United States
Environmental Protection
Agency
Environmental Research
Laboratory
Gulf Breeze FL 32561
Research and Development
EPA-600/S4-84-074 Sept 1984
Project Summary
Development of a Fate/Toxicity
Screening Test
William W. Walker
A shake-flask screening test was
designed to rapidly evaluate the relative
degradation rates of a wide spectrum of
chemicals, each compared to methyl
parathion. Test chemicals evaluated
were bolero, bravo, dibutylphthalate,
dimilin, dursban, endosulfan, hoelon,
pentachlorobenzene, phorate, and
trifluralin. Diverse regimes of salinity,
pH, TOC, and microbial biomass were
encountered across space and time.
The experimental design for the
screening test embodies four
treatments: active sediment, sterile
sediment, active water, and sterile
water. Decay curves were produced
and rate constants and half-life values
determined.
Half-life values for the 10 chemicals
evaluated varied substantially with time
and geographic sampling site. In active
systems, 8 of the 10 chemicals
degraded more rapidly than methyl
parathion. Nine dibutylphthalate
screens were run involving six
geographic sites. Disappearance was
quite rapid in active treatments in all
screens. Disappearance curves
describing DBP abatement either: (1)
appeared to be substrate dependent
with the rate of degradation decreasing
as DBP was depleted; (2) appeared
independent of substrate concentra-
tion; or (3) reflected a marked increase
in degradation rate during the screening
period.
This Project Summary was developed
by EPA's Environmental Research
Laboratory. Gulf Breeze, FL, to
announce key findings of the research
project that is fully documented in a
separate report of the same title (see
Project Report ordering information at
back).
Introduction
Since the passage of the Toxic
Substance Control Act in 1 974, the need
has increased for a rapid but informative
screening test to evaluate the potential
environmental hazard of chemicals
produced and submitted for registration in
the United States each year. A screening
test for this purpose should be relatively
rapid and have the capability of detecting
toxic degradation products without use of
radiolabelled materials. These criteria are
essential in enabling industry, including
commercial laboratories, to utilize the
screening test on an infrequent basis
without prohibitive physical or economic
restraints. Data resulting from the test
should enable scientists to determine the
environmental hazard (or lack thereof) of
the compound tested and whether
additional, more sophisticated, testing is
in order.
Materials and Methods
A shake-flask screening test was
designed to rapidly evaluate a wide
spectrum of chemicals, each comparedto
methyl parathion. Test chemicals
evaluated during the project period were
bolero (thiobencarb), bravo (chloro-
thalonil), dibutylphthalate, dimilin (diflu-
benzuron), dursban (chloropyrifos),
endosulfan, hoelon (diclofop methyl),
pentachloronitrobenzene, phorate (timet),
and trifluralin. Diverse regimes of salinity,
pH, TOC, and microbial biomass were
encountered across space and time.
The screening test is comprised of two
parts: (1) the fate screen which considers
the degradation rate of the chemical; and
(2) the toxicity test which relates loss of
toxicity to disappearance of parent
compound.
-------
The experimental design for the fate
screen embodies four treatments: active
sediment (AS), sterile sediment (SS),
active water (AW), and sterile water (SW).
The AS treatment is natural water
amended with 500 mg/L underlying
sediment. The SS treatment is sediment-
amended water sterilized with two
percent formalin. The AW treatment is
untreated water, and SW formalin-
sterilized site water. The concentration of
each test chemical was 200 fjg/L (500
fjg/L for dibutylphthalate) in each of the
above four treatments, and incubation
was in the dark at 25 C and 140-150
rpm. Salinity and pH were maintained at
that of the site water at the time of
sampling. Disappearance of test chemi-
cal was determined by periodic sampling
and analysis by gas or liquid chromato-
graphic methods. Decay curves were
produced and rate constants determined
for each chemical by linear regression
according to the relationship InC = a +
K,x, where C = fjg testchemical/L, a =Y-
axis intercept, K, = slope, and x = time.
Half-life values (T1/2 = 0.693/K,) for each
chemical were then determined.
Periodically during each fate screen,
generally at time zero and at each half-life
thereafter, the acute, static toxicity of the
AS supernatant was determined with 1-
to 2-day-old Mysidopsis bahia or Daphnia
magna as test animal. The objective here
was to determine whether or not toxicity
declined with disappearance of parent
compound. Lack of toxicity abatement
would indicate the accumulation of toxic
degradation products.
Results and Discussion
Fate Screen
Half-life values for the 10 chemicals
evaluated during the project period are
shown in Table 1. In active systems, half-
lives varied from several hours (hoelon) to
one month (dimilin, bolero). For four of
the 10 compounds (bravo, hoelon, bolero,
DBP), half-life in the active sediment
treatment was less than 50 percent of
that in the sterile sediment counterpart,
indicating involvement of the
sediment/water microflora in the
degradation process. With endosulfan
and dimilin, degradation proceeded more
rapidly in sterile than in active systems.
Duplicate flasks for the four treatments
were used for two endosulfan screens:
formalin was the sterilizing agent in the
first screen and 50 mg/L mercuric
chloride in the second. Results were
identical across all replicates in both
screens. For dimilin, disappearance of
parent compound from aseptic systems
Table 1. Hall-Lite Values for Fate Screen Compounds
Half-Life in Days
Test Chemical
Bolero (March. 1980J
Bolero (August. 1981 f
Bravo (July. 1981)
Dibutylphthalate (April, 1982)
Dibutylphthalate (June 1. 1982)'
Dibutylphthalate (June 9, 1982)'
Dibutylphthalate (June 22. 1982)'
Dibutylphthalate (August. 1982)'
Dibutylphthalate (September, 1982)'
Dibutylphthalate (October, 1982)
Dibutylphthalate (November, 1982)
Dibutylphthalate (December. 1982)
Dimilin (April 2. 1982)'
Dimilin (April 16, 1982)'
Dimilin (April 29, 1982)'
Dursban (March. 1980)
Dursban (May, 1981)'
Endosulfan (October, 1981)'
Endosulfan (November, 1981)'
Hoelon (August, 1980)
Hoelon (October, 1980)'
Pentachloronitrobenzene (July. 1981)
Phorate (August. 1980)
Trilluralin (August. 1981)'
AS
6.7
42.2
2.4
0.6
1.9
10.8
2.0
2.4
2.2
1.5
2.9
2.3
2.8
2.0
14.1
17.7
25.0
15.6
13.6
0.3
0.04
6.3
1.2
3.8
SS
89.6
468.4
5.0
23.0
19.7
61.7
42.1
29.0
12.6
21.1
7.9
12.8
13.2
16.7
1.2
16.5
39.0
6.0
10.4
1.8
2.2
7.6
1.6
6.6
AW
32.1
nd2
8.1
13.6
3.4
17.0
3.8
9.0
3.8
4.0
4.8
3.7
nd
nd
31.5
16.3
26.6
8.2
6.2
0.6
0.2
5.2
1.1
7.4
SW
83.6
nd
10.2
97.6
23.0
61.3
99.2
30.7
49.1
54.9
1578.8
322.0
nd
nd
1.0
24.1
28.6
2.8
4.4
12.4
3.8
6.5
1.5
7.2
'/Wean of duplicate replications.
2Not done.
exceeded the active systems only in the
screen involving Horn Island, MS (April
29).
For methyl parathion, 16 screening
evaluations involving four geographic
sites were conducted. In the active
sediment treatment, methyl parathion
half-life ranged from 0.9 to 29.9 days
with a mean of 12.6 days and a standard
deviation of 6.5. In sterile sediment, half-
life ranged from 2.7 to 87.8 days with a
mean of 47.9 days and a standard
deviation of 25.5. In active water, the
range was 1.1 to 56.3 days, the mean
31.1 days, and the standard deviation
14.6. In the sterile water treatment half-
lives ranged from 2.7 to 108.6 days with a
mean of 44.3 days and a standard
deviation of 29.9. Methyl parathion was
screened concomitantly with each test
material and the degradation rates
compared through derivation of K] ratios
(K, testchemical/K, methyl parathion). In
active systems bravo, DBP, dimilin,
endosulfan, hoelon, PCNB, phorate, and
trifluralin reflected K, ratios greater than
one, indicating a more rapid degradation
rate than observed for methyl parathion.
Dursban abatement was slightly less
than that for methyl parathion in active
sediment and slightly greater than that
for methyl parathion in active water.
Bolero disappearance was somewhat
inconsistent but appeared to degrade at a
rate roughly equal to or slightly slower
than methyl parathion. In sterile systems
most test materials (bolero and DBP were
exceptions) degraded more rapidly than
did methyl parathion. Bolero degradation
was somewhat reduced as compared to
methyl parathion. DBP ratios in sterile
systems were erratic due to difficulties in
explaining DBP behavior by linear
regression. The idea of "standardizing"or
"normalizing" degradation rates using a
benchmark chemical has been proposed
by others and, at the screening level, may
well represent a valid method of
describing the behavior of a relatively
unknown xenobiotic.
Considerable project time was spent
evaluating the behavior of dibutylphtha-
late under the conditions of the fate
screen. Nine DBP screens were run
involving six geographic sites from
Louisiana, Mississippi, and Florida.
Disappearance was quite rapid in active
treatments in all screens. In AS
treatments, the time requiredfor residual
DBP to fall below detection limit ranged
from two to 13 days. In AW, this range
was from two to 17 days. In all sites,
degradation in AS was equal to or greater
than that in AW, emphasizing the
-------
involvement of thesediment microflora in
the degradative process. In the formalin-
sterilized systems, abatement was
substantially reduced. DBF loss in sterile
systems was inconsistent, appeared
resultant of site-specific factors, and as in
the active systems, was enhanced by the
presence of sediment. Neither salinity,
pH, biomass, nor TOC could singly or in
combination adequately explain
differences in DBF degradation either
between or within sites evaluated. Disap-
pearance curves describing DBP
abatement either: (1) appeared to be
substrate dependent with the rate of
degradation decreasing as DBP was
depleted; (2) appeared independent of
substrate concentration; or (3) reflected a
marked increase in degradation rate
during the screening period. Such
increases reflect adaptation of the
microflora present from a population that
can degrade DBP slowly or not at all to
one capable of metabolizing the DBP
molecule at a significantly higher rate.
Data of this type characterized by variable
decay curves are difficult to describe by
linear regression analysis. In no case did
decay curves indicating adaptation
produce coefficient of determination (r2)
values greater than 0.800, indicating that
the behavior of DBP under these
conditions cannot be adequately
described by linear regression.
William W. Walker is with the Gulf Coast Research Laboratory, Ocean Springs,
MS 39564.
C. Richard Cripe is the EPA Project Officer (see below).
The complete report, entitled "Development of a Fate/Toxicity Screening Test,"
(Order No. PB 84-246 370; Cost: $8.50. subject to change) will be available only
from:
National Technical Information Service
5285 Port Royal Road
Springfield, VA 22161
Telephone: 703-487-4650
The EPA Project Officer can be contacted at:
Environmental Research Laboratory
U.S. Environmental Protection Agency
Sabine Island
Gulf Breeze. FL 32561
•ft U. S. GOVERNMENT PRINTING OFFICE; 1984 — 759-015/7828
Toxicity Testing
Dibutylphthalate and hoelon reflected
minimum effective concentration (MEC)
values to mysids and daphnids in the
mg/L range. Toxicity tests were not
included with these fate screens. Toxicity
of the remaining test chemicals was
found to decrease in direct proportion to
abatement of parent compound. This
information indicates that toxic degrada-
tion products either were not produced
or, more likely, did not accumulate to a
toxic level.
------- |