EPA-600/1-76-037
November 1976
FEASIBILITY OF APPLYING FIELD IONIZATION
MASS SPECTROSCOPY TO PESTICIDE RESEARCH
By
R.L. Dyer, H. d'A. Heck, A.C. Scott and M. Anbar
Stanford Research Institute
Menlo Park, California 94025
Contract No. 68-02-1799
Project Officer
August Curley
Environmental Toxicology Division
Health Effects Research Laboratory
Research Triangle Park, N.C. 27711
U.S. ENVIRONMENTAL PROTECTION AGENCY
OFFICE OF RESEARCH AND DEVELOPMENT
HEALTH EFFECTS RESEARCH LABORATORY
RESEARCH TRIANGLE PARK, N.C. 27711
I3RARY
•' cHY!RG,sivlENTAL PRO TEC i (ON AGENCY
^,v4 N. L 08817,
-------
DISCLAIMER
This report has been reviewed by the Health Effects Research
Laboratory, U.S. Environmental Protection Agency, and approved for
publication. Approval does not signify that the contents necessarily
reflect the views and policies of the U.S. Environmental Protection
Agency, nor does mention of trade names or commercial products
constitute endorsement or recommendation for use.
11
-------
FOREWORD
The many benefits of our modern, developing, industrial society are
accompanied by certain hazards. Careful assessment of the relative risk
of existing and new man-made environmental hazards is necessary for the
establishment of sound regulatory policy. These regulations serve to
enhance the quality of our environment in order to promote the public
health and welfare and the productive capacity of our Nation's population.
The Health Effects Research Laboratory, Research Triangle Park
conducts a coordinated environmental health research program in toxicology,
epidemiology, and clinical studies using human volunteer subjects. These
studies address problems in air pollution, non-ionizing radiation,
environmental carcinogenesis and the toxicology of pesticides as well as
other chemical pollutants. The Laboratory develops and revises air quality
criteria documents on pollutants for which national ambient air quality
standards exist or are proposed, provides the data for registration of new
pesticides or proposed suspension of those already in use, conducts research
on hazardous and toxic materials, and is preparing the health basis for
non-ionizing radiation standards. Direct support to the regulatory function
of the Agency is provided in the form of expert testimony and preparation of
affidavits as well as expert advice to the Administrator to assure the
adequacy of health care and surveillance of persons having suffered imminent
and substantial endangerment of their health.
This report presents data on the feasibility of measuring subnanogram
quantities of pesticides in animal blood, tissues and excreta by use of
nonradioactive isotope dilution analysis. Using multilabeled compounds
as isotopic dilutants and nonfragmenting field ionization mass spectrometry
(FIMs) as the analytical instrument, picomole (10-^M) quantities of organic
compounds can be quantitatively determined. This high sensitivity and
quantitative accuracy permits the determination of pesticide residues in
biological samples at very low levels. These data will provide a more
substantiated assessment of persistence and degradation (biotransformation)
of the compound and the potential hazard associated with exposure.
John H. Knelson, M.D.
Director,
Health Effects Research Laboratory
111
-------
ABSTRACT
An isotope dilution methodology was developed for analysis
of an insecticide, parathion, and a herbicide, trifluralin, isolated from
rat tissues and excreta. Sample cleanup was facilitated by use of
high-pressure gel permeation chromatography in conjunction with thin
layer chromatography and reversed-phase high pressure liquid chrom-
atography. Isotope ratio measurements were performed using multilabeled
stable isotopic carriers and nonfragmenting field ionization mass
spectrometry. Parathion and trifluralin were administered intra-
peritoneally and/or orally at the sub-mg/kg level, and the unchanged
materials assayed in tissues and excreta at the ppb level. The
technique was also applied to the determination of parathion and
methyl parathion concentrations in aerosols. The biological
implications of the results of the animal experiments are discussed.
IV
-------
CONTENTS
I INTRODUCTION 1
II EXPERIMENTAL 2
II A SYNTHESIS OF MULTILABELED CARRIERS 2
D -Parathion 2
10
D -Trifluralin 4
7
D -Methyl Parathion 5
6
II B EXTRACTION AND CLEANUP OF PTN AND TFN
IN ANIMAL TISSUES 8
Extraction Methods 8
Cleanup Methods 9
Purification of PTN and TFN Following
Cleanup 14
II C EXPERIMENTAL PROCEDURES 17
Animal Administration Regimens 17
Parathion IP 17
Trifluralin IP 17
Trifluralin Oral 18
Extraction Procedures 21
Gel Permeation Chromatography 23
High Pressure Liquid Chromatography .... 23
Thin Layer Chromatography 25
Mass Spectrometer Isotope Ratio Measurements 28
III RESULTS AND DISCUSSION 29
Cleanup of Extracts from Animal Tissues by High-
Pressure Gel Permeation Chromatography 29
Analysis of Mass Spectrometric Data 30
Calculation of Isotope Ratio 30
-------
Detection Limits 31
Background Correction and Error Analysis 35
Determination of PTN in Rat Tissues by Isotope Dilution 39
Determination of TFN in Rat Tissues by Isotope Dilution 44
IV DETERMINATION OF THE CONCENTRATION OF PTN AND MePTN
IN AEROSOLS 53
Experimental 53
Results and Discussion 54
V CONCLUSIONS 57
REFERENCES 60
APPENDIX: RAW COUNT DATA 62
VI
-------
LIST OF FIGURES
1. FIMS Scan—D -Parathion 3
10
2. FIMS Scan—D -Trifluralin 6
7
3. FIMS Scan—D -Methyl Parathion 7
6
4. Separation of Benzene and Phenol by High-Pressure Gel
Permeation Chromatography 13
5. Parathion Purification—Gel Permeation Chromatograms .... 24
6. Trifluralin Purification—Gel Permeation and Reversed-
Phase High Pressure Liquid Chromatograms 26
7. FIMS Scans—Parathion: Comparison of HPLC and TLC
Purification Techniques 32
8. FIMS Scans—Parathion 33
9. FIMS Scans—Trifluralin IP 34
10. FIMS Scans—Trifluralin Oral 36
11. Trifluralin Concentrations in Adipose Tissue and Liver
of Rats Following IP Injection 51
-------
LIST OF TABLES
1. Animal Data: Parathion IP Regimen 18
2. Animal Data: Trifluralin IP Regimen 20
3. Animal Data: Trifluralin Oral Regimen 22
4. Measured Parathion Levels in Rats 40
5. Summary—Parathion IP Results 41
6. Measured Trifluralin Levels in Rats—IP Administration .... 45
7. Summary—Trif luralin IP Results 46
8. Measured Trifluralin Levels in Rats—Oral Regimen 47
9. Summary—Trifluralin Oral Results 48
10. Inhalation Chamber Measurements 55
viii
-------
AC KNOW LE DGEME NTS
The authors wish to thank Ms. Shirley Madan and
Dr. Roger Earth of the SRI Life Sciences Division
for their skillful assistance in performing the
animal injections, feedings and necropsies. Mr.
David Burkhouse was responsible for the synthesis
of the multilabled materials. Mr. Russell Sperry
and Mr. Gary Gonser carried out the mass spectro-
metric analyses. Mr. Albert Lobato and Mr.
Ferdinand Engesser were responsible for construction
of the gel permeation apparatus and for maintenance
of the mass spectrometers. We especially thank Ms.
Linda Morehead and Ms. Romen Rey for careful
preparation of this manuscript.
IX
-------
I INTRODUCTION
This is the first annual report on a research program sponsored by
the Environmental Protection Agency under Contract #68-02-1799.
The objective of this program has been to test the feasibility of
determining very low quantities of insecticides and herbicides in animal
tissues, urine, feces, and plasma by the use of nonradioactive isotope
dilution analysis. Using multilabeled organic molecules as isotopic
dilutants and nonfragmenting field ionization mass spectrometry (FIMS) as
the analytical instrument, picomole quantities of organic compounds can be
quantitatively determined. This very high sensitivity of detection may
allow the determination of pesticide residues in animal tissues at much
lower levels than has been heretofore possible. Such data will provide
a more substantiated assessment of the long-term persistence and potential
hazards involved, in exposure of animals or humans to toxic chemicals.
During the first year of this research program, we have used the
technique of isotope dilution analysis to study the distribution and
persistence of parathion (PTN) and trifluralin (TFN) in rat tissues,
plasma, urine, and feces following low dose injection of either compound
or repeated oral administration in the case of TFN. The high sensitivity
of the technique enabled us to determine ppb concentrations of PTN or
of TFN in most of the biological materials examined.
Section II A describes the syntheses of multilabeled PTN,. methyl PTN
and TFN. Section II B contains a review of the new methodologies
developed for sample cleanup by high-pressure gel permeation chromatography.
Isotopic ratios were measured by multiscanning field ionization mass
spectrometry using a cooled sample probe. The animal experiments and the
details of the experimental procedures are summarized in Section II C.
Results are compiled, discussed and compared with the literature in
Section III. Section IV describes a brief experiment using isotope dilution
techniques to measure the concentration of PTN and methyl-PTN in aerosols.
Section V contains the conclusions of our stuciv
-------
II EXPERIMENTAL
II A SYNTHESIS OF MULTILABELED CARRIERS
d, -Parathion
The synthetic route was:
C D OH + SPC1 — KL^ SP(OC D ) Cl (I)
A o «.5 & o &
, ___
-/' \V 0-
I + p-NOr phOH — A NO -' 0-P-OC D (d -parathion)
£t £j \ in / I ^ D JL U
• - ' OC D
2 5
O,0-diethyl-d -chlorothiophosphate (I) was prepared by addition of
5 g d -ethanol in pyridine and benzene to 9.3 g thiophosphorylchloride
5 o 1
in benzene at 0-20 C. Pyridine hydrochloride was filtered and washed
with benzene, the filtrate and washings condensed, and 3.32 g of the
product collected by distillation at 80-92 C/20 torr.
Parathion was prepared by refluxing I with 2.32 g p-nitrophenol and
2
18 g sodium carbonate in acetone for 4 hours. The filtrate was dissolved
in benzene, washed with aqueous sodium carbonate, and dried over sodium
sulfate. The product (3.3 g, 65% yield) was distilled and collected at
o
45-50 C/0.03 torr. Identity of the product was confirmed by IR and NMR.
A field ionization mass spectrum of the d -parathion is shown in
Fig. 1. The isotopic purity, i.e., the ratio of counts at mass 291
(unlabeled) to the total of counts at masses 299-304 (labeled) , is
-5
5.7 x 10 .
-------
FIGURE 1
FIMS SCAN: D -PARATHION
d10-PARATHION
ISOTOPIC PURITY 6/105
10
8 K
291
301
MASS NUMBER
SA-4280-1
-------
d -Trifluralin
7
The synthetic route was:
n-propyl-NH + d -n-PrBr
— 2 7
reflux
n-Pr
d -n-Pr
7 —
NH
(I)
CF
CF
HNO
(ID
(d -trifluralin)
NO
N-propyl-d -n-propylamine (I) was prepared by refluxing 4.4 g n-propyl
7 3
amine with 5 g n-propyl-d -n-propyl bromide (Merck). 4N sodium hydroxide
o
was added and the amine distilled up to 110 C, yielding an azeotropic
o
mixture. This was dried over NaOH pellets and distilled at 108-110 C,
giving 3 ml of the desired product.
3,5-dinitro-4-chlorobenzotrifluoride (II) was prepared by dropwise
addition of 80 g 96% H SO and 60 g 90% HNO to a mixture of 20 g 3-nitro-
4
4-chlorobenzotrifluoride in 40 g 30% oleum. The mixture was heated at
o o
100 C for 3 hours and then 105-110 C for 30 rain. The mixture was cooled,
poured over ice, the precipitate filtered, washed and dried under vacuum.
Recrystallization from ethanol gave an 85% yield of the product, melting
point 58 C.
d -trifluralin was prepared by heating a mixture of 2.4 g II and 3 ml
o 5
I at 100 C for 2 hours. The mixture was dried with anhydrous ether, the
amine hydrochloride removed by filtration and the filtrate concentrated
to give 3 g crude product. Recrystallization from hexane gave 1.3 g (43%)
of the final product. Identity was confirmed by melting point, IR and NMR.
-------
A field ionizat^on mass spectrum of the d -trifluralin is shown in
Fig. 2. The material contains a significant amount of unlabeled
_ o
contaminant at mass 335; the isotopic purity is only 1 x 10
The d contaminant may be attributed to the presence of some unlabeled
material in the d -n-propyl bromide (a reagent quality control problem we
have encountered before). Because of the relatively high level of unlabeled
material, which limits the dynamic range of the isotope dilution measure-
ment, we administered the labeled material to the animals, using unlabeled
trifluralin as carrier - the reverse of the parathion procedure.
d -Methyl Parathion
6 _
D -methyl parathion was prepared for use in the inhalation chamber
6
experiments (Section IV). The synthetic route was:
CD OD + SPC1 - > SP(OCD ) Cl (I)
3 3 32
I + p-NO phOH
£1
O,0-dimethyl-d -chlorothiophosphate (I) was prepared by dropwise
6
addition of lOg d -methanol (Merck) to a cooled solution of thiophosphoryl-
chloride, followed by 10 g of NaOH in 15 ml of water. The reaction
o
temperature was maintained from -20-0 C. The sodium chloride was dissolved
in water and the organic layer separated. 7.0 g (14%) of the product was
o
collected from a spinning band distillation at 58-60 /22 torr.
Methyl parathion was prepared by dropwise addition of 3.8 g of I to
a mixture of 3.2 g ]D-nitrophenol, 2 g sodium carbonate, and 23 ml acetone,
2
and refluxed for 4 hr. The mixture was cooled, filtered, and the filtrate
concentrated. 3.5 g(58%) of the product was recrystallized from methanol
in dry ice-acetone. Identity was confirmed by IR, NMR and melting point.
A field ionization mass spectrum of the d -methyl parathion is shown
-5 6
in Figure 3. The isotopic purity is 2 x 10
-------
S5
M
J
o
03
CM
CN
•*
n
IM
tr
OJ
CD
5
o
I— I
En
Sz;
<
o
i
CO
I—I
EM
CO
O
in
CO
n
C t
13 CC
-I D
LL CL
O
CO
-------
W
O
i—i
ft
°
-------
II B EXTRACTION AND CLEANUP OF PTN AND TFN IN ANIMAL TISSUES
The analysis of pesticide residue levels in animal and plant tissues
generally includes the three phases of extraction, cleanup, and determi-
nation. The classical techniques of extraction and cleanup, summarized
6
by Burchfield and Johnson in 1965, continue to be widely used up to
date. Major innovations, however, have arisen in the techniques for
determination through the application of more sensitive or more rapid
analytical tools, such as GC, HPLC, GC/MS, and isotope dilution/FIMS.
The state of the art of many of these techniques as they apply to pesticide
residue analysis, has been reviewed recently, e.g., J» Chromatog. Sci. 13,
No. 5 and 6 (1975); Anal. Chem. 47, 157R (1975).
In the course of this study, we investigated a number of classical
methods used for the extraction and cleanup of residues from lean and
fatty animal tissues, bodily fluids, and feces. The goal in these
investigations was to select or to develop efficient methods, broadly
applicable to many tissues, that require a minimum of labor.
Although the conventional techniques were found to be effective, they
are disappointingly slow and laborious, particularly the techniques used
for cleanup of sample extracts. For this reason, a new approach was
developed in which high-pressure gel permeation chromatography (GPC) was
used as the first step in sample cleanup. Subsequent chromatography,
either by TLC or reversed-phase HPLC , yielded samples having an exception-
ally low level of chemical impurities. This section summarizes the
various methods of extraction and sample cleanup that were tried during
this research program.
Extraction Methods
6
Following Burchfield and Johnson, samples of rat liver (typically
3-4 g) were initially ground in a mortar and pestle in the presence of
about 3 g of Na SO . To test the efficiency of extraction, approximately
-------
40 jg of TFN were added to several samples. In the classical methods,
this st°p is usually followed by boiling the ground tissue in an organic
solvent or by Soxhlet extraction. To avoid such time-consuming procedures,
the ground tissues were homogenized in the presence of an organic
solvent, either acetone or hexane (20 ml).
To reduce the time required for extraction, the initial grinding of
tissue with Na SO was omitted. Homogenization was carried out in the
presence of 20 ml of hexane using a Tekmar Tissumizer. In addition to
being less laborious, this method appeared to be more efficient, as the
liver was more thoroughly dispersed in the organic solvent than
•*as the case following Na SO grinding. The Tissumizer technique was
2 4
tested on a number of different tissues, as well as on feces, urine and
plasma. Following cleanup (see below), analyses were performed by TLC.
Moderately polar organic molecules, such as PTN and TFN, were extracted
efficiently in all cases with 90% hexane/ether.
It is well known that homogenization of adipose tissue in the presence
of organic solvents results in the extraction of substantial quantities of
lipids. The same result is found, although to a much smaller extent, in
the case of brain and liver. When such extracts are dried, an oily residue
remains that cannot, in many cases, be effectively chromatographed, either
by TLC or GC. Consequently, a cleanup procedure is required to remove the
bulk of the lipids before chromatographic purification or analysis can be
performed.
Cleanup Methods
The classical procedures of sample cleanup involve partitioning
between polar and non-polar organic solvents, and/or sorption on various
types of solid supports (e.g., charcoal, Florisil, siloxid) followed by
6
batchwise or continuous elution. A recent review suggests that no
substantial alterations in these methods have been made within the past
decade. It seemed likely that the application of HPLC techniques, gel
permeation chromatography in particular, in the area of sample cleanup
might significantly reduce the time, labor and expense involved in this
phase of the analytical procedure.
9
-------
High-efficiency gel penr.eation columns appeared to offer many
advantages as a first step in sample cleanup. Such columns achieve
separation almost soleLy on the basis of molecular size. Since
adsorption effects are minimal in such systems, the possibility that
constituents in an organic extract would bind strongly to the column
was deemed unlikely. It was expected, therefore, that gel permeation
columns could be used for a large number of samples before column
plugging or loss of resolution became severe.
During the first six months of this project, however, we did not
have access to a gel permeation chromatograph. Although a system had
been designed early in the project, various delays caused by the slow
delivery of components and the construction of new chemistry laboratories
postponed final acquisition of a working system until raid-winter. During
this time, some of the classical techniques for sample cleanup were
investigated, on the possibility that construction of the GPC system
would require more time than expected.
The experiments that were carried out in this area involved partition-
ing trifluralin and parathion, after extraction from fat, liver, brain, or
feces, between different organic solvents. The solvents used were hexane/
acetonitrile, hexane/dimethylformamide, and hexane/methanol. As expected,
the compounds of interest partitioned preferentially into the more polar
phase (except for hexane/methanol), while the bulk of the lipid impurities
remained in the less polar, hexane-rich phase. Although the partition-
ing ratios were favorable, it was found, in general, that at least two,
and preferably three extractions using fresh polar solvent were required
to thoroughly extract PTN and TFN from the hexane phase.
Concentration of the partitioned extracts to dryness revealed that,
whereas most of the lipid materials had been removed by partitioning, the
amount of oily residue remaining was still unacceptably high in the case
of fat. If, however, the acetonitrile phase was first added to an aqueous
10
-------
solution containing 5% Na SO , followed by re-extraction with hexane,
the extracts were found to be sufficiently pure to permit analysis by
TLC. The same method was tried on several other tissues, and satisfactory
results were obtained in all cases.
These procedures involving multiple extractions would have been
extremely tedious had they been used for large numbers of samples. Cleanup
by sorption on solid supports, which is normally used as an alternative or
as an adjunct to cleanup by partitioning, would probably have been at least
as time-consuming and was not, therefore, attempted. Work was directed
instead toward the development of a successful GPC procedure.
A relatively inexpensive GPC system was constructed in this laboratory,
8
the general design being based on that of Snyder and Kirkland. Pumping
is provided by a pneumatic amplifier pump (Haskel Engineering Co., Los
Angeles, Model 28646-4) with a maximum pressure rating of 7000 psi. The
pump is driven by filtered laboratory air, regulated at 80 psi. A second
variable regulator at the pump reduces this pressure to approximately 22
psi, so that with a 46:1 pneumatic amplification, the output pressure is
maintained at 1000 psi. This pressure provides a near-optimal flow rate in
our gel permeation system.
For an isocratic gel permeation system, the Haskel pump provides many
advantages. Its cost is low, and, in contrast to motor-driven, recipro-
cating pumps in the same price range, a stable rate of flow is maintained.
This enables the Haskel pump to be used, if desired, with detectors that
may be sensitive to fluctuations in flow rate or pressure, such as a
refractive index detector. Moreover, the output of the Haskel pump is
variable over a large range, so that it can be used for both analytical
and preparative applications.
Prefiltered solvent at 1 atm pressure is gravity-fed to the pump
through a 1/2" stainless steel tube. Outflowing solvent passes through
1/4" stainless steel tubing to a 35 micron line filter and then to a
cross. Two arms of the cross are connected to high-pressure valves, one
11
-------
of which carries solvent back to the solvent reservoir for rapid recycling
and priming of the pump, while the other goes to waste. The remaining
arm is connected to a 1/4"-1/16" reducer, thence through 1/16" tubing to
a 15 micron line filter, a tee, and a third high-pressure valve. The
third arm of the tee is connected to a flow-through, 0-5000 psi pressure
gauge (American Chain and Cable, Helicoid Div., Bridgeport, Conn., Model
440L). The valves, filters, cross, and tee were purchased from High
Pressure Equipment Co., Erie, Pa.
The sample injector (Rheodyne, Berkeley, Ca., Model 7105, with a
1.1 ml sample loop) follows the third high-pressure valve and is located
as close as possible to the inlet of the first of three gi-Styragel
columns, 100 A pore (Waters Associates, Milford, Mass.), connected in
series. At a pressure of 1000 psi, the flow rate through the system is
approximately 1.25 ml/min, which is close to the optimum for gel permeation
using 10 micron size particles.
A dual wavelength UV detector (Spectra-Physics, Berkeley, Ca., Model
230) connected to a dual chainnel strip chart recorder (Linear Instruments,
Irvine, Ca., Model 285) monitors the absorbance of the column outflow
simultaneously at 254 and 280 run. Following the detector, a Rheodyne
Teflon rotary valve (Model 50-11) is used either to manually collect an
individual peak or to direct the flow of solvent to waste.
An example of the high resolution achievable with three p,-Styragel
columns in series is shown in Fig. 4, which illustrates a size separation
between benzene and phenol with tetrahydrofuran as solvent. Phenol, being
the larger of the two molecules, is eluted first. The number of theoret-
2
ical plates can be calculated from the equation N_ = 16 (V/w) , where V
is the elution volume and w is the baseline width of the peak in the same
units. For this case, the calculated number of plates is 4800. The
running time required for this separation was 20 minutes.
12
-------
FIGURE 4
SEPARATION OF BENZENE AND PHENOL BY
HIGH-PRESSURE GEL PERMEATION CHROMATOGRAPHY
T
Packing: /u-Styragel, 100 A pore
Solvent' Tetrahydrofuran
Column: 7 mm x 30 cm,
3 Columns in Series
Pressure- 1000 psi
Flow Rate: 1.3 ml/mm
BENZENE
ABSORBANCE
0.1
14
16
18
TIME — mm
20
22
SA-4280-1 1
13
-------
A very important feature; of high pressure liquid chromatographic
techniques for the cleanup of crude organic extracts is that large volumes
of sample can be injected without significantly degrading the efficiency
of the column. This property, unique to liquid chromatography, was
9
pointed out by DeStefano and Beachell for chromatography on silica and
10
was reiterated by Moye. We have investigated this point for the case
of gel permeation. Acceptable bandwidths and resolutions were found with
sample sizes as large as 1 ml, although 0.5 ml sample sizes gave somewhat
better resolution. There was practically no difference in the bandwidth
of peaks or in the resolution achieved in our GPC system for any size
sample up to 0.5 ml.
To estimate the sensitivity of an isotope dilution assay following
sample cleanup by GPC, a series of rat fat extracts were prepared contain-
ing various known amounts of unlabeled parathion and a known excess of
the deuterated parathion as carrier. The extracts were chromatographed by
GPC, the parathion peak collected, and the isotopic ratio determined by FIMS.
Details of the chromatographic procedure are given below. The resulting
mass spectra indicated that PTN could be determined in fat at the ppm level
following a single gel permeation chromatography. Not surprisingly,
however, gel permeation did not completely remove all of the potentially
interfering impurities. It was evident that detection of PTN at lower
levels in fat would require a further reduction in the amounts of these
impurities. The final purification techniques used for PTN and TFN are
described in the next section.
Purification of PTN and TFN Following Cleanup
Both TLC and HPLC were used as final purification techniques before
mass spectrometric analysis. Purification by TLC on silica gel plates
has been the standard method used for isotope dilution analysis in this
11 12
laboratory. ' Since an extensive literature exists on the TLC of PTN
and TFN, the development of effective purification procedures was
straightforward.
14
-------
A Waters model COOO high-pressure liquid chromatograph with gradient
programming provided the capability to carry out efficient and rapid liquid
chromatographic separations. High-pressure adsorption chromatography on
silica is physically analogous to TLC, and the information learned from
TLC is directly applicable. This chromatographic technique was not attempted,
however, because of the tendency of silica to exhibit irreversible adsorp-
tion. Such an effect is expected with some components of a complex
extract from animal tissues.
Preliminary experiments were performed in which the HPLC apparatus
was used in a normal-phase partitioning mode for purification of PTN and
TFN. In normal-phase liquid-liquid chromatography (LLC), the support is
impregnated or bonded to a polar stationary phase, while a relatively
non-polar solvent is used as the mobile phase. In reversed-phase LLC,
the two phases are interchanged (i.e., less polar stationary phase,
polar moving phase).
Normal-phase LLC appeared to offer one major advantage over
reversed-phase LLC for the purification of volatile compounds. In
normal-phase LLC, the compounds would be eluted in a non-polar, volatile
solvent. Solvent evaporation would be the only step required before
mass spectrometry. In contrast, in reversed-phase LLC, the compounds
would be eluted in a non-volatile, mixed aqueous/organic solvent.
Extraction into a volatile, organic solvent would be necessary before
the samples could be concentrated to near-dryness.
Unfortunately, chromatography of PTN and TFN in a normal-phase mode
was not successful. The solvent was cyclopentane/tetrahydrofuran in
various proportions. Binding of the compounds to the stationary phase
(u-Bondapak CN) was too weak even in pure cyclopentane to enable this
system to be used as a final purification procedure. This result was
somewhat surprising, since partitioning of PTN and TFN between cyclo-
pentane and the cyanopropyl stationary phase of y,-Bondapak CN was
15
-------
expected to resemble partitioning of the compounds between hexane and
acetonitrile (see above), and thus to favor retention of the compounds
13
in a non-polar solvent such as cyclopentane. Despite its low polarity
and lack of miscibility with acetonitrile, cyclopentane is apparently
a sufficiently strong solvent for PTN and TFN to overcome the expected
solute-column interactions.
Reversed-phase LLC has been used successfully for the purification
of pesticides, including PTN. ' Mass spectrometric analysis of PTN
following reversed-phase chromatography in 30% aqueous methanol solution
indicated that the level of residual impurities, as measured by the
background, was similar to that found following 2—dimensional TLC. The
ratio of counts obtained at the unlabeled and labeled mass peaks of PTN
following extraction of labeled PTN from tissues of control rats, was
-4
in the 10 range. Similar ratios have been obtained with other compounds
11,12
following 2-dimensional TLC.
The chromatographic techniques used do not remove all impurities.
For instance, whether purification of PTN was by silica gel TLC or by
reversed-phase HPLC, major impurities were found in the mass spectrum of
PTN at m/e 297 and 298. These impurities do not interfere, however, with
the determination of PTN, which has a molecular weight of 291. Since
chromatography under the two sets of conditions is not expected to yield
a common impurity, this impurity is probably solvent-related, since
methanol was common to both purification techniques. It should be noted
that the solvents used were of the highest grade commercially available
(Burdick and Jackson Laboratories).
The background in TFN spectra, following reversed-phase HPLC in
30% aqueous methanol, was as much as four-fold lower for a given tissue
than was the case with PTN. No detectable structure characterized the
background in the TFN spectra obtained from diverse sources. With tissue
of control rats, the ratio of counts at the masses of labeled and unlabeled
TFN was in the 10 range. This indicates a very high level of purification
achieved by the reversed-phase chromatography of TFN.
16
-------
II C LXPERIMENTAL PROCEDURES
Animal Administration Regimens
Parathion IP
Rats were 50 day Sprague Dawley males (Simonsen Labs). Upon arrival,
the rats were weighed, ear notched and housed in groups according to
sacrifice schedule. The rats were allowed to acclimate for 3 days before
treatment began.
2.6 mg of parathion (Pfalz & Bauer Chemicals) was taken up in 1 ml
absolute alcohol and 25 ml sesame oil. This solution was injected by
body weight, 100 y,g/kg. Control consisted of 1 ml absolute alcohol
(Gold Shield) and 25 ml sesame oil.
Treatment consisted of a single IP injection. There were 5 groups
of 4 rats (1 control), sacrificed 1, 2, 4, 8 and 24 hr post-injection.
Fat, liver, brain, plasma, and red blood cells were collected from
each rat at the time of sacrifice. The tissues were immediately placed
in dry ice, the heparinized blood was immediately centrifuged, and the
plasma and RBC cells separated and placed in dry ice.
Animal and sample weights are presented in Table 1.
Trifluralin IP
Rats were 60 day Sprague Dawley males (Simonsen Labs). The rats
were allowed to acclimate for 4 days before starting the experiment.
There were five groups of rats:
Sacrifice Schedule -
Group No. No. of Rats Hours after Treatment
14 4
23 8
34 24
43 48
54 72
17
-------
o
ex
rH
§
SS
I-H
rH
s
"-/ CO
e
• in
rH CO
0 rH
^» cx
T3
•H
3 0
rH ,£2
fH rl
C
•H
CO
^^ rl
hh 5^
•H (1)
0) >
?~ -H
S ^
in
in
•H -P
H co
CM
...•
l> t- t- CO
....
r-H 1— 1 rH rH
in rH CO CM
o o o CM
O rH O CM
CO ^ ^Ji CM
O CO t>
• . • •
i — 1 CM ' — ' ' — '
O CO rH CO
....
rH CM CM O
rH CM •* CO
CO T)i CO CO
oo oo t*» to
CM (M CM CM
O O 0 O
to O 'I* CM
N 00 to to
CM CM CM CM
,
O (M tO •*
00 00 ID tO
CM CM CM IN
0
* -r1
rH CM CO C
O
O
0)
o
•H
<*H
•H
0
CO
in
a
CM
-f-
co to t> o
«...
rH rH rH CM
•* rH t> OO
tO CD l> 00
. • .
rH i-H 1— 1 rH
co CM m oo
• • to •
rH O -CM
co en CM co
CM CM CM CO
co to m o
CM (N (M CO
O O O O
to ^ to
t^ in ^51 i
CM CM CM
O (M CM <*
oo to in o
CM CM (N CO
i-H
O
£4
-p
rH CM CO C
0
o
0)
O
•rH
t> t> 0
•
t~ 00 CM CO
• • • •
rH rH CM rH
co CM co m
• . . I
O rH CM CO
a> CM to oo
CO CO CO ^
00 00 O CM
CM N CO CO
O 0 O 0
O CO O 00
t~ to CT> G}
CM w N ca
CM tM CM 00
CO 00 O rH
CM CM CO CO
rH
o
SH
rH CM CO C
o
o
0)
o
•H
I> l> 00
. . • •
rH r-H rH •— 1
to to h- co
* " "
o
00 ^ CD tD
. • . •
rH rH r-H rH
in rH rf CO
• » • •
•^ T}l rH ^
CM ^ CO CD
^ CO tO CO
rH N O O
' CO CO CO CO
o o o o
TJ< o O to
rH CXI CD CD
CO CO CM N
o o to to
i-l CM CD CD
CO CO CM IN
rH
O
rl
•P
rH CM CO C
O
0
V
o
•rl
m
•rl
!H
O
(3
tn
JH
"*•
S
SH
SH
•o
01
0)
•rl
IH
•p
eg
rl
in
co
^
c
•rl
CO
rl
O
cu
43
•P
MH
o
rH
rH
CO
•P
O
18
-------
Each rat received a single intraperitoneal injection of d -TFN, 500 M*g/kg
(500 Jig/ml solution of d -TFN in sesame oil). Groups 1, 3 and 5 contained
one control animal, injected with plain sesame oil, 1 ml/kg. Fat, liver
and plasma were collected from each rat at the time of sacrifice. The
tissues were immediately placed in dry ice and the heparinized blood was
immediately centrifuged, the plasma taken off and placed in dry ice. In
addition, urine and feces samples were collected from groups 3, 4 and 5
as below:
Gr. 3 0-24 hour samples
Gr. 4 24-48 hour samples
Gr. 5 48-72 hour samples
Animal and sample weights are presented in Table 2.
Trifluralin Oral
Rats were 60 day Sprague Dawley males (Simonsen Labs). The rats
were allowed to acclimate for 4 days before the experiment began. The
rats were housed 3 per cage in R3A cages and all animals received food
and water ad lib.
The rats were divided into 4 groups of 3:
Group No. Dosage
1 (Control) Vehicle*
2 20 |JLg/kg
3 100 |J,g/kg
4 500 |jg/kg
*
9g NaCl, 4 ml polysorbate 80, 5 gm carboxymethylcellulose,
9 ml benzyl alcohol
Each rat received the compound daily orally for 21 consecutive days.
An oral gavage needle (18 gauge from Popper and Sons, Inc. New York) was
used attached to a 1 ml disposable syringe. The rats were weighed and
the dosage adjusted to the correct amount with the d -TFN
solutions made up at 20 |0g/ml (Group 2) , 100 yg/ml (Group 3) and
500 jjg/ml (Group 4).
19
-------
CQ
.5
O
M
U
OH
rH
55
rJ
h
rH
e
£
§
/M
in
y
B
-/
y
B B
rH C
rH -H
O rl
0 3
CO
-P
y rH
w ^
T3 CO
rH CO
2
,J
in
in
•H rl
r -. fii
H B
rH
V
CO
r— 1
E
d)
bfl
co
in
0
Q
in
a
o
•P rl
x; y
bb Q)
•H C
Q)
r-> 0!
•§ '
CQ "^
r-H
+-1
a!
05
a
3
0
is
O
t- tO X ffi
c~ co <** m
^ ITS rf
t- oo m
CO CO CO
C' O O
01) 00 C<1
CO CO CO
^P *f 00
t-. 00 Tji
CO CO CO
r-( (N CO
B
y
•rl
rH CJ) rH
o in
• m m •
rH • • rH
rH 00 t- -H
oo m oo oo
O CO Tf t-
in Tjl rH CO
rH CM CM ID
m co M< m
• in «3 o
CO O CM TJI
W ^ JO JO
io m m JO
CO ^ t^« CO
CN CM •* CO
CO CO CO CO
o o o o
m > o JD
O rH r}l CO
CO CO CO CO
JO O JO ^
rH CM CO CO
CO CO CO CO
rH
o
rl
P
rH IN CO C
O
0
B
y
•rl
-------
brine and feces samples were taken from all groups on days 7, 14
and 21. The rats were placed individually in metabolism cages for 24
hours and returned to the R3A cage after the samples were collected.
The rats were sacrificed on the 22nd day after the treatment began.
Fat, liver, plasma, urine and feces were collected at the time of
sacrifice. Tissues were immediately placed in dry ice, the heparinized
blood centrifuged, and the plasma removed and placed in dry ice. Animal
and sample weights are presented in Table 3.
Extraction Procedures
Parathion: All tissues were kept on dry ice for 24 hours prior to
extraction. One gram each of fat, liver, and brain, one ml of plasma and
1.1 to 2.0 ml of packed red blood cells were used per sample.
One ml of 1.5 M phosphate buffer was added to plasma and one ml of
1.5 M phosphate buffer plus 10 cc distilled water was added to the RBC
samples to lyse the cells. All samples were homogenized in 8 ml of
hexane:ether/90:10, using a Tekmar Tissumizer Model SDT, for 30 seconds
at top speed, centrifuged (5 min, medium speed, clinical centrifuge),
and the organic layer filtered through a millipore filter. Some plasma
samples required ~0. 5 cc iso-amyl alcohol to break a heavy emulsion.
Twenty micrograms of d -PTN were then added as the labeled carrier to
each sample. Samples were then dried down at room temperature under
gentle nitrogen flow on a Meyer-N-Evap Organomation Analytical Evaporator
and redissolved in 2.0 ml of tetrahydrofuran (THF) in preparation for
the gel permeation chromatography.
Trifluralin: Fat, liver, and plasma samples were extracted using
the same method as for parathion. Urine was treated the same as plasma.
Feces were extracted as follows: One gram of feces was added to 8 ml
distilled water and one ml of 1.5 M phosphate buffer in a 125 ml
Erlenmeyer flask and allowed to stand for 30 min. to soften the feces.
21
-------
z
J
CO
•c
^-, CM
bC
-H CC
is -c
VJ «*
H> r—
o
0
>1
cc
r-
a
•c
•-I CM
e
• >>
o -c
>
i
•o
r~
lectecl
asma (ml)
0 Q.
U
13 ' "
H
3 *-.
» >
§ -
en
H
H 3
s
>. a
a d
o TJ
Bight
necropsy
&
>, a
S'
CD -H
V)
*
-H
CO
OS
P.
g
SH
0
^ CO C
LO x m
f-
• n uO
-i cn 01
O CM CM
t£> in Gl
0
o • o
• o •
t* CM C5
in in o
O :*: M
LO tO
• OJ O
t- -- -i
CO W N
(D -H <-*
0 *
• 0 0
CM
in in o
CM CM in
^ (D t^
m •-« o
m m ^
-< ro o
x m m
y5 ID LT
o o o
tD C O
t£ CD Ci
CO ro CO
CM 00 X
«> m rp
CO CO CO
•-* CM CO
bll
^4
4
O
o
Cl ^ CM
rj* co tn
to
• to t>
^ 01 m
— * m o
d 01 in
m o
- • o
o o
-* -H 01
m o
• m
co • m
0 0
CM • —H
i— i t* i— 1
t- 00 +
• 1
co <£) in
co m co
^H CO CM
to m ^
CO •? CM
CO CO CO
o c o
CO
-------
Eight ml of hexane:ether/90:10 were added and the mixture thoroughly
shaken by hand for aoout 30 seconds. This mixture was then placed in a
centrifuge tube and centrifuged for 5 minutes at top speed in a clinical
centrifuge. The organic layer was removed and filtered.
All samples received 80 micrograms of unlabeled trifluralin as
carrier, added just prior to homogenization-extraction, rather than
after filtration. All samples were dried down at 30 C under a gentle
N stream and redissolved in 2.0 ml of THE prior to GPC.
£
Gel Permeation Chromatography
All samples were stored in Teflon-capped culture tubes at 5 C until
gel permeation Chromatography. Exactly 0.5 ml of sample was injected
into the liquid chromatograph utilizing three 100 A microstyragel columns
in series. Conditions for operation were as follows: solvent - THF,
pressure - 1000 psi., flow rate - 1.25 ml/min, chart speed - 0.5 inch/min.
Parathion or trifluralin peaks were collected into another culture tube
manually using U.V. absorbance readings. Final volume was approximately
2.5 ml of THF. Figure 5 displays representative gel permeation chroma-
tograms for PTN. No attempt was made to identify the impurity components
of the chromatograms.
Reversed-Phase High Pressure Liquid Chromatography
Preliminary parathion samples were processed by both HPLC and thin
layer Chromatography following GPC. Since TLC appeared to give cleaner
mass spectra (see discussion, page 31), TLC was used for final purification
of the parathion samples.
All trifluralin samples were processed in a Waters Assoc. Model
M6000A HPLC with a Model 440 absorbance detector. The previously
collected GPC sample was dried under a gentle N stream as before and
£
redissolved in 0.5 ml H O-MeOH/30:70.
23
-------
FIGURE 5
PARATHION PURIFICATION
GEL PERMEATION CHROMATOGRAMS
SOLVENT — THF, 1.3 ml/mm
0.5 ml Injection
5 M9 d10-PTN Spike
ABS at 254 nm
ABS at 280 nm
0.32 O.D. FULL SCALE
PTN
PLASMA
BRAIN
PTN
FAT
/PTN
10 15 20
MINUTES
10 15 20
MINUTES
25 30
SA-4280-4
24
-------
Some fat samples would not dissolve in this solvent and were placed
in 100% MeOH, shaken until all the TFN was dissolved, centrifuged to
remove the fat globules, dried again under N , and redissolved in HO-MeOH.
2 £
Conditions for operation for the HPLC were as follows: Solvent 30% HO-
£4
70% MeOh, flow rate - 2 ml/rain., pressure 5000 psi., column - micro -
Bondapak C reversed phase, chart speed 0.5 inch/min. absorbance scale -
18
0.2.
The TFN peak was collected in a clean culture tube in approximately
2 ml of solvent to which sufficient NaCl was added to saturate the solution.
Two ml of cyclopentane were added to each tube which was then capped and
shaken by hand for 15 seconds. The trifluralin, now in the upper cyclo-
pentane layer, was then drawn off into a small culture tube and placed
dropwise on a small glass capillary tube in preparation for mass
spectrometric analysis. Figure 6 displays representative GPC (first
stage cleanup) and HPLC chromatograms for TFN.
Thin Layer Chromatography
Previously collected parathion peaks were dried under N , dissolved
^
in 10 drops of CHC1 :MeOH/l:l and spotted on an Anal tech TLC plate which
\j
o
had been previously washed in CHC1 : MeOH/10:90 and activated at 110 C
O
for cne hour.
Two dimensional TLC using hexane:acetone/85:15 as the first dimension
and benzene as the second gave single spots under U.V. light. These
spots were scraped onto a weighing paper, placed into a glass tube having
a wad of silanized glass wool for a filter and eluted with ~1 ml of
CHC1 :MeOH/l:l into a small glass tube. The sample was then ready for
•j
dropwise transfer to a small glass capillary preparatory to mass
spectrometric analysis.
25
-------
FIGURE 6
TRIFLURALIN PURIFICATION
GEL PERMEATION AND REVERSE-PHASE HIGH PRESSURE LIQUID CHROMATOGRAMS
20 /jg d7-TFN SPIKE
GPC- 0.5 ml Iniection
Solvent — THF, 1 3 ml/mm
-Abs. at 254 nm
Abs. at 280 nm
0.32 O.D. FULL SCALE
HPLC 0.5 ml Injection
Solvent — 30:70 H2O/CH3OH, 2 ml/mm
Abs. at 254 nm
0.2 O.D. FULL SCALE
INJ
10 15
MINUTES
5 10
MINUTES
SA-4280-5
26
-------
FIGURE 6A
TRIFLURALIN PURIFICATION (Continued)
PLASMA GPC
PLASMA HPLC
TFN
r
INJ
FAT GPC
, TFN
10
15
MINUTES
20
25
30 INJ
FAT HPLC
TFN
5 10
MINUTES
15
SA-4280-6
27
-------
Mass Spectrometer Isotope Ratio Measurements
Two sector magnet mass spectrometers were used to obtain the isotope
o
ratio measurements of parathioa and trifluralin. A 30 sector instrument
was used for the parathion and some of the trifluralin measurements. A
o
90 instrument was used for the remaining trifluralin measurements. Each
machine was equipped with an SRI-designed multipoint field ionization source;
the ionizing potential was 1500 volts. The accelerating voltage was scanned
repeatedly to span the parent ion region of the material of interest:
288-308 arau for PTN, 332-346 for TFN. Ion counts detected at the electron
multiplier were stored in a 1024-channel multichannel analyzer and the counts
integrated. When the count rate dropped to background level, the spectrum
was recorded and the total counts in each mass tabulated.
The sample was prepared by evaporating the solvent down to a few drops
with a gentle flow of dry N , which was then placed dropwise on a small glass
2t
capillary, allowing the remaining solvent to evaporate in air. The loaded
capillary was placed in a quartz solid sample probe and introduced into
o
the spectrometer through a vacuum lock. The sample was cooled to -5 with
liquid nitrogen (LN ) introduced through the back of the probe. The sample
£
was then allowed to warm up and the count accumulation started when the
average count rate began to rise above background. PTN came off in the
o o
10-20 C range; TFN in the 0-10 C range. The count rate was maintained at a
reasonable level by occasionally cooling with LN - A typical run took 10
£i
minutes. Memory effects were found to be negligible for both materials.
28
-------
Ill RESULTS AND DISCUSSION
Cleanup of Extracts from Animal Tissues by High-Pressure Gel Permeation
Chromatography
The cleanup of extracts prior to analysis is an essential step in the
determination of pesticide and herbicide residues in many animal tissues due
to the relatively large amounts of lipophilic substances that may be co-
extracted. Conventional methods, summarized recently by Burchfield and
Storrs, generally involve partitioning between polar and non-polar organic
solvents and/or adsorption on columns of charcoal, Florisil, or other materials.
Batchwise or continuous elution yields the pesticide or herbicide in a
sufficiently pure state to permit subsequent analysis.
When applied to large numbers of samples, "chese methods are slow and
15
labor-intensive. Recently, Stalling et al. described a gel permeation
procedure for the cleanup of fish extracts that appeared to achieve a fairly
clean separation between small pesticide molecules and the bulk of the co-
extracted lipids. An even more effective separation could be expected with
the aid of highly efficient, microparticulate column packings. A procedure
involving high-pressure GPC on columns of |j,-Styragel was developed for the
present study.
The major advantages of the high-pressure GPC method over conventional
methods are:
1. Minimal methods development time is required, since separation depends
almost solely on molecular size;
2. Cleanup effectiveness is practically independent of sample size for
volumes smaller than 1 ml;
3. The apparatus can be used repeatedly for large numbers of samples
without column replacement, since adsorption effects are negligible;
4. A high rate of throughput is maintained, typically 30 min. per sample;
29
-------
5. Several thousand theoretical plates obtained with microparticulate
columns result in high resolution separations.
Following high-pressure; GPC, samples can be analyzed immediately by
a variety of chromatographic techniques, or they can be further purified
for subsequent analysis by mass spectrometry. Low level determinations of
pesticides and herbicides in animal tissues by isotope dilution will
generally require that additional purification procedures be executed.
Both 2-dimensional TLC and reversed-phase HPLC were used in the present
study, although other purification techniques, particularly GC, might be
advantageous when dealing with volatile compounds.
Analysis of Mass Spectrometric Data
Calculation of Isotope Ratio:
The general formula for calculating the concentration, L (g/g or g/
ml sample), of PTN or TFN in a sample is:
C ~ C
L = m B . S
CM
where C = ion counts in minor mass (that of the material originally
m
present in the sample)
C = background counts in minor mass (sum of instrument background
B
and of the contribution of trace impurities with the same
molecular weight)
C = counts in major mass (that of the isotopic multilabeled carrier;
M
background correction is negligible)
S = grams of carrier added per g or ml of sample
For PTN the carrier was d -PTN, parent ion mass 301 amu (M+9 to
M+12 are summed). The material to be assayed was unlabeled PTN, 291 amu.
For TFN the carrier was unlabeled TFN, parent ion mass 335 (M to M+2
are summed). The material to be assayed was d -TFN, 342 amu.
30
-------
Detection Limits:
The ultimate sensitivity of the isotope dilution technique is
limited by the background at the minor mass. Total background is the sum
of: isotopic background (residual unlabeled material in the carrier),
chemical background (impurities at the same mass which carry through from
the tissues or are present in the solvents), and instrumental noise
(electronic noise and ion scatter).
As was discussed earlier, the high level of residual d -TFN in the pure
d -TFN would have produced an unacceptably high isotopic background at
mass 335 if d -TFN was used as carrier. Therefore, the reverse dilution
assay was performed since the amount of naturally occuring M+7-TFN is nil.
The amount of d -PTN in the labeled PTN was at the limit of detection.
Inspection of the PTN spectra suggests that the chemical background in
the M region is significant, since at each mass there is a definite peak.
There are notably accentuated impurity peaks at M+7 and M+8, as well as
a peak at M+5 present in the carrier (presumably PTN in which only one
ethyl group is deuterated) . Figure 7 shows mass scans for a PTN sample
purified by reversed-phase HPLC and by 2-dimensional TLC. Since the
chemical background is significantly lower following TLC, we selected this
method for the second-stage purification. Note that d -PTN is clearly
present in the TLC sample, but not significantly above background in the
HPLC sample. Some chemical background was still present even in the TLC-purified
samples, suggesting that a more powerful purification procedure could
improve the ultimate sensitivity of detection of PTN.
Figure 8 displays mass scans of PTN plasma samples purified by TLC
in which PTN peaks are clearly evident.
For TFN, however, there is an essentially "flat" background in the
minor mass region. Figure 9 shows three representative control spectra
for the TFN IP series and corresponding samples in which levels of TFN
were zero (< 1.1 ng/g) , medium (11.5 ±1.5 ng/g) and high (584 ± 5 ng/g).
31
-------
FIGURE 7
FIMS SCANS—PARATHION
(LOG DISPLAY)
COMPARISON OF HPLC AND TLC PURIFICATION TECHNIQUES
(2 HR PLASMA—100 ng/kg PTN IP)
M + 10
HPLC
Bkg level — 10 x 10~4
Bkg level (Average counts in M - 2, M + 1, M + 2) (counts in M + 9 — M + 12.)
SA-4280-7
32
-------
FIGURE 8
FIMS SCANS- -PARATHION
(LOG DISPLAY)
100 ,ug/kg PTN IP
CONTROL PLASMA
Bkg level 86 x 10'4
,M + 10
M + 10
1 HR PLASMA
141 ± 5 ng/ml PTN
2 HR PLASMA
44 ± 2.5 ng/ml PTN
M + 10
SA-4280-10
33
-------
FIGURE 9
FIMS SCANS—TRIFLURALIN IP
(LOG DISPLAY)
REPRESENTATIVE SPECTRA SHOWING UNDETECTABLE, SLIGHT,
AND HIGH d?-TFN VERSUS CONTROLS
PLASMA CONTROL
Counts M + 7
Counts M
24 HR
- 7 x 10~-'
,M + 7
LIVER CONTROL — 24 HR
Counts M + 7
Counts M
3 x 10-5
PLASMA -- 24 HR (500
0 ±1 1 ng/g TFN
IP)
FAT CONTROL — 24 HR
= 5 x 10
,-5
,M + 7
,M
LIVER — 24 HR (500 M9/k9 IP)
11.5 ±1.5 ng/g TFN
M + 7
-M
f\ FAT — 24 HR (500 /ug/kg IP)
/ ', 584 ±5 ng/g TFN
M + 7
SA-4280-8
34
-------
Figure 10 shows representative TFN spectra for each tissue in the TFN
oral series at the 500 yg/kg feed level. The limit of detection of TFN
was determined by the dynamic range attainable with the spectrometer,
which is dependent on ion scatter and electronic noise.
Background Correction and Error Analysis:
For each individual mass spectrometric analysis, a minor mass
background count value, C , must be derived. The overall chemical back-
B
ground "profile" in the minor mass region will differ for each material
and tissue. Total counts over the entire scan are a function of sample
size, which depends on the aggregate yield of the extraction, cleanup
and purification steps. Instrumental noise varies according to the
instrument tuning and the condition of the ion source. Consequently, in
computing the level of material in a sample we cannot simply subtract
an average background count figure taken from the control spectra. We
must instead obtain an "internal" background value for each scan by
inspection of the background level at adjacent mass peaks which bracket
the minor mass of interest. Ideally, the chemical background contribution
at these mass peaks should be completely absent, with only the
unstructured electronic and scattered ion background contributing to the
total ion count.
Control samples for each series were used to derive a correction
factor by use of which a value of C could be calculated for each spectrum
B
according to the scheme described below.
An average background count value, C , was obtained for each scan
nn
by averaging the counts in the two masses which bracket the minor mass.
For PTN the counts in masses 290 and 293 (M-l, M+2) were averaged. For
TFN, masses 341 and 344 (M+6, M+9) were used.
35
-------
FIGURE 10
FIMS SCANS—TRIFLURALIN ORAL
(LOG DISPLAY)
TISSUES AND EXCRETA
500 ug/kg d7-TFN
ORALLY FOR 21 DAYS
M
URINE
< 4.3 ng/g Tf:N
A
*»*%U*vwU«^«_^_
PLASMA
< 1.9 ng/ml TFN
FECES
75.7 z. 5.2
SA-4280-9
36
-------
An iverage background correction factor, F , was obtained for each
control series for a particular material and tissue. For a particular
control spectrum,
C
m
F =
°AB
F for n control spectra.
n ' n
and
n
This factor allows us to account for the chemical background
relative to that in nearby masses for a particular material and tissue.
For example, the background at the minor mass in the PTN red blood cell
control samples was on average higher than that in the nearby masses
(F = 1.3); in PTN plasma, however, the minor mass background was lower
(F = 0.65). Using the background correction factor we can thus "inter-
nally" compute a value for C for each scan which properly reflects the
6
background profile in each tissue, and which accounts for sample size and
instrument noise variations reflected in the absolute magnitude of C :
C = F . C (for each spectrum in a particular
B AV AB
material/tissue series)
The standard deviation in the value for F , a , was also computed for
each control set. This reflects the variations in chemical background
level due to variations in the extraction and purification procedures.
Using the equation
C - C
m B . s
C
M
we may then calculate the level of PTN or TFN in a particular sample in
ng/g or ng/ml.
37
-------
The variance in a particular measurement depends on the variance in
F , the statistical variance in the minor mass and major mass count
values, and on the volumetric error in the carrier spike. Counting
error in the major mass is small relative to that in the minor mass and
can be neglected. The standard deviation in the counts at the minor
mass, Oi'U, was computed by the formula,
r 2, 1/2
m ~ [_ m AB F AV
which expresses the variance due to counting error in the minor mass and
due to variance in chemical background in the controls.
Standard deviation in the measured concentration of PTN or TFN in a
sample, CT ,is then expressed as
L
CT
CT = m . L
L C - C
m B
Each value for the level of PTN or TFN in the tables in the following
section is expressed as L ± CT ; ov does not include the volumetric error
in the spike, which is of the order of 1-2 percent.
In some scans, the number of counts in the minor mass, C , was less
m
than or equal to C , which corresponds to a "negative sample". This
B
impossibility arises because of the uncertainties in F , computed from
/» V
the controls. In those instances, the measured level is expressed as
0 ± CT' where
CT
CT = m . S, CT' being the limit of detection
L - L
m
There are also many marginal cases near the detection limit—below 5 ng/g
or so—where L ^ CT •
L
38
-------
In those series where significant amounts of material were found,
e.g., liver and fat TFN-IP, a grand average was computed for each set of
measurements (3 rats, replicate spectra). The error limits placed on
these values reflect both animal and measurement variances, the former
being predominant. Average detection limits were computed for those sets
where the material was not found to be present.
Determination of PTN in Rat.Tissue by Isotope Dilution
The measured levels of PTN for the 100 |4g/kg IP regimen are displayed
*
in Table 4, and the results summarized in Table 5. Replicate scans \\ere
obtained for some samples. Detection limits differed among the tissues
because of differences in chemical background. Liver and brain were the
cleanest samples, as traces of PTN could be identified at the 1-2 ng/g
level. Chemical background was higher for fat, plasma and red blood cells;
PTN below the 5 ng/g level could be identified in only one sample (2 hr
plasma, rat #2).
Because the PTN levels were near or below the detection limits in
the 2 hr samples (with the exception of one anomalous fat sample) we
analyzed for PTN in only a few samples collected at later times. No
PTN was found in these samples.
A number of methods have been applied for the determination of PTN
7
in animal tissues. Gas chromatography with alkali flame ionization
detection or electron capture detection appears to be among the most
sensitive analytical procedures currently being used for this compound.
Quantitation of PTN down to about 50 ng/ml of plasma is feasible by the
addition of methyl parathion as an internal standard.
With isotope dilution analysis and FIMS, quantities of PTN down to
about 2 ppb were measured in the present study. The sensitivity was
limited primarily by the chemical background in the mass spectrum,
following purification by 2-dimensional TLC or reversed-phase HPLC.
Similar impurities were observed in the mass spectrum of PTN after both
chromatographic techniques had been applied. This observation suggests
*
Raw count data are compiled in the Appendix
39
-------
TABLE 4
MEASURED PARATHION LEVELS IN RATS (ng/g or ng/ml)
SINGLE DOSE, 100 yg/kg PTN IP; RATS SACRIFICED AT 1 AND 2
(3 RATS; MULTIPLE MASS SPECTROMETER RUNS)
LIVER Rat #1
#2
#3
FAT Rat #1
#2
#3
PLASMA Rat #1
#2
#3
BRAIN Rat #1
#2
#3
RBC Rat #1
#2
#3
I HR
0 ± 1-2
0.2 ± 1.2
0 ± 1.2
0.2 ± 2.3
0 ± 3.9
0 ± 3.1
27.0 ± 3.9
141.0 ± 5.0
0 ± 4.3
3.9 ± 1.2
2.7 ± 1.1
2.1 ± 1.2
2.6 ± 1.3
7.1 ± 1.2
0 ±13.5
0 ±3.6
0 ±3.5
0 ± 1.5
0 ± 1-4
1.7 ± 1.9
0 ± 6.0
3.6 ± 1.2
2.0 ± 1.3
1.9 ± 0.9
14.2 ± 1.2
0+± 10.5
0 ± 3.0
0 ± 3.8
2 HR
3.7 ± 1.1
1.8 ± 1.5
2.5 ± 1.2
0 ± 3.0
170 ± 7
9.6 ± 2.8
0+± 10.1
4.4 ± 2.5
1.3 ± 4.1
0.2 ± 0.7
9.4 ± 1.4
10.0 ± 1.6
1.2 ± 1.9
0 ± 4.2
0 ± 4.8
0 ± 4.2
0.3 ± 1.0
0 ± 1.3
4.7 ± 1.4
*
152 ± 5
5.0 ± 1.6
9.3 ± 1.0
1.0 ± 0.6
0 ± 4.1
0 ± 4.6
Possible accidental injection into fat; see Discussion
High background contamination
40
-------
Table 5
SUMMARY
PARATHION IP
Single dose, 100 y,g/kg IP; rats sacrificed at 1 and 2 hr
post-injection.
LIVER: 1 hr— None detectable, < 1.5 ng/g (average 3 rats)
2 hr— Trace, 2.2 ± 1.2 ng/g (average 3 rats)
FAT: 1 hr— None detectable, < 4 ng/g (average 3 rats)
2 hr— None detectable, < 3 ng/g (rat #1)
161 ± 9 ng/g (rat #2) Possible improper injection into fat
9.6 ± 2.8 ng/g (rat
PLASMA: 1 hr—27.0 ±3.9 ng/ml (rat #1)
141 ± 5 ng/ml (rat #2)
None detectable, < 4 ng/ml (rat #3)
2 hr—None detectable, < 6 ng/ml (average 2 rats)
Trace, 4.4 ± 2.5 ng/ml (1 rat)
BRAIN: 1 hr—10.6 ±0.8 ng/g (1 rat)
Trace, 2.6 ± 0.5 ng/g (average 2 rats)
2 hr—9.6 ± 0.8 ng/g (1 rat)
Trace, 1.8 ± 1.1 ng/g (average 2 rats)
RED BLOOD CELLS: 1 and 2 hr—None detectable, < 4 ng/g average
41
-------
the possibility that a solvesnt common to both methods was the source
of the impurities. Methanol was used both for elution of PTN from silica
gel and as a solvent in reversed-phase HPLC.
It seems probable that higher sensitivities of detection might have
been achieved in the case of PTN by the use of gas chrornatography for
purification. Our laboratory is not, at present, equipped for routine,
preparative gas chromatographic separations. We intend, however, to
develop this capability duri.ig the coming year for pesticide applications.
The results of the investigation of PTN levels in different tissues
of rats following IP injection at a dose of 100 fjg/kg indicate that the
pesticide disappears rapidly from plasma. Low but significant levels
were found in the brain tissue of all rats sacrificed 1 and 2 hr post-
injection. One of three rats killed at 2 hr post-injection appeared to
have low levels of PTN in both liver and fat; evidence supporting the
presence of PTN in the liver and fat of the other two animals was not
compelling (see below).
The rapid rate of disappearance of PTN from rat plasma is consistent
16
with the findings of Vukovich et al. Following intravenous injection of
rats with 4 mg/kg of parathion, these investigators determined the PTN
levels in plasma for times up to 90 minutes. An average half-life of
about 50 minutes for PTN in rat plasma can be estimated from their data.
This finding seems to agree with our own results, since it implies that
the PTN concentration would have fallen to an essentially undetectable
level by about four hours post-injection. Unambiguous determinations of
PTN were made in two rats at 1 hr and in one rat at 2 hr, but not in the
samples collected subsequently.
The presence of PTN in brain at 1 and 2 hr post-injection is
consistent with rapid equilibration of the pesticide between that organ
and plasma. This is not surprising, since the brain is well-endowed with
17
vascular tissue and readily absorbs lipophilic substances. A similar
42
-------
result T"ould be expected for liver, since it also is a well-perfused
organ, but the experimental findings indicate that the levels of PTN in
liver were negligible in most cases. It is probable that the high metabolic
activity of liver is responsible for the virtual absence of unchanged PTN
in that organ.
Other workers have established a complex metabolic pathway for PTN
18-25
in mammalian systems. In cases in which heavy doses of PTN were
ingested, sufficient to cause death, it has been possible to demonstrate
26-28
high levels of PTN in human liver at autopsy and in rat liver. In
these unusual cases, however, it is likely that the normal metabolic
pathways, which depend on an 0 supply, were essentially inoperative.
£i
It is somewhat surprising that PTN was not detected in any of the
1 hr fat samples, when it is still present at significant levels in
plasma. No positive explanation can be given for the absence of PTN in
most of the 1 and 2 hr fat samples, although slow uptake due to the low
degree of vascularization of adipose tissue, together with the rapid
disappearance of PTN due to metabolism and reaction, is probably responsible.
One rat sacrificed at 2 hr did exhibit a very high level of PTN in fat.
This case seems, however, to be anomalous, and could have been caused
by accidental injection of the PTN directly into the fat tissue, in
which it would be expected to remain as a deposit for an extended period
of time.
The disappearance of PTN in rats is evidently quite rapid, and the
compound seems to be distributed extensively among the well-perfused
tissues and organs of the animal. Storage of PTN in adipose tissue,
however, does not appear to be significant, at least at low dose levels.
This is encouraging from the viewpoint of exposure of humans or animals
to PTN, since it implies that gradual build-up of the compound to toxic
levels due to repeated exposures is unlikely.
43
-------
Although PTN has been found at generally low levels in tissues
following IP injection, it is quite probable that metabolites of PTN
would be present in the same tissues at much higher quantities. A
sensitive analytical procedure that assays for one or more metabolites
of PTN might employ similar methods to those used here. Such a procedure
could conceivably detect residues of PTN metabolites for extended periods
of time after exposure.
Determination of TFN in Rat Tissues by Isotope Dilution
The measured TFN levels for the 500 |jg/kg IP regimen are displayed
*
in Table 6 and the results summarized in Table 7. All samples were run
in duplicate. In the liver series, with TFN concentrations in the 5-50
ng/g range, the agreement between duplicate runs is good, in all cases
within the standard deviation limits calculated for each value. Average
relative deviation from the mean is ~7%. In the high level fat series,
agreement is excellent; average deviation from the mean is ~2%. (There
is one exception, 4 hr fat, rat #1, in which the discrepancy is 22%; this
may be due to a mistake in recording the counts).
Animal variation in measured TFN value is, however, considerable—
a predictable result.
One rat in the 8 hr set had no measurable TFN in liver or fat, most
probably because that animal was mistakenly not injected with TFN. This
rat was omitted when calculating the averages.
The measured TFN levels for the daily oral administration of TFN at
*
3 levels are displayed in Table 8 and summarized in Table 9. TFN was
found in only a few samples. Because of time constraints, and because the
reproducibility of the mass spectrometric determinations was already
established by the TFN IP measurements, only those samples in which TFN
was present were run in duplicate.
Raw count data are compiled in the Appendix
44
-------
I!
H B
CXI
W
rH
55
t-H
s
00
w
H
- CJ
S rr
M tj
r^
B
Q
Ed CO
CJ 55
to
W
M M
c os os
^ o w
< EH
CO CO W
«! co O
OS EH OS
< H
55 OS CJ
rH W
•- a*
CO On CO
> 55 CO
W fc, <
J H S
55 bfl •-
M ^ CO
J \ H
J O CO
PTJ O ^^
M in
OS
EH -
w
Q co
53 3
H O
os
w
CXI
t*.
OS
a
00
as
X
CM
OS
SM
00
os
«
?
rH tO CM 00 tO CO
-H -H -H -H -H -H
CXI O CXI O 00 rH
. • • •
CXI O CM CXI
rr cxi co in rr to
rH rH rH rH rH rH
-ft -H -H ft -H fl
oo to en co rH in
O rH t- tO 00 t~
r- to t~ in ,-H rr
rT rH rT rH in rH
-H -H ft +1 -H ft
rr cxi rr in co ,-H
rH co cxi rH o en
CO IN rH rH l-H
rH t- 00 CXI 00 CXI
CM rH CXI CXI rH CXI
-H -H -H +1 -H -H
* *
o o rr cxi co cn
• • • •
co co oo i~-
CXI CM
in in rH o o oo
in in to t~ to t-
-H -H -H -H -H -H
O •* CM O CM rr
aa t~ co cxi o oo
rr co in rr to to
rH CXI CO
ft ft ft
•p -p -p
c3 cfl cJ
OS OS 03
OS
£>
M
rP
o
oo cn cxi co i> cn
-H +1 -H fl +1 ft
in rH CO CO in rH
CO O) • • CXI CM
cn oo rH to t> t^
rH rH 00 00
to o t*
00 rH in m rH rH
-H +1 -H H fl H
t~ r- rf to cxi CM
rr to CM cxi oo o
00 00 tO tO CM CO
rH rH
in tO rH CXI rH rH
ft -H ft ft fl H
I* 00 CO rH tO CO
00 t^ CO O] 00 rT
in in oo to co co
rH rH CXI CXI
lO t**
• • CJ) CO
co n r* m ,H ,-n
-H -H -H -H H H
u
CXI O rH rH t- CXI
to m cn co
CM CXI CXI tO tD
rH rH 00 rH 00 rH
-H -H -H "H -H -H
oo to co co rr cn
t^ •* CM cxi rr co
rH CXI CO
ft ft ft
•P -r> -4->
03 ci rt
os os os
H
rt<
o o oo cn oo o
rH rH O O O rH
H ft ft +1 H ft
O O O O O O
O 00 t» f- rH t-
CO CXI O CM rH CM
H fl H H H fl
O O 0 O O O
rH CT) oo cn cxi to
rH O O O rH rH
fl ft ft H fl H
00 00 O O
CM O O CM 00 tO
co rr rr rH CM cxi
H -H ft fl H H
* *
o o en o o o
in
00 t- tO CO (DO
CXI CM CM CO CXI rT
-H -H -H -fl -ft -H
f- O O O O O
-r
rH CXI CO
ft ft ft
•p -p +->
rt cd c3
OS OS 03
s
w
t- tD t~
rH O O O O O
-H -ft -H -H -H -H
o cn o o o o
O rH rH
to oo oo oo o cn
O O O O rH O
-H -H -H +1 -H -H
in tO rH O t- t»
d o o o o
rH rH rT rH rH t-
CM CXI CM CXI CM CM
-H -H +1 H -H +1
cn to co tn rH rH
O CXI CXI in O rH
rH CXI CO
ft ft ft
•P -P -P
cJ nj cd
os as os
w
rH
OS
o co cxi cn to
rT CXI rH rH f-
fl H H H H
o cxi in co oo
co d rH cn
t~ oo t- cn oo to
o o o o o o
•H ft H ft H +1
in o o o c- o
d ' d
rH t^ CO rT Cn rH
CO CXI rH CO CXI rH
-H -H H -H H -H
m o o o to o
in d
i
*
-t
1
4
rH CM , CO
+->+-> V
oj aj 03
os os os
w ••
CJ c
w
En •*
45
-------
TABLE 7
SUMMARY
TRIFLURALIN IP
Single dose, 500 y,g/kg IP; rats sacrificed at 4, 8, 24, 48,
72 hr post-injection.
LIVER: 4 hr—51.5 ±9.6 ng/g
8 hr--15.7 ±7.7 ng/g
24 hr—16.3 ± 9.7 ng/g
48 hr— 5.4 ± 3.0 ng/g
(average 3 rats)
(average 2 rats)
(average 3 rats)
(average 3 rats)
72 hr— None detectable, < 2.4 ng/g (average 2 rats)
Trace, 2.5 ± 1.0 ng/g (1 rat)
FAT: 4 hr— 417 ± 138 ng/g
8 hr— 461 ± 205 ng/g
24 hr—1558 ± 737 ng/g
48 hr— 925 ± 274 ng/g
72 hr— 9OT ± 750 ng/g
(average 3 rats)
(average 2 rats)
(average 3 rats)
(average 3 rats)
(average 3 rats)
PLASMA: None detectable, < 2.8 ng/ml average
URINE: 24 hr— Trace, 2.1 ± 1.7 ng/ml (average 3 rats)
48 and 72 hr— None detectable, < 1.8 ng/ml average
FECES;
None detectable, < 1.6 ng/g average
46
-------
oo
o
o
1
rH
OT
w
w
g
ho co
s 5
w
a, ^
bfl Q
W
H
w
tD M
£ 1
5 ^
EH S5
§
FECES
URINE
^ri
^^
H
PH
os
rH
*
Q
rH
Q
rH
Q
t>
Q
rH
IM
Q
i-H
1
t>
t?- <5> C>
CXI rH CO
1 1 _44 _1_J
-j-j "Tl ~Tl
05 00 CXI
• » •
CXI rH N
-H 41 -H
O O O
O 00 00
00 CO ^
41 41 -H
t^ t^ ^*
C^ CO CM
41 41 41
O O O
<* co
to m rf
41 -H 41
rH 0 0
in
00 rH H
o • •
41 41 -H
to t>
o <• •
O CXI
to co 01
rH CXI CO
41 41 41
to o 01
rH "^
G} I> O
CM co co
-H 41 41
o o o
rH N CO
-P
d
OS
bO
O
(M
*
in in t^
CO (M CXI
•+* 41 "^
t^ CO
^^ ^5
t~ rH O
^f CXI CM
-H H 41
o O O
CXI CO O
• * •
(N CO (M
H -H 41
o"
co m "j1
rH rH CO
2 •" 5
o
0 O
in CO rH m rH
(M rH
O5 IM (M
-H 41 -H
O O O
rH CM CO
d
(A
bO
O
O
i-H
Jte « rH IM CO
C9 in
O
tO ^f N rH • N
. . O .
co (M N in IM m
41 41 41 41 41 41
N IM in rH rH rH
tD to rH CO 00 CO
CO CO t~
0
(M rH ff)
• • •
rH Cvl rH
-H 4) 41
CXI rH
O
O rH
•* 00 00 00 CO IM
00 !M rH CO CO CM
41 41 41 41 41 41
o in in m in oo
t- m in rH oo cs
i-H rH
CM
o <°. m.
CM c^ co
41 41 H
00°
rH CM CO
-P
d
1
o
0
in
o
•H
-P
«
G
•H
•H
§
O
2 3
bo
.* rH
O CO
CS +J
X2 O
* +
47
-------
TABLE 9
SUMMARY
TRIFLURALIN ORAL
Daily oral dose (by stomach tube) for 21 days at 20, 100 and
500 )j,g/kg levels. Rats sacrificed at 21 days. 24 br urine
and feces collected at 7, 14, and 21 days.
LIVER: None detectable at any level, < 3.2 ng/g average
FAT: 20 ^g/kg: None detectable, < 3.2 ng/g (average 2 rats)
Trace, 4.9 ± 3.9 ng/g (1 rat)
100 |j,g/kg: 9.1 ± 7.1 ng/g (average 3 rats)
500 M*gA.g: 9.6 ± 5.5 ng/g (average 3 rats)
PLASMA: None detectable at any level, < 1.9 ng/ml average
URINE: None detectable at any level/time, < 2.3 ng/ml average
FECES: 20 and 100 |j,g/kg: None detectable at any time, < 3.1 ng/g average
500 jag/kg: rat #1—6.2 ± 2.2 ng/g (7 days); 191 ± 6 ng/g (14 days);
29.5 ±5.0 ng/g (21 days)
rat #2— < 3.5 ng/g (7 days) ; 7.5 ± 2.5 ng/g (14 days) ;
61.3 i 4.8 ng/g (21 days)
rat #3—73.4 i: 5.2 ng/g (7 days); 2.6 ± 1.9 ng/g (14 days);
3.6 ± 3.3 ng/g (21 days)
48
-------
Conventional techniques for the detection or analysis of TFN
include TLC and GC. The latter can be made quantitative with the
29
addition of diisobutyl phthalate as an internal standard. The limit
of sensitivity of these methods, when applied to animal tissues, has
not been reported to our knowledge, but is probably similar to that of
PTN (50 ng/g). A novel method involving spark source mass spectrometry
30
has been described for the analysis of TFN in soils. This method
takes advantage of the relatively low fluorine background in soils; a
sensitivity limit of about 1 ppb is claimed. This method, however, is
non-specific.
A sensitivity of the order of 1 ppb was achieved in the present study
using isotope dilution analysis and FIMS. Following cleanup by high-
pressure GPC and purification by reversed-phase HPLC, background
impurities in TFN extracts were below the point of detection. Scattered ion
noise in the mass spectrometer set the limit of detectability at about
1 ppb in animal tissues. Identification of trifluralin is, moreover,
unambiguous in this method.
The scattered ion background in TFN samples was slightly higher
11
than has been found in some other applications of the method. This
was probably due to an instrumental alteration that preceded the analyses
of TFN. A retarding lens that is normally situated near the detector had
been removed to allow some other instrumental modifications to be made.
In prior applications, this lens had reduced the scattered ion background
by a factor of 5 without significantly reducing the signal. An improve-
ment in the signal/noise ratio by this factor would allow analyses of
pesticides and herbicides to be performed in the subnanograra range.
Following IP injection of TFN (500 ^jg/kg) , significant levels were
found in liver for at least 48 hr post-injection. High levels of TFN
were found in fat at 72 hr post-injection, the longest time that was
studied. No detectable unchanged TFN was present in plasma or feces at
any time up to 72 hours. Trace levels of TFN appeared in urine during the
first 24 hours, but not at later times.
49
-------
For pharraacokinetic purposes, it is generally assumed that the
well-perfused tissues, e.g., liver, lung, heart, spleen, and brain,
constitute together with plasme. a single "compartment" within which a
drug or other foreign compound is equilibrated rapidly according to its
relative affinities for the different tissues in the compartment (see
for example, discussions in refs. 31-33). Poorly perfused tissues, such
as muscle, bone, and fat depots, constitute a second "compartment".
Interactions between these compartments, together with metabolism and
excretion, define the kinetics of the compound in question.
The compartmental hypothesis is an over-simplification, since
equilibration among the different tissues of a compartment is not
instantaneous. Nevertheless, surprisingly good agreement among different
33
tissues has been found in the kinetics of at least one compound when
the levels were determined in several different organs of rat.
To the extent that the liver may be regarded as representative of
perfused tissues in general, and fat as representative of the less well-
perfused tissues, the terminal half-life of TFN in the rat can be
estimated (Fig. 11). For the two tissues, the estimated terminal
half-lives are 20 and 25 hours, respectively. Considering the paucity of
data, the assumptions involved Ln the model, and the fact that the lines
were drawn visually rather than by least-squares, the agreement between
these values, although predicted, is perhaps fortuitous. These results,
in essence, provide the basic information needed to undertake a more
thorough study of the persistence, kinetics, and distribution of TFN in rats.
The major identified metabolites of TFN in rats result from dealkylation
34-36
of the N-propyl group and reduction of the nitro group(s). Since one
of the two propyl groups of the TFN injected into the rats was labeled
with deuterium, it is probable that the rate of metabolism of the injected
compound via N-dealkylation is riot exactly equal to the rate of metabolism
of unlabeled TFN. The half-lives of the two isotopic variants should not
50
-------
FIGURE 11
TRIFLURALIN CONCENTRATIONS IN ADIPOSE TISSUE AND LIVER OF ADULT MALE
SPRAGUE-DAWLEY RATS FOLLOWING INTRAPERITONEAL INJECTION WITH
TRIFLURALIN. DOSE 500 ^ug/kg EACH POINT REPRESENTS THE AVERAGE VALUE
OBTAINED FROM 3 RATS ERROR BARS ARE ESTIMATED STANDARD DEVIATIONS.
10,000
1000
(T
I-
Z
LU
u
p
100
tr.
36 48
TIME -- hours
SA-4280-12
51
-------
differ substantially, however, because (1) only one of the two propyl
groups was labeled with deuterium, and (2) N-dealkylation accounts for
a relatively minor fraction of the total metabolites produced in rats.
Following chronic oral administration of TFN at three different
levels to rats (20 pg/kg , 100 ug/kg, and 500 yg/kg) for 21 days, the
compound was detected both in fat and feces at the higher dose levels,
but was not found in liver, plasma, or urine. The levels found in fat
were low in comparison to the results of IP injection, and the levels in
feces were also low in comparison to the quantities administered.
A similar study was carried out several years ago by Emmerson and
34
Anderson, in which rats were chronically fed high doses of TFN (100 mg/
14
kg), radiolabeled with C. These authors also observed very little
unchanged TFN in the feces (< 3% of an oral dose), but found that
approximately 70% of the radioactivity was eliminated as metabolites in
feces, the remainder being excreted as metabolites in urine. They
observed only low levels of TFN in fat, following chronic oral toxicity
studies.
The results of our study are seen to be in substantial agreement
34
with those of Emmerson and Anderson. The latter authors postulated
that TFN is poorly absorbed from the gut and is extensively metabolized
within the intestines, probably by micro-organisms. Such an hypothesis
provides a reasonable explanation for the present findings. The oral route
of TFN administration evidently results in little systemic absorption of
the unchanged herbicide. However, the IP studies suggest that, once
absorbed, the compound remains in tissues for rather lengthy periods of
time. A study of the long-term persistence of TFN in these and other
tissues of rat could be undertaken readily using the techniques developed
in this research program.
52
-------
IV DETERMINATION OF THE CONCENTRATION
OF PTN AND MePTN IN AEROSOLS
The concentrations of parathion and methyl parathion in the inhalation
37
exposure chambers in the Toxicology Department of SRI were measured by
an isotope dilution technique similar to that used to measure the concentra-
38
tion of explosive vapors in air. There are four inhalation chambers in
series; the material was dispersed in an aerosol by a nebulizer (methyl
parathion 80% in xylene) and the airflows adjusted to reduce the concen-
tration by a factor of two in each succeeding chamber in the line. Running
conditions were selected to match those of previous toxicology experiments
with these materials.
Experimental
The chamber air was sampled with 50 ml glass bulbs with Teflon stopcocks.
The stopcock was removed and 0.5 ml of a THF solution of the isotopic diluent
added through a flexible Teflon tube attached to a syringe. 136 pg of d -PTN
(determined by absorbance) or 100 ^g of d -MePTN (determined gravimetrically)
were loaded into the sample bulb. The solvent was evaporated by a stream
of dry N and the stopcock replaced. The bulb was then immersed in LN
2 - 2
and evacuated on a small vacuum line.
The evacuated bulb was attached to the sampling port of the inhalation
chamber, allowed to equilibrate for 5 minutes, the stopcock opened briefly
to sample the air, and then closed.
After a 10 min. wait to allow the sampled air to equilibrate with the
multilabeled carrier, the stopcock was removed and the bulb interior washed
out with two 1 ml portions of spectrograde THF. , introduced and withdrawn
with a syringe equipped with a flexible Teflon tube. The solvent was
evaporated off and the sample transferred to a small glass capillary for
mass spectrometer ratio determination in the previously described manner.
No prior purification of the samples was attempted.
-------
The chamber air flows were set to give nominal concentrations of
3 3
250 mg/m PTN and 380 mg/m MePTN in chamber #1. These values were
based on glc determinations made' many months previously when these compounds
were being run on a routine basis. Flow split factor was 1/2. Duplicate
samples were taken from chamber #1, chamber #2 (1/2 cone. #1) and
chamber #4 (1/8 cone. #1). One sample was inadvertently lost in transfer.
Results and Discussion
The results of the inhalation chamber measurements are summarized in
Table 10. These is a large discrepancy between the concentrations as
measured by mass spectrometry and the nominal values previously determined
by collecting pesticide samples on nucleopore filters and analyzing by
gas chromatography. The measured values are consistently low by about a
factor of 6. However, the values are internally consistent. The factor
of 6 obtains for both materials, and the flow split factors of 1/2 and 1/8
are in good agreement with the measured values.
One possible explanation for the discrepancy could be that a major
portion of the air sampled in the bulb was insufficiently equilibrated with
the air flowing in the chambers. Each sampling port was connected to a 20 cm
long tube mounted in the conical upper portion of each chamber. Since the
air does not flow through this tube, the concentration of the aerosol may be
low. However, if the values are recalculated assuming that the sampling
tube and stopcock attachment "dead volume" contained zero aerosol, it would
only raise the measured values by a factor of ~1.6. The presence of any
impurities in the mass region of interest would contribute to the counts in
the minor peak (unlabeled) and give falsely high values.
Since the nominal values of the aerosol concentrations are based on
measurements made many months earlier, it is quite possible that the PTN and
MePTN used in the chambers had degraded considerably over time (they were
stored at room temperature). In order to elucidate the origin of the
54
-------
o
rH
3
CQ
^J
H
OS
w
w
OS
w
"S.
X
o
iz;
0
2
O
rH
K
<;
OS
a,
in
ra
0
^
E
O
o
0
o
rH
tfc
^
S
CO
u
o
fl
o
o
-a
0
3
in
OS
0
S
o
c
o
o
rH
^
•
S
03
£
O
O
c
o
o
rH
CO
G
•H
S
o
00 CO O
to to m
^-v y— V
lO rH
rH O O
CO
e
in t> co
CM O O
-H -H -H
m t> CM
...
CD OS to
CO rH
m
/-> CM
in rH
r^ O 0
CO
e
he in
£ CM
O in rH
in CM co
CM rH
rH CM ^
4t 4t -tt
.
&
£
CO
U
OS
w
CM OS C^
CM TJ<
m 1-1
CH 0 0
m in
CO O
-H -H
03
m o m
rH CM OS
to co
m
/•v CM
in rH
/^•N
rH 0 0
m
o a t~
oo os ^
CO rH
rH CM ^
:»t tfc tt
•
e
CO
&
o
g
CM
4)
in
3
rH
CO
>
o
r"H
bC
to
c
o
•rH
03
C
•H
£
rH
0
0
•O
rH
0
•H
rH
rH
03
0
G
0
•a
in
03
CQ
*
in
he
•H
rH
a.
s
03
in
CM
tH
O
o
M
03
rH
0
J>
<£
-(-
-P
in
O
rH
01
rH
a
s
CO
in
a)
c
o
a
55
-------
discrepancy, it would be necessary to repeat the experiment with fresh
PTN and MePTN, and with an improved sampling geometry, making GC
determinations of aerosol concentration at the same time that the mass
spectrometric samples are taken.
56
-------
V CONCLUSIONS
In the first year of this research program we have achieved the
following goals:
1) Synthesized multilabeled isotopic variants of parathion and
trifluralin, as well as methyl parathion;
2) Developed rapid and efficient methods for the isolation and
cleanup of pesticide and herbicide residues in animal tissues, based on
high-pressure gel permeation chromatography;
3) Worked out final purification methods for parathion and trifluralin,
using either TLC or reversed-phase HPLC.
4) Carried out analyses for these compounds on approximately 200
samples of rat tissues, plasma, urine or feces, most of which were done
in duplicate;
5) Made preliminary measurements of the concentration in air of
parathion and methyl parathion in the inhalation exposure chambers of the
SRI Toxicology department.
It can thus be concluded that the isotope dilution analysis methodology
works when applied to the kinds of problems posed by this program. There
are, however, a few points that require comment and that will be considered
as our program progresses.
The isotopic purity of the compounds was not uniformly high. In the
case of d -PTN, no detectable unlabeled material was present in the carrier.
In d -TPN , however, the d /d ratio was unacceptably high, caused apparently
by the presence of some unlabeled n-propyl bromide in the deuterated starting
material. The deuterated TFN was, therefore, not used as an isotopic
dilutant and was instead administered to the animals, while unlabeled TFN
served as the carrier. The kinetics of TFN and d -TFN metabolism are
7
probably not very different, for reasons already discussed. In other cases,
however, it may be necessary to rigorously avoid such occurences.
57
-------
We have not encountered any major problems in the extraction and
cleanup procedures. High-pressure, multiple-column GPC proved to be
an efficient cleanup method for a variety of tissues, and it is expected
that it will find numerous applications in pesticide residue analysis.
A sequential HPLC method utilizing both GPC and reversed-phase LLC
was very effective in purifying TFN samples for mass spectrometric analysis.
This method was not, however, as successful when applied to PTN samples,
and for this reason, TLC was used as the final purification step. It is
likely that a variety of chromatographic techniques will be called upon as
different pesticides and herbicides are analyzed, and we intend to develop
a broad capability in this area.
The limiting factors in the analyses of PTN and TFN were chemical
impurities in the samples and noise due to scattered ions in the mass
spectrometer, respectively. Both of these can, in principle, be reduced.
A decrease in the chemical background of PTN can be achieved either with
the use of higher purity solvents or by a more efficient chromatographic
technique, such as gas chromatography. The scattered ion background can
be reduced by addition of a retarding lens near the detector.
The biological information that was obtained can be summarized as
follows:
PTN was detected at low nanogram levels in plasma and brain of rats
at 1 and 2 hr post-injection with 100 [jg/kg of the insecticide, but was
not found in the liver, fat, or red blood cells of most of the animals. Unchanged
PTN was not detected in tissues of animals sacrificed at later times. These
data suggest that PTN undergoes rapid metabolism or reaction in rats and is
essentially not stored as such in lipophilic depots.
In contrast, TFN was found at high concentrations in fat and liver
following IP injection (500 pg/kg). Traces of the unchanged compound also
appeared in urine during the first 24 hr post-injection. Determinations of
the fat and liver concentrations of TFN in rats at times as long as 72 hr
post-injection suggest a biological half-life for TFN of 20 to 25 hr.
58
-------
Daily oral administration of TFN at three dose levels (20 )jg/kg,
100 fjg/kg, 500 |ag/kg) for three weeks led to detectable uptake of TFN into
fat at the higher dose levels. Small quantities of TFN were also found in
feces during this regimen, but the compound was not detected in urine at
any time, or in plasma or liver at sacrifice (21 days).
The oral TFN data suggest that the compound is poorly absorbed from
and extensively metabolized within the gut. Once absorbed, however, TFN
remains in fat for an apparently lengthy time period, as suggested by the
IP data. Further study of the distribution and kinetics of TFN in other
tissues would provide valuable information concerning the toxicology of
this important herbicide.
59
-------
REFERENCES
1. T. W. Mastin et al , JACS 67, 1662 (1945).
2. J. H. Fletcher et al, J4CS 7j2, 246 (1950).
3. S. R. Sandier and T. Faro, "Organic Functional Group Preparations,"
Organic Chemistry, Academic Press, New York, 12, 325.
4. L. M. Yagupol'skii and V. S. Mospau, Ukrain, Khim, Zhur, 21, 81-5 (1955).
5. Q. F. Soper, U. S. 3, 257, 190, June 21, 1966.
6. H. P. Burchfield and D. E. Johnson, Guide to the Analysis of Pesticide
Residues, Vol. I, U. S. Dept of Health, Education, and Welfare, Bureau
of State Services, Office of Pesticides, Washington, D. C. (1965).
7. H. P. Burchfield and E. E. Storrs, J. Chromatog. Sci. 13, 202 (1975).
8. L. R. Snyder and J. J. Kirkland, Introduction to Modern Liquid
Chromatography, John Wiley & Sons, New York (1974), p. 95.
9. J. J. DeStefano and H. C. Beachell, J. Chromatog. Sci. 10, 654 (1972).
10. H. A. Moye, J. Chromatog. Sci. 13, 268 (1975).
11. J. H. McReynolds, H. d'A. Heck and M. Anbar, Biomed. Mass Spectrom. 2,
299 (1975).
12. H. d'A. Heck, J. H. McRejynolds and M. Anbar, Cell Tissue Kinet. (submitted).
13. L. R. Snyder and J. J. Kirkland, Introduction to Modern Liquid Chroma-
tog raphy , John Wiley, New York (1974), p. 217.
14. R. Stillman and T. S. Ma, Mikrochim. Acta, 491 (1973).
15. D. L. Stalling, R. C. Tindle and J. L. Johnson, J. Assoc. Offic. Anal.
Chemists 55, 32 (1972).
16. R. A. Vukovich, A0 J. Triolo, and J. M. Coon, J. Agr. Food Chem. 17,
1190 (1969).
17. B. B. Brodie, in Absorption and Distribution of Drugs, (T. B. Binns, ed.)
Williams & Wilkins, Baltimore (1964), p. 16.
18. F. W. Plapp and J. E. Casida, J. Econ. Entom. 51_, 800 (1958).
19. S. D. Murphy, Residue Rev. 25, 201 (1969).
20. M. T. Shafik, D. Bradway, and H. F. Enos, J. Agr. Food Chem. 19, 885 (1971),
60
-------
21. T. Nakatsugawa, N. M. Tolman and P. A. Dahm, Biochem. Pharmacol. 18,
1103 (1969).
22. H. T. Appleton and T. Nakatsugawa, Pesticide Biochem. Physiol. 2_, 286 (1972)
23. E. P. Lichtenstein, T. W. Fuhremann, A. A. Hochberg, R. N. Zahlten, and
F. W. Stratman, J. Agr. Food Chem. 21_, 416 (1973).
24. L. W. Whitehouse and D. J. Ecobichon, Pesticide Biochem. Physiol. 5,
314 (1975).
25. G. M. Benke and S. D. Murphy, Toxicol. Appl. Pharmacol. 31, 254 (1975).
26. G. Bohn, G. Ruecker and K. H. Luckas, Z. Rechtsmedizin 68, 45 (1971).
27. S. N. Tewari and S. P. Harplani, Mikrochim. Acta, 321 (1973).
28. A. C. Koutselinis, G. D. Dimopoulos, and Z. I. Smirnakis, Med. Sci. and
Law 1£, 178 (1970).
29. L. G. Hambleton, J. Assoc. Offic. Anal. Chem. 54, 125 (1971).
30. S. C. Tong, W. H. Sutenmann, L. E. St. John, Jr., and D. J. Lisk,
Anal. Chem. 44, 1069 (1972).
31. S. Riegelman, J. C. K. Loo, and M. Rowland, J. Pharm. Sci. 57, 117 (1968).
32. R. Nagashima. G. Levy and R. A. O'Reilly, J. Pharm. Sci. 57, 1888 (1968).
33. T. J. Benya and J. G. Wagner, J. Pharmacokin. Biopharm. 3_, 237 (1975).
34. J. L. Emmerson and R. C. Anderson, Toxicol. Appl. Pharmacol. 9, 84 (1966).
35. T. Golab, R. J. Herberg, E. W. Day, A. Po Raun, R. J. Holzer and
G. W. Probst, J. Agr. Food Chem. 1/7, 576 (1969).
36. J. R. Plimmer and U. I. Klingebiel, in Mass Spectrometry and NMR
Spectroscopy in Pesticide Chemistry (R. Haque and F. J. Biros, eds.),
37. G. Newell, "inhalation Toxicological Studies of Chemical Agents", SRI
Annual Report to EPA (Dr. R. L. Baron, Contracting Officer), on
Contract No. 68-02-1751, January 1976.
38. G. A. St. John, J. H. McReynolds, W. G. Blucher, A. C. Scott and
M. Anbar, Forensic Science 6, 53-66 (1975).
61
-------
Parathion-IP
1 hr Liver
control
APPENDIX
RAW COUNT DATA
89
245K
•"AB
124
Spectrometer
all 30° sector
rat #1
#2
#3
2 hr Liver
control
rat #1
#2
#3
1 hr Fat
control
rat #1
#2
#3
2 hr Fat
control
rat #1
#2
#3
70
7L
128
96
98
114
92
178
84
128
126
156
29
38
124
94
284
72
20
137
103
177
89
2014
2127
329
148
253K
197K
312K
237K
264K
260K
277K
35 OK
229K
235K
26 IK
25 3K
55K
15K
257K
214K
413K
135K
3 IK
299K
140K
277K
21K
208K
253K
312K
24K
152
144
192
165
163
181
153
173
122
165
143
150
28
63
139
87
417
95
36
147
115
205
143
286
232
207
255
62
-------
Parathion-IP
Spectrometer
1 hr Plasma
control
rat #1
#2
#3
2 hr Plasma
control
rat #1
#2
#3
1 hr Brain
control
rat #1
#2
#3
2 hr Brain
control
rat #1
#2
#3
ill
621
2365
4055
678
405
230
875
1048
186
87
92
170
194
125
157
140
229
130
386
403
*
108
67
212
209
122
66
63
1Y1
835K
1190K
518K
1400K
470K
200K
1380K
1360K
360K
280K
270K
290K
310K
260K
285K
280K
475K
230K
46 IK
390K
*
382K
112K
25 8K
312K
154K
115K
32 OK
•".J-J
831 all 30° sector
802
430
998
354
351
600
1012
313
131
137
178
214
140
201
173
284
155
345
197
*
162
56
141
100
69
92
74
* Control contaminated by reference spot on TLC plate
63
-------
Parathion-IP
AB
Spectrometer
1 hr RBC
control
rat #1
#2
#3
2 hr RBC
control
rat #1
#2
#3
264
281
323
277
125
116
151
149
203
145
528
137
150
85
in.
254K
254K
257K
239K
262K
296K
262K
260K
312K
309K
547K
287K
316K
18 IK
164 all 30° sector
222
568
406
142
135
142
151
201
204
426
186
207
127
Background Correction Factors: Parathion-IP
F (Liver): 0.65 ± 0.061
AV
F (Fat): 0.869 ±0.18
AV
F (Plasma): 0.945 ± 0.28
AV
F. (Brain): 0.643 ± 0.043
AV
F (RBC): 1.30 ± 0.30
AV
64
-------
Triflgralin-IP
4 br Liver
control
m
112
"AB
Spectrometer
all 30° sector
rat #1
#2
#3
8 hr Liver
rat #1
#2
#3
24 hr Liver
control
rat #1
#2
#3
48 Lr Liver
rat #1
#2
#3
72 hr Liver
control
rat #1
#2
#3
313
219
434
247
397
188
83
153
560
977
288
263
89
58
194
848
237
465
148
375
141
140
371
239
365
217
33
24
97
46
97
108
89
154
400K
340K
490K
320K
417K
180K
1000K
2440K
1400K
2520K
165 OK
1390K
484K
229K
359K
2395K
680K
2280K
45 OK
2236K
2120K
204 OK
2490K
1740K
2360K
1460K
990K
624K
1430K
817K
1890K
1400K
966K
19 6 OK
62
51
92
68
71
29
74 all 90° sector
153
129
192
100
108
75 30° sector
36 30°
45 30°
128 90°
113 30°
116 90°
77 30°
104 90°
103 all 90° sector
84
110
88
108
72
45 all 90° sector
22
50
40
78
92
47
87
65
-------
Trifluralin-IP
4 hr Fat
control
rat #1
#2
#3
8 hr Fat
rat #1
#2
#3
24 hr Fat
control
rat #1
#2
#3
48 hr Fat
rat #1
#2
#3
72 hr Fat
control
rat #1
#2
#3
Cm
63
71
1591
1021
1396
633
3349
1327
123
127
1833
2805
1543
1785
53
15330
9322
36560
5380
32450
16010
12476
2922
13662
4931
15986
5755
53
17
58352
42944
2302
1090
12696
6207
CM
590K
409K
17 IK
169K
389K
200K
634K
277K
556K
593K
527K
877K
171K
226K
990K
2100K
1290K
1590K
264K
1088K
544K
1170K
267K
1740K
625K
993K
35 IK
977K
395K
2400K
1810K
2140K
945K
1390K
683K
CAB
73
76
35
38
70
39
71
51
108
131
116
55
56
78
60
130
85
134
32
147
84
96
30
102
39
78
43
42
20
220
170
130
77
98
52
Spectrometer
all 30° sector
all 30° sector
all 90° sector
all 90° sector
all 90° sector
66
-------
Trifluralin-IP
-M
-AB
Spectrometer
4 hr Plasma
control
rat #1
#2
#3
8 hr Plasma
rat #1
#2
#3
24 hr Plasma
control
rat #1
#2
#3
48 hr Plasma
rat #1
#2
#3
72 hr Plasma
control
rat #1
#2
#3
64
89
26
84
54
47
24
53
17
45
57
65
111
67
50
26
63
62
82
101
36
68
86
63
79
49
66
53
70
48
56
58
53
47
78
610K
440K
408K
587K
338K
45 OK
186K
394K
170K
204K
1200K
490K
670K
930K
730K
830K
1600K
1800K
1600K
1700K
570K
460K
570K
1300K
580K
1040K
560K
1030K
980K
115 OK
1160K
1780K
1160K
1500K
160 OK
48
42
48
68
43
57
23
53
20
20
67
60
80
39
63
35
65
68
60
100
32
58
74
54
70
41
68
33
50
44
45
63
40
54
75
90°
30°
30°
30°
30°
30°
30°
30°
30°
30°
90°
30°
90°
all
30°
30°
90°
30°
90°
30°
all
sector
sector
90° sector
sector
90° sector
67
-------
Trifluralin-IP
m
CM
"AB
Spectrometer
24 hr Urine
control
rat #1
#2
#3
48 hr Urine
rat #1
#2
#3
72 hr Urine
control
rat #1
#2
#3
83
56
83
70
64
178
83
57
70
50
75
53
31
39
53
42
K3
47
98
16
68
57
490K
400K
615K
500K
420K
780K
620K
370K
2000K
1000K
2000K
1400K
690K
780K
HOOK
830K
511K
980K
1800K
580K
2100K
1500K
80 all 30° sector
56
73
52
50
115
79
50
56 all 90° sector
40
70
63
24
31
43 all 90° sector
46
17
35
73
9
70
61
68
-------
Trifluralin-IP
m
-AB
Spectrometer
24 hr Feces
control
rat #1
#2
#3
48 hr Feces
rat #1
#2
#3
72 hr Feces
control
rat #1
#2
#3
48
86
57
93
94
103
92
54
68
36
48
39
74
38
146
68
23
115
65
89
17
250K
470K
290K
600K
1300K
560K
550K
115 OK
1540K
1160K
1330K
1000K
1490K
1790K
710K
520K
180K
690K
88 6K
730K
57K
46
70
36
90
93
112
85
71
56
47
47
42
59
54
171
67
31
84
57
74
10
30° sector
30°
30°
30°
90°
30°
30°
90°
all 90° sector
30° sector
30°
30°
30°
90°
90°
30°
Background Correction Factors: Trifluralin-IP
F (Liver): 1.17 ± 0.31
AV
F (Fat) : 0.97 ± 0.17
AV
F (Plasma): 1.49 ± 0.17
AV
F (Urine): 1.04 ± 0.13
AV
F (Feces): 1.04 ± 0.13
AV
69
-------
Trifluralin-Oral
CM
Spectrometer
CONTROL GROUP
Liver
rat
Fat
rat
Plasma
rat
#1
#2
#3
#1
#2
#3
#1
#2
#3
7 day Urine
rat #1
#2
#3
14 day Urine
rat #1
#2
#3*
21 day Urine
rat #1
#2
#3
19
2(1
13
67
58
45
34
22
14
48
36
46
36
16
35
19
684
26
14
17
m
16 2K
135K
83K
930K
1000K
720K
670K
15 OK
135K
390K
195K
278K
820K
9 IK
246K
110K
780K
193K
5 IK
12 7K
25
13
15
61
75
29
37
46
18
54
26
43
38
14
29
12
568
27
12
26
30° sector
30°
30°
90°
90°
90°
90°
90°
30°
90°
30°
30°
90°
30°
30°
30°
90°
30°
30°
30°
* high background contamination
70
-------
Trifluralin-Oral
M
-AB
Spectrometer
CONTROL GROUP
7 day Feces
rat #1
#2
#3
14 day Feces
rat #1
#2
#3
21 day Feces
rat #1
#2
#3
50
66
33
19
35
27
22
60
28
26
322K
360K
16 2K
227K
213K
130K
295K
410K
48K
14 3K
49
62
24
27
25
20
36
51
36
24
30° sector
30°
30°
30°
30°
30°
90°
90°
90°
90°
Background Correction Factors: Trifluralin-Oral
F (Liver): 1.22 ± 0.67
AV
F (Fat): 0.97 ± 0.41
AV
F (Plasma): 1.03 ± 0.32
AV
F (Urine): 1.09 ± 0.27
AV
F (Feces): 1.06 ± 0.28
AV
71
-------
Trifluralin-Oral
20
(jtg/kg GROUP
Liver
rat #1
#2
#3
Fat
rat #1
#2
#3
Plasma
rat #1
7
14
21
#2
#3
day Urine
rat #1
#2
#3
day Urine
rat #1
#2
#3
day Urine
rat #1
#2
#3
m in ~AB
19 490K 22
30 476K 28
27 600K 28
27 530K 17
46 850K 49
28 210K 17
22 590K 31
28 450K 24
56 610K 36
31 126K 21
28 180K 38
19 189K 27
49 HOOK 60
63 1270K 59
63 910K 53
14 250K 19
9 119K 9
22 367K 27
"^^^ tri WlU^ UC J.
90° sector
90°
90°
90°
90°
90°
90°
90°
90°
30°
30°
30°
90°
90°
90°
90°
90°
90°
72
-------
Trifluralin-Oral
Spectrometer
20 y,g/kg GROUP
•"•""
7 day Feces
rat
#1
#2
#3
11
32
28
56K
260K
194K
10 30° sector
33 30°
31 30°
14 day Feces
rat
#1
#2
#3
38
21
12
480K
372K
290K
53 90°
19 90°
19 90°
21 day Feces
rat
100 (ag/kg
Liver
rat
Fat
rat
Plasma
rat
#1
#2
#3
GROUP
#1
#2
#3
#1
#2
#3
#1
#2
#3
43
30
7
22
19
86
138
89
96
36
325
149
2
13
19
480K
510K
13 6K
72 OK
660K
1500K
990K
776K
810K
470K
600K
710K
160K
490K
390K
73
48 90°
30 90°
8 90°
31 all 90° sector
31
72
57
37
66
28
148
57
9
26
21
-------
Trifluralin-Oral
Spe c trome ter
100
7
14
21
7
14
21
y,g/kg GROUP
day Urine
rat #1
#2
#3
day Urine
rat #1
#2
#3
day Urine
rat #1
#2
#3
day Feces
rat #1
#2
#3
day Feces
rat #1
#2
#3
day Feces
rat #1
#2
#3
111
51
33
52
52
34
41
43
14
12
58
24
52
22
41
32
43
12
23
m
790K
635K
920K
1030K
900K
HOOK
850K
560K
150K
710K
250K
660K
200K
720K
560K
270K
245K
330K
11 J-»
47 all 90° sector
46
57
44
47
35
37
26
10
58
25
45
29
50
36
26
16
21
74
-------
Trif luralin-Ora^.
500 ug/kg GROUP
Liver
rat #1
#2
Fat
m
CM
"AB
Spectrometer
17 54K 16
51 790K 47
30 380K 22
rat #1
#2
#3
Plasma
rat #1
#2
#3
7 day Urine
rat #1
#2
#3
14 day Urine
rat #1
#2
#3
21 day Urine
rat #1
#2
#3
36
233
140
51
68
139
^ 51
55
33
149
182
72
76
53
90
63
61
39
100K
940K
1290K
447K
403K
856K
1030K
710K
72 OK
1600K
680K
750K
930K
790K
1300K
596K
620K
740K
17
64
59
44
29
42
46
44
42
161
104
62
51
47
66
44
46
40
all 90° sector
75
-------
Trifluralin-Oral
"AB
Spectrometer
500 ug/kg GROUP
7 day Feces
#1
#2
#3
sees
#1
#2
#3
sees
#1
#2
#3
48
108
53
37
194
524
887
1068
87
46
123
160
215
172
423
27
275K
700K
550K
190K
12 OK
52 OK
360K
430K
540K
630K
710K
260K
430K
216K
420K
215K
26
53
41
28
79
72
86
78
39
45
75
83
34
50
47
17
all 90 sector
76
-------
TbCHNICAL REPORT DATA
. /ii.'Sf miL/ /u^j-iit'tions on the reverse before completing)
1
4
7.
9.
12
15
16
17.
a.
REPORT NO '?
EPA-600/1 -76-037
I 1TLE AND SUBTITLE
FEASIBILITY OF APPLYING FIELD IONIZATION l<
SPECTROMETRY TO PESTICIDE RESEARCH
AUTHOR(S)
R.L. Dyer, H. d'A. Heck, A.C. Scott and M
PERFORMING ORGANIZATION NAV AND ADDRESS
Stanford REsearch Institute
Menlo Park, California 94025
. SPONSORING AGENCY NAME AND ADDRESS
Health Effects Research Laboratory
Office of Research and Development
U.S. Environmental Protection Agency
Research Triangle Park. N.C. 27711
. SUPPLEMENTARY NOTES
3 RECIPIENT'S ACCESSIOf+NO.
5 REPORT DATE
/I/\SS i November 1976
•6 PERFORMING ORGANIZATION CODE
8. PERFORMING ORGANIZATION REPORT NO.
. Anbar i
10. PROGRAM ELEMENT NO.
1EA615
11. CONTRACT/GRANT NO.
68-02-1799
13. TYPE OF REPORT AND PERIOD COVERED
First Annual Report
14. SPONSORING AGENCY CODE
EPA-ORD
. ABSTRACT
An isotope dilution methodology was develped for analysis of an insecticide,
parathion, and a herbicide, trifluralin, isolated from rat tissues and excreta.
Sample cleanup was facilitated by use of high-pressure gel permeation chromatography
in conjunction with thin layer chromatography and reversed-phase high pressure
liquid chromatography. Isotope ratio measurements were performed using multi-
labeled stable isotopic carriers and nonfragmenting field ionization mass
spectrometry. Parathion and trifluralin were administered intraperitoneally and/
or orally at the sub-mg/kg level, and the unchanged materials assayed in tissues
and excreta at the ppb level. The technique was also applied to the determination
of parathion and methyl parathion concentrations in aerosols. The biological
implications of the results of the animal experiments are discussed.
KEY WORDS AND DOCUMENT ANALYSIS
DESCRIPTORS
Pesticides
Herbicides
Chemical analysis
lonization
Mass spectroscopy
Biochemistry
13
DISTRIBUTION STATEMENT
RELEASE TO PUBLIC
b IDENTIFIERS/OPEN ENDED TERMS c. COSATI I icld/Group
Parathion 07 C
Trifluralin 06 A
Methyl parathion
'<• ^ RCURITY CLASS f i his Report) 21 NO. OF PAGES
UNCI ASSTFTFD 8fi
jo SECURITY CLASS 'T/i.'i pa^ 22 PRICF
UNCLASSIFIED
EPA Form 2220-1 (9-73)
77
------- |