TOXICOLOGICAL SCIENCES
66, 176 –184 (2002)
Interaction of Exposure Concentration and Duration in Determining
Acute Toxic Effects of Sarin Vapor in Rats
R. Mioduszewski, 1 J. Manthei, R. Way, D. Burnett, B. Gaviola, W. Muse, S. Thomson, D. Sommerville, and R. Crosier
U.S. Army Edgewood Chemical Biological Center, AMSSB-RRT-TT (E3150), 5183 Blackhawk Road, Aberdeen Proving Ground, Maryland 21010-5424
Received August 14, 2001; accepted November 1, 2001
Sarin (GB) vapor exposure is associated with both systemic and
local toxic effects occurring primarily via the inhalation and ocular routes. The objective of these studies was to develop models for
predicting dose-response effects of GB vapor concentrations as a
function of exposure duration. Thus, the probability of GB vaporinduced lethality was estimated in rats exposed to various combinations of exposure concentration and duration. Groups of male
and female Sprague-Dawley rats were exposed to one of a series of
GB vapor concentrations for a single duration (5–360 min) in a
whole-body dynamic chamber. The onset of clinical signs and
changes in blood cholinesterase activity were measured with each
exposure. Separate effective concentrations for lethality in 50% of
the exposed population (LC50) and corresponding dose-response
slopes were determined for each exposure duration by the Bliss
probit method. Contrary to that predicted by Haber’s rule, the
interaction of LC50 ⴛ time (LCT50) values increased with exposure duration (i.e., the CT for 50% lethality in the exposed population and corresponding dose-response slope was not constant
over time). A plot of log (LCT50) versus log (exposure time)
showed significant curvature. Predictive models derived from multifactor probit analysis of results describing the relationship between exposure conditions and probability of lethality in the rat
are discussed. Overall, female rats were more sensitive to GB
vapor toxicity than male rats over the range of exposure concentration and duration studied. Miosis was the initial clinical sign
noted after the start of GB vapor exposure. Although blood cholinesterase activity was significantly inhibited by GB vapor exposure, poor correlation between cholinesterase inhibition and exposure conditions or cholinesterase inhibition and severity of clinical
signs was noted.
Key Words: Sarin; inhalation; exposure concentration; rat; lethality; miosis; mydriasis; cholinesterase; LC50; LCT50.
Historically, it has been standard practice to use a linear
time-integrated concentration (i.e., C ⫻ T, CT, or dosage) to
predict mortality-response relationships for chemical vapor
exposures. The above concept has persisted as an accepted
principle in military hazard assessment, currently serving as
the basis for estimating injury from exposure to chemical
1
To whom correspondence should be addressed. Fax: (410) 436-7129.
E-mail: [email protected].
warfare (CW) agents. This measure of exposure was first
attributed to Haber (1924), who found that phosgene gas–
induced toxic effects appeared to be correlated with dosage (at
least between 5-min and 8-h durations). Haber’s law or rule, as
commonly understood in inhalation toxicology, states
C⫻T⫽k
(1)
with regard to the incidence of a particular biological effect. A
linear time-integrated concentration implies concentration and
duration are equally important components in quantifying the
toxic potential of an exposure.
Alternatively, ten Berge et al. (1986) demonstrated that:
C nT ⫽ k
(2)
correlates well with the degree of injury; where C is concentration, T is exposure time, n is an index (the so-called toxic
load exponent) that depends on the particular gas/aerosol or
exposure scenario, and k is a constant. In general, for most
gases for which experiments have been conducted, n was found
greater than one. In order to determine the value for n, an
experiment must be designed such that both exposure concentration and duration are varied in the same study. If the above
concepts for quantifying health risks of toxic gases are also true
for chemical nerve agents, the traditional use of dosage (i.e.,
C ⫻ T) in estimating casualty/risk is not appropriate.
The primary objective of the present study was to model the
relationship between GB vapor exposure concentration (C),
duration of exposure (T) and the probability of a toxic effect
(lethality). Developing such models is hampered by the scarcity of published data involving acute chemical agent vapor
exposures beyond several minutes (NRC, 1997; Yee, 1996).
Estimating responses to exposures beyond a few minutes currently requires extrapolation based upon the assumptions of the
theoretical dosage models cited above. This study tested
whether the relationship between exposure concentration time
and lethal response in rats exposed to GB vapor for 5–360 min,
could be adequately described by the above models.
176
INTERACTION OF EXPOSURE CONCENTRATION AND DURATION
MATERIALS AND METHODS
Chemicals. Isopropyl methyl phosphono fluoridate (Sarin, or GB) was
used for all vapor exposures in this study. Chemical agent standard analytical
reagent material (CASARM)-grade Sarin (lot #GB-U-6814-CTF-N [GB2035])
was verified as 97.2 ⫹ 0.2 wt. % pure (as determined by quantitative NMR
31P) and stored in sealed ampoules containing nitrogen. Ampoules were
opened as needed to prepare external standards or to be used as neat agent for
vapor dissemination. All external standards for GB vapor quantification were
prepared on a daily basis. Triethylphosphate (99.9% purity), obtained from
Aldrich Chemicals, Milwaukee, WI, was used as the internal standard for the
GB purity assay. Hexane (purity ⬎ 85% n-hexane and 99.9% n-hexane and
isomers), purchased from Burdick and Jackson, Muskegon, MI, was the
solvent used for standard preparation and bubbler collection procedures. It
should be noted that it was necessary to use hexane only for solid sorbent
sampling calibration and not for animal exposures.
The majority of the impurities in the CASARM GB consisted of 0.2%
o,o⬘-diisopropyl methyl phosphonate (DIMP), 0.2% methyl phosphonic difluoride (DF), 0.3% methyl fluoridic acid (Fluor Acid), and 0.3% excess HF/F
ion. Impurity percentages were based on mole ratios from acid-base titration.
Vapor generation. The vapor generation system (Muse et al., 2000) consisted of a gas-tight syringe (Hamilton, Reno, NV), variable-rate syringe drive
(Model 22, Harvard Apparatus Inc., South Natick, MA), and a vaporization
apparatus. This system, located at the chamber inlet, was contained within a
stainless steel glove box (23 in. long ⫻ 14 in. wide ⫻ 18 in. high) maintained
under negative pressure (0.25 in. H 2O). A Plexiglas door at the front of the
glove box facilitated syringe loading and syringe drive adjustments during
setup operations. Prior to chamber operation, liquid GB (undiluted) was loaded
into a gas-tight syringe (Hamilton, Reno, NV), then the syringe was mounted
onto the syringe drive. Once activated, the syringe drive provided a constant
dispersal rate of GB (␮l/min) through a flexible plastic line (⬃8 in. long) into
a spray atomization system (Spray Atomization Nozzle 41 J SS, Spraying
Systems Co., Wheaton, IL). The atomizer was modified by retrofitting a
syringe needle (SS 25 gauge 3 in.) into the top of the sprayer to provide a
smaller orifice. As liquid GB entered through the top of the sprayer, compressed air (30 – 40 psi) entered through the side. The compressed air broke the
liquid GB into fine droplets (⬍ 15 ␮m diameter.), which then entered the
chamber inlet. Due to the high volatility of GB (2.2 ⫻ 10 4 mg/m 3 at 25°C),
these droplets quickly evaporated, with the resulting vapor drawn through the
exposure chamber.
Sampling/Monitoring Exposure Chamber GB Vapor
Vapor sampling/analysis. Three methods were used to sample/monitor
and analyze GB vapor concentration in the exposure chamber: (1) Edgewood
bubblers (containing hexane)/ gas chromatograph with flame photometric
detection (GC-FPD); (2) solid sorbent tubes (Tenax-TA)/gas chromatograph
with flame ionization detection (GC-FID); and (3) a phosphorus monitor
(HYFED, Model PH262) that provided a continuous strip chart record of rise,
equilibrium, and decay of the chamber vapor concentration during an exposure.
GB vapor was sampled as it was drawn through a 750-liter dynamic airflow,
whole-body inhalation exposure chamber located within a 20-m 3 containment
chamber. The exposure chamber was constructed of stainless steel with Plexiglas windows on each of the six sides. The interior of the exposure chamber
was maintained under negative pressure (0.25 in. H 2O), which was monitored
with a calibrated magnehelix (Dwyer, Michigan City, IN). Room air (500 – 650
l/min) was drawn through the chamber under negative pressure. A thermoanemometer (Model 8565, Alnor, Skokie, IL) was used to monitor chamber
airflow at the chamber outlet. All samples were drawn from the same area
(middle) of the chamber. Bubbler and solid sorbent tube samples were drawn
after the chamber attained equilibration (t 99) while the HYFED monitored the
entire run. Bubbler samples were drawn from the chamber every hour, with
each sampling period lasting 1–12 min. Solid sorbent tube samples were drawn
177
from the chamber approximately every 15 min, with each sampling lasting 2–3
min. All sample flow rates for the bubbler and solid sorbent tube systems were
controlled with calibrated mass flow controllers (Matheson Gas Products,
Montgomeryville, PA). Typical flow rates were 0.9 –1.0 l/min for the bubblers
and 100 standard cubic centimeters per minute for the sorbent tubes. Due to
solvent (hexane) evaporation during sampling, an in-line charcoal filter was
installed between the bubbler and mass flow controller. This was to prevent the
cooling effect of the solvent from affecting the mass flow sensor. Flow rates
from both systems were verified before and after sampling by temporarily
connecting a calibrated flow meter (DryCal威, Bios International, Pompton
Plains, NJ) in-line to the sample stream.
Bubbler tube. The concentration of GB in the chamber was determined by
collecting chamber air samples into Edgewood bubblers containing hexane,
connected in series (Muse et al., 2000). The second bubbler (downstream), in
a series of two, contained one-half the solvent volume of the upstream bubbler
to increase sensitivity for analyzing the lower GB concentration (5– 8% of the
upstream bubbler) present in the downstream bubbler. Samples were drawn
through glass sample lines (0.25 in. o.d.) at the rate of 0.9 –1.0 l/min. The
collected solvent was diluted to a known volume and injected into a gas
chromatograph with flame photometric detection, (GC-FPD) phosphorus
mode. External standards (GB/hexane) were injected into the GC-FPD to
generate a calibration curve. A linear regression fit (r 2 ⫽ 0.999) of the standard
data was used to compute GB concentration.
Solid sorbent tube. The automated solid sorbent tube sampling system
consisted of four parts: a heated sample transfer line; heated external switching
valve; thermal desorption unit; and gas chromatograph. A stainless steel
sample line (1/16 in. o.d. ⫻ 0.004 in. i.d. ⫻ 6 feet long) extended from the
middle of the chamber to an external sample valve. The sample line was
commercially treated with a silica coating (Silicasteel威 Restek, Bellefonte, PA)
and covered with a heated (60°C) sample transfer line (CMS, Birmingham,
AL). The combination line coating and heating minimized GB absorption onto
sample surfaces. From the transfer line, the sample entered a heated (125°C)
6-port gas-switching valve (UWP, Valco Instruments, Houston, TX). In the
bypass mode, GB vapor from the chamber continuously purged through the
sample line and out to a charcoal filter. In the sample mode, the gas sample
valve redirected GB vapors from the sample line to a Tenax TA sorbent tube
located in the thermal desorption unit (ACEM-900, Dynatherm Analytical
Instruments, Kelton, PA). The solid sorbent material used to trap GB vapor
consisted of Tenax-TA. Temperature and flow programming within the Dynatherm desorbed GB from the sorbent tube for vapor injection directly onto the
gas chromatograph (GC) for quantitation. Either flame ionization detection
(FID) or flame photometric detection (FPD) was used, depending upon the
level of sensitivity required.
To calibrate the solid sorbent tube sampling system, external standards
(GB/hexane) were injected directly into the heated sample line of the Dynatherm. In this way, injected GB standards were put through the same sampling
and analysis stream as the chamber samples. A linear regression fit (r 2 ⫽
0.999) of the standard data was used to compute chamber sample GB concentration.
Phosphorus monitor (HYFED). GB levels in the chamber were continuously monitored with a phosphorus analyzer (HYFED, Model PH262, Columbia Scientific, Austin, TX). The analyzer output was recorded on a strip chart
recorder, which showed the rise, equilibrium, and decay of the chamber vapor
concentration during a run. In addition, it gave a close approximation of the
amount of GB (␮g/l) in the chamber based on data (bubbler and solid sorbent
tube quantification with HYFED response) from previous chamber runs.
Animal Exposures
In conducting this study, investigators adhered to the Guide for the Care and
Use of Laboratory Animals, National Institutes of Health Publication No.
86-23, 1985, as promulgated by the Committee on Revision of the Guide for
Laboratory Animal Facilities and Care of the Institute of Laboratory Animal
Resources, Commission of Life Sciences, National Research Council, Wash-
178
MIODUSZEWSKI ET AL.
TABLE 1
Summary of GB Vapor-Induced Clinical Signs in Male and Female Rats at Various Exposure Durations and Concentrations
Salivation
Exp. dur.
(t 99)
Tremors
Convulsions
Lethality
Conc.
Male
Female
Male
Female
Male
Female
Male
Female
25.6
28.2
31.5
36.3
44.0
48.1
51.4
54.4
—
—
—
5/10
6/10
5/10
9/10
9/10
4/10
1/10
5/10
7/10
9/10
—
—
—
—
—
—
10/10
10/10
10/10
10/10
10/10
10/10
8/10
10/10
10/10
10/10
—
—
—
—
—
—
9/10
7/10
10/10
8/10
10/10
5/10
1/10
9/10
9/10
10/10
—
—
—
—
—
—
2/10
4/10
5/10
7/10
8/10
2/10
0/10
7/10
6/10
9/10
—
—
—
9.6
12.0
15.3
18.7
21.8
27.1
34.3
—
—
1/10
2/10
4/10
8/10
10/10
0/10
2/10
7/10
6/10
0/10
—
—
—
—
9/10
8/10
8/10
10/10
10/10
0/10
8/10
10/10
10/10
10/10
—
—
—
—
1/10
1/10
8/10
10/10
10/10
0/10
1/10
7/10
8/10
10/10
—
—
—
—
0/10
1/10
4/10
9/10
10/10
0/10
0/10
3/10
4/10
9/10
—
—
6.0
7.4
9.0
10.3
12.1
2/10
2/10
8/10
10/10
10/10
5/10
4/10
9/10
9/10
9/10
7/10
10/10
10/10
10/10
10/10
9/10
10/10
10/10
10/10
10/10
0/10
2/10
7/10
10/10
10/10
2/10
8/10
10/10
10/10
10/10
0/10
1/10
4/10
10/10
10/10
0/10
1/10
6/10
8/10
9/10
6.0
6.4
7.0
7.6
8.1
5/10
7/10
6/10
8/10
7/10
8/10
6/10
9/10
8/10
10/10
10/10
10/10
10/10
10/10
10/10
10/10
10/10
10/10
10/10
10/10
710
7/10
9/10
9/10
7/10
9/10
10/10
9/10
10/10
10/10
3/10
4/10
2/10
6/10
6/10
3/10
6/10
3/10
5/10
10/10
4.0
4.1
4.5
4.9
5.5
4/10
1/10
7/10
9/10
8/10
6/10
4/10
9/10
7/10
10/10
10/10
10/10
10/10
10/10
10/10
10/10
10/10
10/10
10/10
10/10
4/10
1/10
9/10
10/10
10/10
9/10
5/10
10/10
8/10
10/10
0/10
0/10
5/10
6/10
8/10
2/10
1/10
7/10
7/10
10/10
2.1
2.7
3.3
4.2
4.4
0/10
2/10
3/10
8/10
5/10
0/10
10/10
10/10
8/10
10/10
0/10
10/10
10/10
10/10
10/10
6/10
10/10
10/10
10/10
10/10
0/10
1/10
2/10
10/10
9/10
0/10
10/10
10/10
10/10
10/10
0/10
1/10
1/10
6/10
6/10
0/10
4/10
7/10
8/10
10/10
2.3
2.4
2.7
2.8
3.0
3.5
4/10
—
8/10
7/10
9/10
10/10
5/10
8/10
6/10
9/10
9/10
—
10/10
—
10/10
1010
10/10
10/10
10/10
10/10
10/10
10/10
10/10
—
5/10
—
8/10
6/10
10/10
10/10
5/10
10/10
8/10
10/10
10/10
—
2/10
—
2/10
2/10
8/10
9/10
3/10
2/10
5/10
6/10
9/10
—
5 (2.1)
10 (5.3)
30 (5.3)
60 (6.3)
90 (5.3)
240 (5.3)
360 (6.9)
Note. Exp. dur. (exposure duration) and t 99 are given in min; concentration values are in mg/m 3. Clinical signs are given as no. affected/total; lethality assessed
at 14 days postexposure.
ington, DC. These investigations were also performed in accordance with the
requirements of AR 70-18, Laboratory Animals, Procurement, Transportation,
Use, Care, and Public Affairs, and the U.S. Army Edgewood Chemical and
Biological Center Institutional Animal Care and Use Committee (IACUC).
Animal model. Young adult male and female Sprague-Dawley rats (8 –10
weeks) were purchased from Charles River Laboratories, Inc., Wilmington,
MA. Rats were identified by tail tattoo and housed individually in plastic
shoebox cages in an American Association for Accreditation of Laboratory
179
INTERACTION OF EXPOSURE CONCENTRATION AND DURATION
TABLE 2
Summary of LC50, LCT50, Slopes and Fiducial Limits for GB Vapor-Induced Lethality (14 Days Postexposure)
at Each of Seven Exposure Durations
Exp. Dur.
LC50
95% FL
Slope
LCT50
95% FL
Slope
5
10
30
60
90
240
360
32.8
18.1
8.51
6.39
4.46
3.03
2.63
29.8–36.6
16.3–20.4
7.79–9.31
5.72–6.95
4.24–4.69
2.66–3.37
2.44–2.82
10.3
11.9
12.9
13.0
22.1
9.9
14.9
164
181
255
383
401
727
947
149–183
163–204
234–279
343–417
382–422
638–809
878–1015
10.3
11.9
12.9
13.0
22.1
9.8
14.9
5
10
30
60
90
240
360
45.9
22.6
8.84
7.55
4.81
4.09
2.89
40.0–51.3
20.8–24.8
8.20–9.47
—a
4.58–5.12
3.66–5.00
2.69–3.15
9.4
16.4
21.6
6.1
21.5
8.0
13.2
230
226
265
453
433
982
1040
200–257
208–248
246–284
—a
412–461
878–1200
968–1134
9.4
16.4
21.6
6.1
21.5
8.0
13.2
Females
Males
Note. Exp. dur. (exposure duration) given in min; LC50 given in mg/m 3; LCT50 given in mg.min/m 3. FL, fiducial limits.
a
Not able to calculate due to nonsignificant slope at 95% level.
Animal Care (AAALAC) accredited facility. Ambient holding conditions were
maintained at 21 ⫾ 3°C, 40 –70% relative humidity, and a 12:12 h light-dark
cycle. Rats were provided with certified laboratory rat chow and water ad
libitum (automatic watering system using a reverse osmosis process), except
during vapor exposure. Animals were quarantined for at least 5 days prior to
exposure.
Whole-body inhalation exposures. All animals were exposed (wholebody) to GB vapor in a 750-liter dynamic airflow inhalation chamber located
within a 20-m 3 containment chamber. During inhalation chamber operations,
the airflow through the chamber was kept constant. The concentration-time
profile generated with this type of inhalation chamber is described in a review
by MacFarland (1987). His definition of exposure duration was the one used in
this study: the interval from the start of the flow of agent into the chamber to
the time point when the agent supply is stopped. The time required for the
vapor concentration to reach 99% (denoted as t 99) of its equilibrium value for
each of the exposure durations is listed in Table 1. The t 99 is also the time
required for the chamber to lose 99% of its equilibrium concentration after the
agent supply is stopped. The concentrations recorded in Table 1 are the
equilibrium concentrations.
Rats were exposed to GB vapor for one of several exposure durations (from
5 to 360 min, inclusive) and observed for mortality/clinical signs for up to 14
days postexposure. In most cases, groups of 20 rats (10 male and 10 female )
were assigned to each of five concentrations per exposure duration. However,
in a few cases where an exposure concentration was likely to result in an
all-or-none lethality outcome for a particular sex, male (10) and female (10)
rats were tested separately.
Exposure groups consisted of 10 rats of one sex or 20 rats (10 of each sex)
placed in stainless steel compartmentalized cages (20 in. wide ⫻ 14 in. long ⫻
4 in. high), with each rat in a separate compartment. Rats were exposed to one
of five concentrations (2–54 mg/m 3) of GB vapor for one of seven exposure
times (5, 10, 30, 60, 90, 240, or 360 min). Lethality and sublethal clinical signs
(e.g., miosis, tremors, salivation, labored breathing, and convulsions) were
monitored from an observation point outside the chamber during exposure,
afterward within the first hour postexposure, and once daily thereafter for up
to 14 days. Physical parameters monitored during exposure included chamber
airflow (monitored continuously), as well as chamber room temperature and
relative humidity. Immediately after completion of the exposure period, the
chamber was purged with air for a minimum time of t 99 (t 99 ⫽ time for chamber
to attain or lose 99% of its equilibrium concentration).
Experimental Design
This study consisted of two parts. In Part I, concentration-response curves
were determined for each of four GB vapor exposure durations (10, 30, 90, and
240 min), with separate shipments of rats used for each duration. The first
shipment was used for 10-min duration exposures, the next shipment for
30-min duration, and the last two shipments for 90- and 240-min durations, in
that order. Additional exposure durations (5, 60, and 360 min) were conducted
in Part II. In Part II, five shipments of rats were used. Each shipment was
divided into three equal-sized groups, one group for each exposure duration.
Thus, the five shipments yielded five separate exposure concentrations per
exposure duration. The three groups were tested in random order. Concentrations for the three exposure durations (i.e., the 5, 60, and 360 min) were varied
independently of each other. Selection of exposure concentration depended on
the fractional mortality at the concentrations used for the previous shipments.
Blood sample collection. Two blood samples were collected (within 24 h
prior to exposure and within 60 min after exposure) from each surviving
animal for the purpose of measuring cholinesterase activity in both RBC and
plasma components. Tail vein blood samples were collected into glass tubes
containing EDTA. Assays of red blood cell acetylcholinesterase (AChE) and
plasma butyrylcholinesterase (BuChE) activity were performed by the U.S.
Army Medical Research Institute of Chemical Defense (USAMRICD), APG,
Maryland, using a modification of the Ellman reference method (Ellman et al.,
1961).
Observation of clinical signs. Observations for toxic clinical signs were
carried out during, immediately after, and once daily after exposure. Clinical
signs of clonic or tonic movements (e.g., convulsions, tremors, muscle fasciculation), piloerection, gait abnormalities, overt respiratory abnormalities, and
general appearance were also noted. Pupil sizes were monitored at least 24 h
prior to exposure, at 1–2 h following whole-body GB vapor exposure, and at
1, 2, 3, and 7 days postexposure. Pupil size (diameter) was assessed using a
simple microscope (Bausch and Lomb, 20⫻) with a reticule eyepiece insert
under a 200-foot-candle light source, as monitored by a light meter (Davis,
Model 401025, Extech Instruments, Waltham, MA). This procedure consisted
180
MIODUSZEWSKI ET AL.
of counting the number of reticule lines covering the pupil diameter (20
lines/mm or 0.05 mm between lines).
Data analysis. A probit analysis program (a component of MINITAB威,
Version 13) was used to generate a separate dose-response curve (with slope,
intercept, and 95% fiducial limits) for each duration of exposure tested and to
determine if sex differences exist in the sensitivity to the toxic effects of GB
vapor exposure. Binary normal regression (multifactor probit analysis) was
used to model the relative effects of exposure concentration and duration on
probability of lethality. Differences in preexposure versus postexposure cholinesterase levels were expressed as a percent change resulting from treatment;
thus, each individual rats served as its own control. This graded response was
plotted against CT using linear regression to determine if correlations exist as
indicated by significant regression coefficients.
RESULTS
Lethal and Sublethal Responses
Table 1 summarizes the fraction of exposed male and female
rats in which salivation, tremors, convulsions, and lethality
were recorded. The results of probit analyses for each duration
of GB vapor-induced lethality occurring within 14 days after
exposure can be found in Table 2 and include estimates of
LC50s, LCT50s, 95% fiducial limits, and probit slopes. The
LC50 values for both male and female rats as a function of
exposure duration are displayed in Figure 1.
The exposure conditions used in the present study were selected to optimize estimates of the LC50 and LCT50. Consequently, high (i.e., maximal) incidences of sublethal toxic signs
were often observed under these conditions (Table 1) and did not
permit estimates of EC50 values for mild sublethal signs.
Blood Cholinesterase Activity
Both red blood cell acetylcholinesterase (AChE) and plasma
butyrylcholinesterase (BuChE) activities were inhibited as a
result of exposure to various combinations of GB vapor concentration and time (Fig. 2). Most responses appeared to fall
between 5 and 30% of pretreatment baseline. These data are
limited to samples collected from rats surviving up to 60 min
after exposure to GB vapor. Each postexposure response is
expressed as percent of baseline (i.e., each animal’s postexposure AChE or BuChE activity is expressed as a percent of its
pretreatment value collected at 24 h prior to exposure). Except
for AChE activity in female rats, there was a statistically
significant (p ⬍ 0.05) correlation between ChE activity and
CT. However, the fits for male AChE (r 2 ⫽ 0.04), female
AChE (r 2 ⫽ 0.00), male BuChE (r 2 ⫽ 0.31), and female
BuChE (r 2 ⫽ 0.51) versus CT were very poor and could not be
used for predictive purposes. Median pretreatment levels of
BuChE activity were consistently higher (p ⬍ 0.001) in female
(1750 U/ml) than in male rats (424 U/ml), as determined by the
Mann-Whitney Rank Sum test. However, no differences were
noted between pretreatment male and female AChE activity.
Pupil Size
Mean pupil sizes in rats (survivors) before and after 5-, 60-,
240-, and 360-min exposures are presented in Figure 3. Anal-
FIG. 1. Fits of Interaction (solid line) and Toxic Load models (dashed
line) to LC50 (symbols) versus duration GB vapor exposure are compared for
male and female rats.
ysis of variance results indicated significant changes (vs. preexposure) in pupil size in rats exposed to GB vapor. The results
of Bonferonni and Tukey tests indicated that all combinations
of GB vapor exposure concentrations and times resulted in
complete miosis (pinpoint pupil, p ⬍ 0.01) in male and female
rats, as measured at the first hour postexposure. This was
followed by a transient mydriasis (dilated pupil) between 24
and 48 h postexposure (p ⬍ 0.01) lasting several days. By 7
days postexposure, pupil diameters decreased in size (p ⬍0.01)
but were still larger (p ⬍ 0.01) than preexposure sizes.
DISCUSSION
Evaluating the Adequacy of Traditional Models in Predicting
the Toxicity of GB Vapor Exposure in the Rat
When the data from Part I were analyzed, it was found that
LCT50 was not constant with respect to time and that there was
statistically significant curvature in the plot of log(LCT50)
181
INTERACTION OF EXPOSURE CONCENTRATION AND DURATION
FIG. 2. Blood AChE (left) and
BuChE (right) activity is expressed as
percent of preexposure baseline activity
of individual male and female rats and
plotted versus CT of GB Vapor Exposure. A least squares fit of ChE versus CT
is indicated by a solid line.
versus log(Time). All logs in this paper are base 10. Although
there appeared to be sufficient statistical evidence to reject
Haber’s rule and the toxic load model, this conclusion can be
questioned because the order of exposure times in Part I was
not randomized. Because the Part I exposures were conducted
over a 6-month period, seasonal differences or differences
among shipments of rats may have produced the curvature in
the plot of log(LCT50) versus log(Time). Part II exposures
solved this problem by the use of a different randomization
procedure.
To examine the curvature in the plot of log(LCT50)
versus log(Time), only the data from Part II were used.
Starting with the 12 terms constant, cLogC, cLogT,
(cLogC) 2 , (cLogT) 2 , cLogC*cLogT, Sex (coded 1 for male
and –1 for female), Sex*cLogC, Sex*cLogT, Sex*(cLogC) 2 ,
Sex*(cLogT) 2 , and Sex*cLogC*cLogT, where cLogC ⫽
centered log(Concentration) and cLogT ⫽ centered log(Time), the least significant term (largest p value) was
deleted followed by reanalysis. This process was iterated
until all terms were significant (p ⬍ 0.05). The centering
(subtracting the mean) of log(C) and log(T) reduces multicollinearity in the model. The backwards elimination procedure resulted in the following model:
For probability of lethality, let Y ⫽ normit (normit ⫽ Probit
–5), then
Y ⫽ 0.6691 ⫺ 0.42381*Sex ⫹ 11.171*cLogC
⫹ 6.7223*cLogT ⫹ 1.8936*cLogC*cLogT
⫺ 0.3813*Sex*cLogC.
(3)
For males, this reduces to
Y ⫽ 0.2453 ⫹ 10.7897*cLogC ⫹ 6.7223*cLogT
⫹ 1.8936*cLogC*cLogT,
(4)
and for females, to
Y ⫽ 1.0929 ⫹ 11.5523*cLogC ⫹ 6.7223*cLogT
⫹ 1.8936*cLogC*cLogT,
(5)
where the means for centering are 0.951702 for LogC and
1.67781 for LogT.
The Sex*cLogC term in the interaction model, Equation 3, is
182
MIODUSZEWSKI ET AL.
not be lognormal. The model (3) is an empirical model and is
limited to the conditions of the experiment from which it was
derived. These conditions include the strain of rat, the age, diet,
and health status of the rats, the activity level of the rats during
exposure, and concentration-time profiles to which the rats
were exposed.
The curvature in the interaction model fit of Part II data is
roughly equivalent to the curvature in the fit of Part I data. The
experimental design for Part I could not definitively rule out
the possible effects of season and shipment of rat. However,
the similarity in curvature between the fits of Part I and Part II
indicates that these effects are minor.
For comparison, a toxic load expression was also used to fit
the data from both Part I and Part II:
LC50 ⫽ (93.1)T (⫺.601) or L(C (1.66)T)50
⫽ 1855 共for male rats兲
(6)
LC50 ⫽ (76.4)T (⫺.601) or L(C (1.66)T)50
⫽ 1336 共for female rats兲.
(7)
In Figure 1, the fits from the interaction model [Equations
(4)and (5)] and the toxic load model (Equations 6 and 7) are
compared with the LC50 values from Table 2. The degree of
curvature in the interaction model fit is small (though statistically significant) relative to the straight-line fit of the toxic load
model. The findings of the present study are consistent with the
need to expand beyond dependence on Haber’s rule as advocated by ten Berge et al. (1986), Yee (1996), and Miller et al.
(2000).
FIG. 3. Mean Pupil diameter (symbols) in male and female rats before (24
h) and after (1 h, 1 day, 2 days, and 7 days) exposure to GB vapor. Average
pupil diameters are plotted for 5-, 60-, 240-, and 360-min exposures and their
associated range of CT (mg.min/m 3).
of marginal significance (p ⫽ .033). When the model (3) is fit
to the Part I data, the Sex*cLogC term is not significant (p ⫽
.184), whereas the other terms of Euqation 3 are highly significant (p ⬍ .001). Part I data should be valid to test the
Sex*cLogC term, because both sexes were exposed together in
nearly all the chamber runs. Thus, the data are unclear about
whether the Sex*cLogC term should be included in the model.
Equations 4 and 5 are extensions of the toxic load model. They
are referred to here as the interaction model because of the
presence of the interaction term cLogC*cLogT.
The interaction term, log(T)*log(C), allows the distribution
of individual tolerances to both concentration (at a fixed exposure duration) and exposure duration (at a fixed concentration) to be lognormal. The presence of the term [log(Time)]
squared or the term [log(Concentration)] squared in the model
would imply that the distribution of individual tolerances to
exposure duration or exposure concentration, respectively, can-
Male versus Female Sensitivity to GB Vapor-Induced
Lethality
Female rats were more sensitive to the lethal effects of GB
vapor than males in this study, based on the significance (p ⬍
0.001) of the Sex term in Equation 3. Also, probit analysis with
sex as a factor was performed for each duration (Mioduszewski
et al., 2001), and it was found that the concentration-response
for each exposure time could be differentiated by sex (with the
exception of a few instances [30 and 60 min]). In addition, a
review of the clinical sign data suggests that clinical signs of
toxicity appeared earliest in females, as a group, and progressed to more severe levels earlier than in male rats. These
findings are consistent with those of Callaway and Blackburn
(1954), who reported (for 1-min exposures) that female rats
were almost twice as sensitive to the lethal effects of GB vapor
than males. McPhail (1953) reported that the male mouse was
more sensitive to GB vapor than the female, but the reverse
was true for the hamster and the rat. In addition, female rats
have been shown to be more sensitive than males to the lethal
effects of Soman (Sket, 1993). In female rats, the LD50 for
Soman (GD) was only about half that of males. This pattern
INTERACTION OF EXPOSURE CONCENTRATION AND DURATION
was also reported for lethal exposure of rats to some organophosphate (OP) insecticides (Sheets et al., 1997).
GB Vapor Effects on Blood AChE and BuChE
The most commonly accepted mechanism by which nerve
agents are believed to induce acute toxicity is by inhibition of
AChE activity in target tissues. AChE normally limits the
action of endogenous acetylcholine (the chemical mediator of
junctional transmission at cholinergic synapses). The resulting
cholinergic hyperactivity is often expressed in a variety of
toxic clinical signs. Although red blood cell (RBC) and plasma
cholinesterase activities are routinely monitored as a sensitive
index of exposure to anticholinesterase agents, they by no
means imply anticholinesterase intoxication (Koelle, 1994).
Sidell (1992) suggests that activity of the circulating ChE does
not parallel the activity of ChE in tissue and that tissue function
can be reasonably normal even with minimal blood ChE activity. If an OP compound is administered in low concentration
levels over a long period, blood levels of ChE of an animal can
drop to near zero, yet the animal survives. If blood levels of
ChE are caused to rapidly drop to zero, the animal dies.
Consequently, as confirmed by the results of the present study,
poor correlation is expected between probability of toxic signs
in the lethal range and exposure dosage.
It has been speculated that serum esterases (phosphatases,
carboxylesterases, plasma BuChE, and red blood cell AChE)
and plasma proteins may reduce OP availability by binding it
or hydrolyzing the compound to a less toxic metabolite (Ecobichon and Comeau, 1973). The most important OP scavengers
in rodents are carboxylesterases (CaEs) in plasma and organs,
whereas endogenous plasma BuChE has only a minor role as a
detoxification route for OPs (Clement, 1984; Grubic et al.,
1988). Maxwell et al. (1987a) suggested this may be due to the
greater frequency of CaE binding sites.
Sex differences in the activity of BuChE and CaE but not
(AChE) have been reported both in rodents (Sterri and Fonnum, 1989; Sterri et al., 1985) and in man (Jensen et al., 1995).
Sterri and Fonnum (1989) speculated that the 2-fold higher
BuChE activity in female rat plasma is probably due to a
doubling in the number of enzyme molecules, which also
should double the number of binding sites for OPs. In the
present study, female rats were more sensitive than males to
GB vapor-induced toxic effects, even though the female rat
preexposure mean BuChE activity was approximately four
times higher than the corresponding BuChE activity in male
rats. If BuChE was effectively minimizing the availability of
GB, less toxicity would be expected in female rats. Although
Sarin appears to be binding to BuChE in the present study (Fig.
2), BuChE activity was not inhibited as effectively in female
rats as in males, as suggested by higher incidences of toxic
signs in females (Table 1). Because CaE activity was not
measured in this study, it would be difficult to speculate about
the possible role of CaE in accounting for sex differences in
183
sensitivity to GB toxicity. The greater importance of CaEs
versus BuChEs as a detoxifying resource of OPs, has been
emphasized by several investigators (Clement, 1984; Maxwell
et al., 1987a,b).
Pupil Response to GB Vapor Exposure
Insofar as the GB vapor concentrations used in this study
were selected for estimating the LC50, it is not surprising that
maximal pupil constriction was seen in all rats during the first
24 h after exposure. However, the marked and consistent
reversal of this response (Fig. 3), progressing temporarily to
mydriasis, was not expected, as such responses are rarely
reported in the literature. In Figure 3, there is a clear separation
of pupil response (at 2 days postexposure) based upon exposure time group (5, 60, 240, and 360 min). Nevertheless, there
are difficulties in interpreting the dose-response relationship
for mydriasis due to possible confounding of sex and concentration, because the males were sometimes exposed to higher
vapor concentrations. In addition, there were different proportions of survivors among various categories (male vs. female,
and different concentration/exposure time groups [see Table
1]). Lastly, the entire time profile of pupil diameter change was
not recorded. Measurements were only made preexposure and
postexposures at 1 h, 1 day, 2 days, and 7 days. Thus, it is
unclear whether the maximum mydriasis was observed, and
therefore it cannot be determined how it depends on exposure
conditions.
GB vapor-induced changes in pupil diameter in the present
study may likely be a local ocular effect of GB. It is possible
that GB exposure altered the balance between sympathetic
versus parasympathetic control over pupil size, which changed
over time after exposure. It can be speculated that sometime
after the start of exposure (within 24 h), the cholinergic component predominated in the absence of AChE activity, resulting in miosis. Within 2–3 days after exposure, cholinergic
desensitization occurred, resulting in noradrenergic dominance
and subsequent mydriasis. Beyond 3 days postexposure, homeostasis between opposing sympathetic and parasympathetic
control of pupil size was restored.
Summary
This study evaluated the adequacy of traditional models in
predicting the toxicity of Sarin vapor exposure in the rat, given
various exposure conditions. Incidental data were also collected on changes in blood cholinesterase activity and pupil
response due to Sarin vapor exposure. The lethality data generated from this study were used in formulating models for
estimating the probability of Sarin vapor-induced lethality in a
rodent model, given a combination of exposure concentration
and duration. The findings of the present study are consistent
with the need to expand beyond dependence on Haber’s rule,
as advocated by ten Berge et al. (1986), Yee (1996), and Miller
et al. (2000).
184
MIODUSZEWSKI ET AL.
ACKNOWLEDGMENTS
McPhail, M. K. (1953). Sex and the response to G agents. Suffield Technical
Paper No. 38. Suffield Experimental Station, Ralston Alberta, CA.
The authors thank Dr. Newton Foster, Dr. Julie Watson, Mr. Dennis Johnson, Ms. Jackie Scotto, Mr. Carl Kurnas, and Mr. Dean Bona of the Veterinary
Services Team (ECBC) for their support in the care and handling of the
animals used in these studies and for quality assurance assistance. In addition,
the clinical laboratory support of Dr. John Skvorak, Ms. Connie Clark, and Mr.
Steve Tucker (USAMRICD) is greatly appreciated. Lastly, the authors recognize the expertise of Messrs. Lee Crouse, Charles Crouse (GeoCenters), and
Steve Anthony and Mark Haley (ECBC) for their valuable help in vapor
generation and analysis, and Ms. Michelle Murray (GeoCenters) for help with
data processing.
Miller, F. J., Schlosser, P. M., and Janszen, D. B. (2000). Haber’s rule: A
special case in a family of curves relating concentration and duration of
exposure to a fixed level of response for a given endpoint. Toxicology 149,
21–34.
REFERENCES
Bliss, C. I. (1934). The method of probits. Science 79, 38 –39.
Callaway, S., and Blackburn, J. W. (1954). A comparative assessment of the
vapour toxicities of GB, GD, GF, T.2132, T.2137 and T.2146 to male and
female rats, PTP 404, CDE, Porton Down, UK.
Clement, J. G. (1984). Importance of aliesterase as a detoxification mechanism
for soman (pinacolyl methyl phosphonoflouridate) in mice. Biochem. Pharmacol. 33, 148 –157.
Ecobichon, D. J., and Comeau, A. M. (1973). Pseudocholinesterase of mammalian plasma, physiochemical properties and organophosphate inhibition
in eleven species. Toxicol. Appl. Pharmacol. 24, 92–100.
Ellman, G. L., Courtney, K. D., Andres, V., Jr., and Featherstone, R. M.
(1961). A new rapid colormetric determination of acetylcholinesterase activity. Biochem. Pharmacol. 7, 88 –95.
Grubic, Z., Sket, D., and Brzin, M. (1988). Iso-OMPA induced potentiation of
soman toxicity in rats correlates with the inhibition of serum carboxylesterases. Arch. Toxicol. 62, 398 –399.
Haber, F. R. (1924). Zur geschichte des gaskrieges. In Funf Vortrage aus
Jahren 1920–1923, pp. 23–29. Julius Springer, Berlin.
Jensen, F. S., Skovgaard, L. T., and Viby-Mogensen, J. (1995). Identification
of human plasma cholinesterase variants in 6688 individuals using biochemical analysis. Acta Anaesthiol. Scand. 39, 157–162.
Koelle, G. B. (1994). Pharmacology of organophosphates. J. Appl. Toxicol. 14,
105–109.
Maxwell, D. M., Brecht, K. M., and O’Neill, B. L. (1987a). The effect of
carboxylesterase inhibition on interspecies difference in soman toxicity.
Toxicol. Lett. 39, 35– 42.
Maxwell, D. M., Lenz, D. E., Groff, W. A., Kaminskis, A., and Froehlich,
H. L. (1987b). The effects of blood flow and detoxification on in vivo
cholinesterase inhibition by soman in rats. Toxicol. Appl. Pharmacol. 88,
66 –76.
MacFarland, H. N. (1987). Designs and operational characteristics of inhalation exposure equipment. In Inhalation Toxicology (H. Salem, Ed.), pp.
93–120. Marcel Dekker, New York.
MINITAB (1999).Version 13, Minitab Inc., State College, PA.
Mioduszewski, R. J., Mathei, J. H., Way, R. A., Burnett, D. C., Gaviola, B. P.,
Muse, W. T. Jr., Anthony, J. S., Durst, H. D., Sommerville, D. R., Crosier,
R. B., Thomson, S. A., and Crouse, C. L. (2001). ECBC Low Level
Operational Toxicology Program: Phase I—inhalation toxicity of sarin
vapor in rats as a function of exposure concentration and duration. ECBCTR-183. U.S. Army Edgewood Chemical Biological Center, Aberdeen Proving Ground, MD.
Muse, W., Anthony, S., Buettner, L., Crouse, C., Crouse, L., Mioduszewski,
R., and Thomson, S. (2000). Generation, sampling and analysis of Sarin
(GB) vapor for inhalation studies. Proceedings of the International Chemical
Weapons Demilitarization Conference (CWD 2000), sponsored by DERA,
Porton Down, UK, 21–24 May 2000, The Hague, Netherlands.
NRC (1997). Review of Acute Human-Toxicity Estimates For Selected Chemical-Warfare Agents. Committee on Toxicology, Subcommittee on Toxicity
Values for Selected Nerve and Vesicant Agents, National Research Council.
National Academy Press, Washington, DC.
Sheets, L. P., Hamilton, B. F., Sangha, G. K., and Thyssen, J. H. (1997).
Subchronic neurotoxicity screening studies with six organophosphate insecticides: An assessment of behavior and morphology relative to cholinesterase inhibition. Fundam. Appl. Toxicol. 35, 101–119.
Sidell, R. F. (1992). Clinical considerations in nerve agent intoxication. In
Chemical Warfare Agents (S. M. Somani, Ed.), pp. 129 –179. Academic
Press, New York.
Sket, D. (1993). Efficacy of antidotes against soman poisoning in female
physostigmine-protected rats. Pharmacol. Toxicol. 72, 25–30.
Sterri, S. H., and Fonnum, F. (1989). Carboxylesterase—the soman scavenger
in rodents; heterogeneity and hormonal influence. In Enzymes Hydrolyzing
Organophosphorus Compounds (E. Reiner, W. N. Aldridge, and F. C. G.
Hoski, Eds.), pp. 155–164. Ellis Horwood Publishing, West Sussex, UK.
Sterri, S. H., Johnsen, B. A., and Fonnum, A. (1985). A radiochemical assay
method for carboxylesterase, and comparison of enzyme activity towards the
substrates methyl[1-14c]butyrate and 4-nitrophenyl butyrate. Biochem.
Pharmacol. 34, 2779 –2785.
ten Berge, W. F., Zwart, A., and Appelman, L. M. (1986). Concentration-time
mortality response relationship of irritant and systemically acting vapours
and gases. J. Hazard. Mater. 13, 301–309.
Yee, E. (1996). A non-linear dose-response model with an application to the
reconstruction of the human mortality response surface from acute inhalation toxicity with Sarin. DRES-SM 1476, Defence Research Establishment
Suffield, Rallston, Alberta, Canada.