-56. Fleming, W. J., Clark, D. R., Jr., and Henny, C. J.: 1983, 'Organochlorine Pesticides and PCBs: A Continuing Problem for the 1980's', Trans. N.A. Wildl. Conf. (in press). French, M. C. and Jefferies, D. J.: 1971, 'The Preservation of Biological Tissue for Organochlorine Insecticide Analysis', Bull. Environ. Contain. Toxicol. 6, 460-463. Nakanishi, M., Wilson, A. C.; Nolan, R. A., Gorman, G. C., and Bailey, G. S.: 1969, 'Phenoxyethanol: Protein Preservative for Taxonomists', Science 163, 681-683.
OF M E T H O D S
OF P R E S E R V I N G
PESTICIDE WILLIAM
H. S T I C K E L ,
LUCILLE
TISSUES
FOR
ANALYSIS
F. S T I C K E L ,
RUSSELL
A. D Y R L A N D
Patuxent Wildlife Research Center, U. S. Fish and WtTdlife Service, Laurel, MD 20708. U.S.A. and DONALD
L. H U G H E S
Hazleton-Raltech, Inc., Madison, WI 53707, U.S.A.
(Received May 27, 1983) Abstract. Formalin preservation, freezing, spoiling followed by freezing, and phenoxyethanol were compared in terms of concentrations of DDT, DDD, DDE, endrin, and heptachlor epoxide measured in brain, liver and carcass of birds fed dietary dosages of pesticides and in spiked egg homogenate. Phenoxyethanol proved to be an unsatisfactory preservative; the amount of'extractable lipid' was excessive, and measurements of concentrations in replicates were erratic. Concentrations of residues in formalin-preserved and frozen samples did not differ significantly in any tissue. Percentage lipid in brains and eggs, however, were significantly lower in formalin-preserved samples. Samples of muscle and liver that had been spoiled before freezing yielded less DDD, and muscle samples yielded more DDT than formalin-preserved samples. We conclude that formalin preservation is a satisfactory method for preservation of field samples and that the warming and spoiling of samples that may occur unavoidably in the field will not result in misleading analytical results.
1. Introduction
Freezing remains the customary way of storing samples for residue analysis, but it often is impossible for collectors in the field. Even in the laboratory, freezing creates serious problems of freezer space and presents constant worries about freezer failure and thawing of samples in shipment. Also, it is known that some microbial action continues in frozen samples, and this might cause a gradual degradation of residues. For these reasons, samples from several of our studies were preserved in a 4 percent formaldehyde solution. Many workers have used this technique in the field and several favorable laboratory tests have been reported (Cox, 1970; French et al., 1971; Deubert et al., 1973). The percentage of lipid extracted from brains, however, proved to be consistently lower after formalin preservation than after freezing. The possibilities of 2-phenoxyethanol as described by Nakanishi et al. (1969) also seemed promising. A test therefore was planned to compare four methods of preserving and storing tissues: freezing, formalin, phenoxyethanol, and spoilage followed by freezing (to learn what happens when tissues remain unfrozen for considerable periods). The chemicals selected were: p,p'-DDT, which metabolizes into DDE and DDD; endrin, which is said to degrade rapidly in living animals; and heptachlor, which changes rapidly to heptachlor epoxide in animal tissues. Dosage levels were selected with the goal EnvironmentalMonitoring and Assessment 4 (1984) 113-118. 9 1984 by D. Reidel Publishing Company.
0167-6369/84/0042-0113500.90.
114
W. H, STICKEL ET AL.
of having realistic but readily measurable residues of each chemical. Although these chemicals are no longer used extensively in the United States, they are used in other countries, including wintering areas of birds that breed in the United States. Organochlorine residues continue to occur in environmental samples and the need for monitoring continues (Fleming et al., 1983).
2. Experimental Methods Dietary dosages of heptachlor, dieldrin, and endrin were fed to adult male grackles
(Quiscalus quiscula), resulting in natural incorporation of chemicals into the tissues and producing samples comparable to those from the field. Birds were fed heptachlor and p , p ' - D D T at 5 ppm (dry weight) in the diet for 9 days, June 15-24, 1971. For the last 3 days of this period, endrin was added to the diet at 4.36 ppm. Chemicals were dissolved in Wesson oil and blended into diet of turkey starter crumbles. Birds were sacrificed with chloroform vapor on June 24 and dissected at once. They appeared healthy and were in good flesh and fat. Upon dissection, each brain was bisected longitudinally and each half was randomly assigned to a different preservation method, with the restriction that the two halves be assigned to the same two methods only twice. The same random assignment procedure and restrictions were applied to liver and breast muscle samples, except that these tissues were cut into strips, then into chunks and alternate chunks were assigned to two preservatives. Breast muscle also was trimmed of all visible fat before subdivision. All samples were weighed to 0.1 mg before preservation. Residue readings were based on these weights. Brain samples weighed 1.0 to 1.5 g, liver samples 1.0 to 1.9 g, and muscle samples 3.6 to 6.3 g. The jars used had a capacity of 34 ml and were nearly filled when fluid was used. A 4~o formaldehyde solution was made by diluting chemically pure formalin with distilled water. One part of 2-phenoxyethanol was used with 50 parts of distilled water. The freezer operated at - 17 to - 14 ~ The samples to be spoiled were left open in the laboratory for about 2 hr and were then kept loosely capped in a refrigerator at 7 ~C for 21 days, when mold appeared on tissues;jars then were sealed and frozen. Egg samples, in contrast to tissue samples, were spiked directly, providing a comparison of preservation methods, but very likely resulting in a higher percentage recovery than would occur with naturally incorporated chemicals. Contents of commercial hens' eggs were stirred and put through a 3-ram-mesh sieve to remove the thick parts, chiefly chalazae. Of the fluid egg, 250 g was weighed into a 1000 ml beaker. One ml of hexane containing DDT, endrin, and heptachlor was added to provide spiking at the rate of 4 ppm (wet weight) of each chemical. The mixture was stirred mechanically to assure distribution of chemical. Approximately 11 g of the mixture was poured into each of 12 small, tared jars and weighed to 0.1 mg. Three of the jars were then filled with formalin and three with phenoxyethanol. Three were promptly frozen. Three that were designated for spoiling were left open in the laboratory for about 2 hr and were then placed in a refrigerator at 7 ~C, on August 4, 1971. The bottles were kept loosely capped
COMPARISON OF METHODS OF PRESERVING TISSUES
ll5
and were inspected frequently, but no apparent change appeared until 41 days later. At that time, a brownish darkening was clearly visible, but there was no odor of decay and no macroscopically visible mold. Bottles were then sealed and frozen. All samples were sent to WARF Institute (now Hazleton-Raltech, Inc.) late in 1971 and were analyzed early in 1972. Frozen samples, including those previously spoiled, were kept frozen in shipment and in storage until analysis. Fluid-preserved samples were kept at ambient temperature.
3. Chemical Analytical Methods Frozen samples, including those previously spoiled, were ground with anhydrous sodium sulfate and allowed to air-dry for 48 to 72 hr. They were then placed in 33 x 94 mm Whatman extraction thimbles and extracted by Soxhlet for 8 hr using 70 ml ethyl ether and 170 ml petroleum ether. The fluid was concentrated to 5 to 10 ml by steam bath and made to 50 ml with petroleum ether. Tissues in formalin or phenoxyethanol were treated like the frozen tissues. In addition, the preservative fluid from each sample was shaken three times with the Soxhlet solution. The two extracts for each sample were combined, concentrated to 5 to 10 ml and made to 50 ml as stated above. Egg in formalin or phenoxyethanol was placed in a l-liter separatory funnel, preservative and all, with 100 ml of H 2 0 and shaken for 1 min. Then 100 ml ofisopropyl alcohol was added and the funnel was shaken for another minute. The mixture was extracted with 150 ml of a 50:50 mixture of ethyl ether and petroleum ether. The lower layer (water) was drained into a second separatory funnel and extracted with 50 ml of isopropyl alcohol and 100 ml of a 50 : 50 mixture of ethyl ether and petroleum ether. The layers were allowed to separate, and the lower layer was discarded. The two solvent layers were combined and washed twice with water to remove the alcohol. The extract was transferred to an Erlenmeyer flask, concentrated by steam bath, and made to 50 ml with petroleum ether. For cleanup, an aliquot of the extract was placed on standardized florisil (Pesticide Analytical Manual, Vol. 1, Section 211.15, 1/1/68). A typical elution was 150 ml of 5 ~o ethyl ether in petroleum ether followed by 240 ml of 15 ~o ethyl ether in petroleum ether. Eluates were concentrated by steam bath to 10 to 15 ml and made to 25 ml with hexane. For analysis, 10 gl or less were injected into a Barber Colman Model 500 gas chromatograph. Instrument conditions were: column temperature, 200 ~ injector, 225 ~ detector, 245 ~ column, 1.2 m x 3 mm glass, packed with 5~o DC-200 on Gas Chrom Q 80/100 mesh; nitrogen cartier gas with flow rate of 80 ml rain - 1. Quantification was by peak height in comparison with analytical standards. Limit of detection was 0.05 ppm. Lipids were determined by placing an aliquot of extract in a tared 50 ml beaker and reducing it to dryness on a steam bath followed by 2 to 4 hr in an oven at 40 ~C. The weight of content remaining was taken to be the amount of lipid in the aliquot. The percentage of lipid was calculated on the basis of the fresh wet weight of the sample.
116
w . H . STICKEL ET AL.
All residues are expressed as ppm based on the fresh wet weights taken before preservation. The ppm readings would have been a little higher if based on preserved wet weights, for samples lost weight in storage. Of six muscle samples used for duplicate analyses, one in formalin lost 0.8~o of fresh wet weight, one frozen lost 5.0~o, one spoiled and then frozen lost 8.3~, and three in phenoxyethanol varied from a gain of 0.7~ to a loss of 12.9~/o (average loss, 7.2~o). 4. Results and Discussion
Phenoxyethanol proved to be unsatisfactory as a preservative (Table I): the amount of 'extractable lipid' was excessive; detectable levels of DDT were obtained from only one of six samples of muscle, and quantities of both endrin and heptachlor epoxide were extremely low in comparison with other preservatives; detectable levels of endrin were obtained from only three of six liver samples, and DDT was not obtained from any of the six. The other three methods of preservation differed little from each other in the concentrations of chemicals measured in brain tissues, but the percentage of lipid determined TABLE I C o n c e n t r a t i o n s of pesticides in tissue s a m p l e s p r e s e r v e d by different m e t h o d s a Preservation b
Lipid percent
DDT ppm
DDD ppm
DDE ppm
endrin ppm
HE ppm
Brain tissue Fr Fo SFr Ph
5.0 3.1 5.2 43.0
-+ 0.08 + 0.13 + 0.14 _+ 3.13
b a b c
0.54 0.46 0.57 0.47
a a a a
0.11 0.15 0.13 0.15
+ 0.024 -+ 0.013 + 0.034 + 0.022
a a a a
0.31 0.25 0.28 0.25
+ 0.036 _+ 0.020 + 0.049 _+ 0.028
a a a a
0.31 0.29 0.24 0.24
+ 0.030 + 0.014 _+ 0.019 + 0.032
a a a a
0.83+0.094 0.77 + 0.105 0.67 _+ 0.089 0.77 -+ 0.09~
M u s c l e tissue Fr Fo SFr Ph
1.0 1.1 1.0 11.2
+_ 0.14 + 0.11 + 0.07 _+ 1.18
a a a b
0.54 + 0.085 a 0.33 + 0.043 a 0.85 + 0.047 b -
0.28 0.34 0.18 0.38
-+ 0.021 + 0.060 -+ 0.038 + 0.049
ab b a b
0.20 0.29 0.25 0.25
_+ 0.017 + 0.049 _+ 0.022 + 0.047
a a a a
0.30 0.28 0.29 0.08
+ 0.025 -+ 0.044 + 0.033 _+ 0.023
b b b a
0.83 0.78 0.84 0.49
+ 0.07( + 0.07; _+ 0.07( + 0.02.'
2.2 2.1 2.6 41.0
+ 0.20 +_ 0.15 + 0.20 + 3.19
a a a b
0.96 + 0.184 a 1.09 + 0.184 a 0.67 + 0.137 a -
0.63 0.54 0.92 0.63
+ + + +
0.079 0.040 0.126 0.065
ab a b ab
0.48 0.54 0.48 0.59
+ + + +
0.063 0.007 0.054 0.069
a a a a
0.58 + 0.070 a 0.57 + 0.072 a 0.53 + 0.061 a -
1.91 1.43 1.54 1.09
+ 0.40~ _+ 0.08; + 0.18, + 0.29'
12.5 11.1 t2.4 15.0
+ 0.07 _+ 0.16 + 0.03 _+ 0.16
b a b c
4.30 3.05 3.90 3.08
0.19 0.30 0.24 0.15
+ + + +
0.012 0.073 0.049 0.006
a a a a
0.10 0.06 0.11 0.07
+ + + +
0.003 0.009 0.017 0.007
a a a a
3.42 3.16 3.48 2.99
4.67 + 0.51 3.83 + 0.62 4.08_+0.34 3.68 -+ 0.20
Liver tissue Fr Fo SFr Ph Egg contents Fr Fo SFr Ph
-+ 0.074 + 0.033 -+ 0.062 + 0.072
+ 0.309 + 0.352 _+ 0.338 + 0.066
a a a a
+ 0.292 + 0.190 + 0.209 -+ 0.155
a a a a
C o m p a r i s o n s by analysis of v a r i a n c e and t-tests (P < 0.05). Values followed by the same letter do not differ significa~ from each other. b F r = frozen, - 17 to - 14 ~ F o = 4~o f o r m a l d e h y d e solution; SFr = left open in l a b o r a t o r y for 2 hr, in refriger: at 7 ~ for 21 days, t h e n frozen, - 17 to - 14 ~ Ph = 1 p a r t p h e n o x y e t h a n o l : 50 p a r t s distilled water. a
COMPARISON OF METHODS OF PRESERVING TISSUES
| 17
for samples preserved in formalin, 3.1 ~o, was significantly lower than from samples that were frozen, 5.0~0, or spoiled and then frozen, 5.2~o (Table I). Concentrations of DDE, endrin, and heptachlor epoxide measured in muscle tissue did not differ among the three methods; the concentration of DDT, however, was significantly higher from spoiled-frozen samples than from either frozen or formalinpreserved samples; that of DDD was significantly lower than from formalin-preserved samples (although formalin-preserved and frozen samples did not differ from each other). Percentages of lipid did not differ from each other. Concentrations of DDE, endrin, and heptachlor epoxide measured in liver tissue also did not differ among the three methods. Concentrations of DDD, however, were significantly higher among spoiled samples than among formalin-preserved samples; that of DDT was correspondingly lower, although not significantly so. Percentages of lipid extracted did not differ. As with brain tissue, percentage of lipid extracted from eggs preserved in formalin, 11.1 ~o, was significantly lower than from samples that were frozen, 12.5 ~o, or spoiled and then frozen, 12.4~o. Concentrations of residues, however, did not differ. Because chemicals in egg samples resulted from spiking, traditional recovery data can be calculated from the quantitative data in Table I: recoveries of total DDT (DDT + DDD + DDE) were 115 ~o from frozen samples, 85 ~o from formalin samples, 106~o from spoiled-frozen and 83~o from phenoxyethanol. They were 86, 79, 87, and 75~o for endrin and 117, 96, 102, and 92~o for heptachlor epoxide, listed in the same
TABLE II Variability of concentrations of pesticides in tissue samples preserved by different methods Tissue
Preservation method ~ Fr CV b
Fo CV b
SFr CV u
Ph CV b
Parts per million residue Brain 33.3 Muscle 24.0 Liver 38.5 Egg 12.4
21.0 37.1 26.9 29.4
37.0 30.3 33.8 25.3
32.6 12.8
Percentage lipid Brain Muscle Liver Egg
10.2 22.9 17.1 2.5
6.7 18.8 18.9 0.4
17.8 25.7 19.0 1.9
3.9 34.2 22.5 1.0
a Abbreviations as in Table 1. b Combined average coefficients of variation for DDT, D D D , DDE, endrin, and heptachlor epoxide. CV = (standard deviation/ mean) x 100.
118
w.H. STICKEL ET AL.
order. These values may help to explain a tendency for residues in other tissues (brain, muscle, liver) to be slightly higher in frozen than in formalin-preserved samples. Variability of concentrations did not differ greatly among samples preserved by different methods (Table II). Tests were not made of ethanol or of sodium chloride as preservatives because work by Dr Richard Prouty (personal communication) at Patuxent demonstrated that ethanol is not superior to formalin and that salt yields poor results. Isopropyl alcohol also has proven to be a poor preservative for samples to be analyzed for mirex (Carlson et aL, 1973). Both 4~o formaldehyde solution and Bouin's solution proved satisfactory for DDT and dieldrin analysis in fish samples (Deubert et al., 1973); 3~o formaldehyde solution gave satisfactory results in DDT analysis of marine phytoplankton (Cox, 1970). A 10% solution of phenol provided satisfactory preservation of liver samples for dieldrin and DDT analysis, without proportional changes of DDT and DDD with time; it was believed, however, to possibly produce analytical column difficulties (French et aL, 1971). In the same study, a 4% formaldehyde solution was satisfactory for dieldrin and DDE. DDT, however, decreased with storage time and DDD increased, although the combined total of the two did not change significantly. The authors suggested that a higher percentage of formaldehyde would slow any such change. We conclude that preservation in 4 ~o formaldehyde solution is a satisfactory method for field samples and that the warming and spoiling of samples that may occur unavoidably in the field will not result in misleading analytical results. References Carlson, D. A., Banks, W. A.0 and Wojcik, D. P.: 1973, 'Distortion of Mirex Residues in Insects Owing to Use of Isopropyl Alcohol as a Collection Solvent', Bull. Environ. Contain. Toxicol. 9, 365-369. Cox, J. L.: 1970, 'DDT Residues in Marine Phytoplankton: Increase from 1955 to 1969', Science 170, 71-73. Deubert, K. H., Timmerman, J. S., and McCloskey, L. R.: 1973, 'Effects of Fixation on the Extraction of Dieldrin and p,p'-DDT from Muscle Tissue', Bull. Environ. Contam. Toxicol. 9, 54-56. Fleming, W. J., Clark, D. R., Jr., and Henny, C. J.: 1983, 'Organochlorine Pesticides and PCBs: A Continuing Problem for the 1980's', Trans. N.A. Wildl. Conf. (in press). French, M. C. and Jefferies, D. J.: 1971, 'The Preservation of Biological Tissue for Organochlorine Insecticide Analysis', Bull. Environ. Contain. Toxicol. 6, 460-463. Nakanishi, M., Wilson, A. C.; Nolan, R. A., Gorman, G. C., and Bailey, G. S.: 1969, 'Phenoxyethanol: Protein Preservative for Taxonomists', Science 163, 681-683.
