Journal of Analytical Toxicology, Vol.
19, July/August 1995
Toxicologyof 2,4-DichlorophenoxyaceticAcid (2,4-D)
and Its Determination in Serum and Brain TissueUsing
Gas Chromatography-Electron-CaptureDetection
G.H. Oliveira 1 and J. Palermo-Neto 2
1Departmentof Natural Active Principles, School of PharmaceuticalSciencesUNESP,Araraquara, SP,and 2Applied Pharma-
cology and ToxicologyLaboratory,Schoolof VeterinaryMedicine, University of S~o Paulo, Av. Corifeu de AzevedoMarques,
2720, CEP05340-901, S~o Paulo, SP, Brazil
[ Abstract ]
A gas-liquid chromatographic method with an electron-capture
detector was applied for 2,4-dichlorophenoxyacetic acid (2,4-D)
determination in the serum and brain tissue of rats acutely
intoxicated with the dimethylamine salt of 2,4-D. After extraction
with ethyl ether, 2,4-D derivatization was performed using 2-
chloroethanol and BCI3. The average recovery values found for
serum and brain tissue were 98.5 • 4.8 and 93.3 • 7.5,
respectively. The sensitivity limit of the method was 250 ng/mL for
serum and 300 ng/g for brain tissue. The toxic effects of 2,4-D in
rats were observed within one-half hour after its oral
administration. Results suggest that the toxic mechanism of 2,4-D
is related to an action on the central nervous system.
Introduction
The widespreaduse of pesticides in agriculture in an attempt
to increase crop production has been generating a series of
toxicological and environmental problems. The use of herbi-
cides has increased in the last few years (1), replacing both
mechanical weeding in agriculture and weed control along
railroad borders (2,3). Chlorinated phenoxy acid derivatives
(esters and amines) are among the most common and most
widely used herbicides for these purposes. 2,4-Dichloro-
phenoxyaceticacid (2,4-D) is one of them and has been used to
defoliatejungle areas in South Vietnam as a component of the
"Orange Agent" (2,4). In Brazil, a lethal accident with this
herbicide involving both animals and people in Tucurui,
Amazon, was reported (5).
Despite these occurrences, only a limited number of studies
have been specificallydesigned to measure 2,4-D absorption
and distribution inside humans and other animals after non-
lethal poisoning. Some of these studies used radioactive-labeled
carbon (6), which makes analytical studies difficult for many
laboratories. The objective of the present study is to describe a
method to quantitate 2,4-D in the serum of rats acutely intox-
icated with nonlethal oral doses of 2,4-D. Because some signs
Reproduction(photocopying)of editorialcontentof this journal is prohibitedwithout publisher'spermission.
and symptoms of 2,4-D intoxication are related to behavioral or
motor dysfunctions or both (7,8,9), the presence of the herbi-
cide within the central nervous system (CNS)was also studied.
Materials and Methods
Chemicals
Standards for 2,4-D and 2,4,5-T (99.9% purity) were supplied
by the United States Environmental Protection Agency (EPA).
The commercial herbicide U 46 D Fluid, 2,4-D dimethylamine
(dimethylamine2,4-D salt) was purchased from BASFdo Brazil
S/A. A 1M boron trichloride solution in hexane (Sigma) and
ethyl ether (Merck) were also employed, as well as other sol-
vents and chemicals of analytical or chromatographic grade.
Apparatus
We used a model 370 GC gas chromatograph equipped with
an electron-capture 3H detector and a coiled borosilicate glass
column (6 ft x 1/8-in.o.d.) packed with 1.5% OV-17and 1.95%
QF-1 on 80-100 mesh, acid-washed chromosorb WHP. Quan-
titations were performed using a Hewlett-Packard model 3396
series I! integrator interfaced with the gas chromatograph.
The injector, column, and detector temperatures were 220,
210, and 255~ respectively.Nitrogen (high purity) was used
as the carrier gas at a flow rate of 56.8 mL/min.
Solutions
Standard stocks of 2,4-D and 2,4,5-T solutions (1000 pg/mL)
were prepared in acetone and kept frozen at -20~ A 2%
potassium bicarbonate solution was prepared in twice-distilled
water and was washed with hexane before use.
Derivatization
The standard solutions prepared in acetone and containing
2,4-D (50, 100, 250, 500, and 1000 ng/mL) were made by serial
dilutions of the 2,4-D stock solution. 2,4,5-T (250 ng) was
added as internal standard (IS) to each milliliter of these stan-
dard solutions. One milliliter was transferred from each stan-
251

dard solution to a stoppered reaction tube (15 mL) and evapo-
rated to dryness on a water bath (40-42~ under a gentle
nitrogen flow. Next, 50 IJL 2-chloroethanol and 100 ]JL BCI3
reagent were added to the stoppered tube, the contents were
thoroughly mixed, and the mixture was allowed to react
overnight at 60~ Ethyl ether (3 mL) was then added to the
mixture, and the extract was washed twice, first with a
KHCOg-water solution (3 mL) and then with water (3 mL). The
aqueous layer was discarded using a disposable pipette, and the
remaining ethyl ether extract was dried over anhydrous Na2SO4
in a bath (32-33~ under a mild nitrogen flow. The residue
was dissolved in 1 mL hexane, and 11JL of the solution was in-
jected into the chromatograph.
Blood and brain samples
Rats were killed by decapitation, and blood was collected in a
glass centrifuge tube at room temperature. Just after collection,
blood was allowed to clot in a refrigerator (5~ and then was
centrifuged (1000 x g); the serum was then removed and stored
for further use. Simultaneously, the brains of the animals were
removed, washed in a cold 0.9% NaCl-water solution, dried on
filter paper, frozen with solid CO2, and stored at -20~
Recovery
Brain tissue was homogenized for 8-10 s in 5N HCI (4 mL)
with the help of a model TE 099 Tecnal homogenizer. The
serum was acidified to a pH of less than 2.0 using 0.5N HCI.
Each stoppered tube containing homogenized brain (1 g)
was enriched with 0.5, 1.0, 5.0, or 10.0 IJg 2,4-D plus 1.0 IJg
2,4,5-T as IS. The serum (1.0 mL) was also enriched with 0.5,
2.5, 5.0, 10.0, or 50.0 ~g 2,4-D plus 5.0 IJg 2,4,5-T as IS. After
the addition of ethyl ether (5 mL), each tube was shaken vig-
orously for 1 rain and then centrifuged (1000 xg) for 3 rain;
after centrifugation, the supernatant was transferred and the
extraction procedure was repeated. Combinations of both 5-mL
extracts were concentrated to 3-4 mL, and IN NaOH (3-5
drops) was added to each tube until a pH of greater than 10 was
reached. Then, a 20% NaCl-water solution (5 mL) was added
to the tubes, which were shaken vigorously for 30 s and cen-
trifuged for 3 rain (1000 x g). After discarding the ethyl ether
layer, 1N HC1 (4-6 drops) was added in order to achieve a pH of
less than 2; ethyl ether (5 mL) was then added, and the tubes
were again vigorously shaken for 1 rain and centrifuged for 3
rain (1000 xg). Finally, using a disposable pipette, the aqueous
phase was discarded, and the organic layer was dried over
anhydrous Na2SO4. The ethyl ether extract was evaporated to
residue in a water bath (32-33~ with the help of a mild
nitrogen flow. The residue was then ready for derivatization.
Experimental intoxication
Adult male Wistar rats (195-280 g) of the same strain were
used. Water and food were provided ad libitum to the animals,
which were housed in a temperature-controlled room
(21-23~ Food was withdrawn 12 h before experimental
2,4-D oral intoxication. The animals used in this investigation
were maintained in accordance with the guidelines of the Com-
mittee on Care and Use of Laboratory Animal Resources,
School of Veterinary Medicine of the University of S~o Paulo,
252
Journal of Analytical Toxicology, Vol. 19, July/August 1995
Brazil; these guidelines were based on those reported and used
by the National Research Council (United States).
To study the absorbance rate of 2,4-D and the rate at which
it crossed over into the brain, we randomly divided the rats into
four groups of five animals each that received a single oral dose
of dimethylamine 2,4-D (200 mg/kg). Within each group, the
animals were killed 1, 2, 3, or 4 h later, respectively, for blood
and brain collection.
To analyze the effects of different 2,4-D doses, we randomly
and equally divided the rats into four groups of five animals
each that received a single oral treatment with dimethylamine
2,4-D (10, 60, 100, and 200 mg/kg, respectively). Three hours
after treatment, all animals were killed, and their brains and
serums were collected for chemical analyses. Some other rats
were treated with distilled water only (2 mL/kg), and their
brains and serums were used as controls.
Results
Figure I shows the calibration curve of 2,4-D 2-chloroethyl
ester, using 2,4,5-T as IS. Linearity was obtained in the
50-1000 pg/l~L range; the equation calculated for the curve was
y = 0.512385x - 0.046134.
The electron-capture detection (ECD) chromatograms
obtained for controls of serum and brain tissue and for 2,4-D
(0.5 IJg/mL) are illustrated in Figure 2. As mentioned earlier,
2,4,5-T was used as IS at concentrations of 5.01Jg/mL and 1.0
lJg/g for serum and brain tissue, respectively. The recoveries
observed were as expected, with respective retention times of
3.30 and 5.38 rain for 2,4-D and 2,4,5-T.
Absolute values of 2,4-D recoveries are presented in Table I
for serum and brain tissue. The mean absolute recovery values
for enriched samples of serum and brain tissue were 67.1 _+3.3
and 55.7 _+3.6, respectively; the coefficients of variation (CVs)
observed were 4.9% for serum and 6.3% for brain tissue. Rel-
ative recovery values are presented in Table II for serum and
brain tissue. Thus, the mean relative recoveries obtained for
enriched samples of serum and brain tissue were 98.5% + 4.8
500-
400.
2@)-
IOO-
O 9 - , 9
50 100 250 500 1000
2,4- D-ester (pgl
Figure 1.2,4-D calibration curve drawn using 2,4,5-T as internal standard.
Each point represents the mean of five determinations. AS, sample area;
AIS, internal standard area.
Journal o f Analytical Toxicology, Vol. 19, July/August 1995

and 93.3% • 7.5, respectively. The CV values were 4.8% for it is known that 2-chloroethanol increases ECD sensitivity to
serum and 8.0% for brain tissue. esters. The linear equation obtained for the standard curve
Table III shows 2,4-D absorption and its crossing into the validates this method.
CNS after a single oral dose (200 mg/kg). A direct relationship Using 2,4-D concentrations ranging from 1.0 to 100.0 pg/L,
could be seen between 2,4-D levelsof serum and brain tissue 1, Mierwa and Witek (14) obtained 92.1% recovery and Lee et al.
2, 3, and 4 h after administration. Table IV displays the ab- (15) obtained 73.4-94.9% recoveries.These values were higher
sorption of 2,4-D and its crossing into the CNS calculated 3 h than those reported here. However, those results were ob-
after oral administration of different doses. 2,4-D was detected tained from water samples. Working with biological materials,
both in serum and brain tissue after treatment with these such as serum, and after a very difficult extraction, Earne (11)
single doses.Thus, 2,4-D serum and brain levels each depended obtained recoveries of 79% • 4.2 with a CV of 5.3% and a
on dose and time elapsed after treatment. detection limit of 0.3 parts per million (ppm) or 1 pg/3 g of
sample. In this way, 2,4-D detection limits of 0.02 ppm,
] IJg/mL, and 0.05 rng/L in enriched samples of plant tissue,
serum, and urine, respectively, have already been reported
Discussion (16-18). In the present study, 2,4-D absolute recoveries from
serum and brain tissue were 67.1% • 3.3 and 55.7% • 3.6,
Despite the fact that some analytical methods use chloro- respectively; the sensitivity limit found was 250 ng/mL for
form (10) or benzene (11) for 2,4-D extraction, ethyl ether serum and 300 ng/g for brain. Thus, the recoveries detected
proved to be the most effective solvent (12). In the present ex-
periment, ethyl ether was used not only for that reason but also
for its fast evaporation and minimal toxicity for people in the Table I. Absolute 2,4-Dichlorophenoxyacetic Acid
laboratory.We also used a 50-pL quantity of 2-chloroethanol in (2,4-D) Recovery Values in Enriched Samples of Serum
and Brain Tissue
this study, which permitted us to completely derivatize at least
1000 IJg 2,4-D; however,2-chloroethanol should not be used in 2,4-D
excess or impurities may occur. According to Agemian and addition
Chau (13), the BCI3 method is the most sensitive one for to enrich 2,4-D pV Area
samples' injedion observed %Recovery
2,4-D determination and is the best choice for environmental
studies of water contamination. Figure 1 shows that the BC13 (pg) (pg) expected serum brain serum brain
reaction provides good results for biological materials too, at 0.5 100 2224 1536 1089 69.1 59.8
least up to a detection limit of 50 ng. Furthermore, BCI3 in 1.0 100 2224 - 1031 57.6
hexane solution was also preferred because it allows the use of 2.5 100 2224 1512 - 68.0
2-chloroethanol instead of methanol. This is relevant because 5.0 100 2224 1499 918 67.4 52.9
10.0 500 8945 6244 4694 69.8 52.5
50.0 1000 17,347 10,651 - 61.4
E F. E
* 2,4,5-T (I.0 and 5.0 pg) was used as internal standard in each brain or serum
tube, respectively.

"-2.
Table II. Relative 2,4-Dichlorophenoxyacetic Acid
I.,. (2,4-D) Recovery Values for Serum and Brain Tissue*
i
lal
ql. 2,4-D Mean 2,4-D recoveryt
concentration
I- (pg/g or serum brain
w,i
pg/mL) SD* CVS SD CV
o ?,,,-
0.5 101.5 _+8.42 8.02 93.8 _+3.11 3.32
1,11
.=. 9
,~
1.0 - - 88.2 + 11.10 12.58
2.5 91.6 _+3.47 3.79
Z i'
5.0 100.2 _ 1.51 1.51 87.4 + 5.02 5.74
10.0 99.0 + 2.95 2.98 103.8 + 4.45 4.29
50.0 100.0+4.12 4.12

A B 1: 0 Mean 98.5 + 4.82 4.86 93.3 _+7.55 8.09

Figure 2. Typical electron-capture detection gas chromatogram of 2,4-D
* Relative to 1.0 or 5.0 pg 2,4,5-T used as internal standard for brain or serum,
and 2,4,5-T2-chloroethanol esters.(A) Control serum, (B) 2,4-D recovery
respectively.
from serum(0.5 pg/mLserum plus 5.0 pg internal standard), (C) control i. Data are the mean of three determinations.
brain, and (D) 2,4-D recovery from brain (0.5 pg/g brain plus 1.0 pg in- SD = Standard deviation.
w CV = Coefficient of variation.
ternal standard).

253
 Journal of Analytical Toxicology, Vol. 19, July/August 1995

here were not as high as those reported in the literature. How-
ever, we emphasize that the extraction method employed was
precise, as clearly demonstrated by the CVvalues of 4.9 and 6.3
for serum and brain, respectively. Furthermore, the relative
recoveries of 2,4-D were 99.16% _+4.82 and 93.3% _+7.55 for
serum and brain. Because no peaks were detected in the con-
trol samples (Figure 2), the relative recovery values were used
to correct the real 2,4-D values for the remaining samples.
The LD50of oral 2,4-D in rats is 945 mg/kg (9). Within one-
half hour of acute administration of dimethylamine 2,4-D, the
animals started to show signs and symptoms of CNS depres-
sion, and 1 h after intoxication they exhibited tachypnea and
ataxia, as well as a motor function impairment more noticeable
in the hind limbs.
The peak effect of 2,4-D oral intoxication in rats occurred
between 2 and 4 h after administration. According to
Papenheimer et al. (19) and Kim et al. (8), 2,4-D reaches the
CNS through an organic anion transportation system.
Accordingto Desi and Sos (20) and Elo and Ylitalo (7), signs of
depressive CNS states in animals severely poisoned with 2,4-D
were a consequence of a partial break of the blood-brain barrier
and subsequent herbicide accumulation within the brain.
Mattesson et al. (21) stated that only higher 2,4-D doses
(greater than 150 mg/kg) could injure the blood-brain barrier.
In the present experiment, the 2,4-D doses used to intoxicate
the animals were at least 4 times lower than the LD50found for
Table III. 2,4-Dichlorophenoxyacetic Acid (2,4-D)
Concentrations from Rat Serum and Brain Tissue Taken
at Different Times*
Time after 2,4-D concentrationt
administration serum brain
(h) (pg/mL) (pg/g)
1 478.8 +_31 .l 22.5 _+3.6
2 499.0 _+34.1 25.6 _+3.9
3 514~6 +_53.9 28.7 + 2.3
4 596.9 _+55.6 36.9 + 1.0
the herbicide.Afteradministration of these nonlethal doses, the
herbicide reached the CNS, and concentrations found in brain
tissue increased in accordance with the doses used to intoxicate
the rats. The percentage of 2,4-D that reached the CNS in
relation to that detected in the serum varied from 1.6% for the
lower herbicide dose (10 mg/kg) to 5.6% for the highest one
(200 mg/kg). Serum 2,4-D concentrations detected 1 h after
oral administration (200 mg/kg)were 478.8 1Jg/mL;this value
was very close to those reported by Osterloh et al. (18) and Park
et al. (22) in patients severely intoxicated with lethal over-
doses of 2,4-D (540.9 and 400 IJg/mL, respectively). In the
latter report, as well as in many others, 2,4,5-T was used as IS.
As illustrated in Figure 2, this choice seems appropriate; in-
deed, under the present experimental conditions, 2,4-D and
2,4,5-T peaks were not only separate but also quite distinct.
However, although 2,4,5-T is a reasonable choice for IS,
caution should be taken with its use, as coexposure to this
substance is also possible in environmental field experiments.
The herbicide was observed in blood and brain 1 h after its
oral administration, and 2,4-D concentrations in those com-
partments increased within 1-4 h after intoxication. These
results suggest, first, that the crossing of 2,4-D from serum to
the brain is slow and, second, that 2,4-D can reach the CNS at
doses not as high as those reported by Mattesson et al. (21). The
exact mechanism by which 2,4-D reaches the brain cannot be
explained by the present data. We have previouslyreported (9)
that 2,4-D at doses similar to those used in the present exper-
iment modifiednot only the open-fieldbehavior of rats but also
the functional activity of 5-HT within the CNS.
Taken together, the present data suggest that the method
used in this experiment to determine 2,4-D in biological
samples from rats orally and acutely intoxicated with this
herbicide is efficient. In fact, the smaller dose (10 mg/kg),
which is almost the LDsJ100 and which did not induce any
noteworthy toxic effect, resulted in 2,4-D concentrations of
approximately 143 1Jg/mL and 2.3 IJg/g in serum and brain
tissue, respectively.

* 200 mg/kg oral dose administered. References

Mean +_standard deviation of five animals.

1. A. Alves. Defensivos agricolas ou agrotoxicos? Consumo usos e

abusos. Cienc. Hoje 4:49-58 (1986).
2. International Agency for Researchon Cancer (IARC). Monographs
Table IV. 2,4-Dichlorophenoxyacetic Acid (2,4-D) on the evaluation of the carcinogenic risk of chemicals to man.
Concentrations in Rat Serum and Brain Tissue Taken Some fumigants, the herbicides 2,4-D and 2,4,5-chlorinated
Three Hours after Treatment with Different Oral Doses dibenzodioxins and miscellaneous industrial chemicals. IARC
of the Herbicide Monogr. Set. 15:111-38 (1977).
3. World Health Organization. 2,4-Dichlorophenoxyacetic acid
Dimethylamine 2,4-D concentration' (2,4-D). Geneva, environmental health criteria. WHO Monogr.
Ser. 29:1-44 (1984).
2,4-D serum brain
4. C. McGourty. Agent orange study blasted. Nature 340:179-83
(mg/kg) (pg/mt) (pg/g)
(1989).
10 143.0 _+12.5 2.3 _+0.2 5. F.D.A. Zambrone. Defensivos agffcolas ou agrot6xicos. Perigosa
famflia. Cienc. Hoje 4:42-48 (1986).
60 365.6 + 15.1 8.3 _+1.2
6. K. Tyynela, H.A. EIo, and P. Ylitalo. Distribution of three common
100 413.7 +_10.5 14.4 + 1.8
chlorophenoxyacetic acid herbicides into the rat brain. Arch.
200 514.6 _+53.9 28.7 + 2.3 Toxicol. 64:61-65 (1990).
7. H.A. EIo and P. Ylitalo. Distribution of 2-methyl-4-chloro-
* Mean _+standard deviation of five animals.
phenoxyacetic acid and 2,4-dichlorophenoxyacetic acid in male

254
Journal of Analytical Toxicology,Vol. 19, July/August1995
rats: evidence for the involvement of the central nervous system
in their toxicity. Toxicol. Appl. Pharmacol. 51 : 439-46 (1979).
8. C.S. Kim, R.F. Keizer, and J.B. Pritcard. 2,4-Dichlorophenoxy-
acetic acid intoxication increases its accumulation within the
brain. Brain Res. 440:216-26 (1988).
9. G.H. Oliveira and J. Palermo-Neto. Effects of 2,4-dichloro-
phenoxyacetic acid (2,4-D) on open-field behavior and neuro-
chemical parameters of rats. Pharmacol. Toxicol. 73:79-85
(1993).
10. W.L. Zielinski, Jr. and L. Fishbein. Gas chromatographic mea-
surement of disappearance rates of 2,4-D and 2,4,5-T acids and
2,4-D esters in mice. J. Agric. Food Chem. 15:8411-14 (1967).
11. K. Earne. Determination of phenoxyacetic herbicide residues in
biological materials. Acta Vet. Scand. 7:77-96 (1966a).
12. I.H. Sufflet. The p-value approach to quantitative liquid-liquid ex-
traction of pesticides and herbicides from water. 3. Liquid-liquid
extraction of phenoxy acid herbicides from water. J. Agric. Food
Chem. 21:591-98 (1973).
13. H. Agemian and A.S.Y. Chau. Analysis of pesticide residues by
chemical derivatization. V. Multiresidue analysis of eight
phenoxyalkanoic acid herbicides in natural waters. J. Assoc. Off.
Anal. Chem. 60:1070-77 (1977).
14. S. Mierwa and S. Witek. Gas-liquid chromatographic method
with electron-capture detection for the determination of residues
of some phenoxyacetic acid herbicides in water as their 2,2,2-
trichloroetyl esters. J. Chromatogr. 136:105-11 (1977).
15. H.B. Lee, Y.D. Stokker, and A.S.Y. Chau. Chemical derivatization
analysis of pesticide residues. X. Analysis of ten acid herbicides
in natural waters. J. Assoc. Off. Anal. Chem. 69" 557-60 (1986).
16. W.J. Allender. Determination of chlorophenoxy herbicides in
plant material exposed to spray drift. J. Chromatogr. Sci. 27:
193-96 (1989).
17. W.M. Draper. A multiresidue procedure for the determination
and confirmation of acid herbicide residues in human urine. J.
Agric. Food Chem. 30:227-31 (1982).
18. J. Osterloh, M. Lotti, and S.M. Pond. Toxicologic studies in a
fatal overdose of 2,4-D, MCPP, and chlorpyrifos J. Anal. Toxicol.
7:125-29 (1983).
19. J.R. Papenheimer, S.R. Heisey, and E.F. Jordan. Transport of
diodrast and phenolsulfonphthalein from cerebrospinal fluid to
blood. Am. J. Physiol. 200:1-10 (1961).
20. I. Desi and J. Sos. Central nervous injury by a chemical herbicide.
Acta Med. Acad. Sci. Hung. 18:429-33 (1962).
21. J.L. Mattesson, K.A. Johnson, and R.R. Albee. Lack of neuro-
pathologic consequences of repeated dermal exposure to 2,4-
dichlorophenoxyacetic acid in rats. Fundam. Appl. Toxicol. 6:
175-78 (1986).
22. J. Park, I. Darrien, and L.F. Prescot. Pharmacokinetic studies in
severe intoxication with 2,4-D and mecoprop. Proc. Eur. Soc.
ToxicoL 18:154-55 (1977).
Manuscript received September 30, 1992;
revision received October 12, 1994.
255