Mackay et al 2017 Chemical exposure toxicity space model.pdf

Environmental Toxicology and Chemistry, Vol. 36, No. 5, pp. 1389–1396, 2017
Published 2016 SETAC
Printed in the USA
THE CHEMICAL EXPOSURE TOXICITY SPACE (CETS) MODEL: DISPLAYING EXPOSURE
TIME, AQUEOUS AND ORGANIC CONCENTRATION, ACTIVITY, AND ONSET OF TOXICITY
DONALD MACKAY,y ALENA K.D. CELSIE,*yz J. MARK PARNIS,yz LYNN S. MCCARTY,x JON A. ARNOT,k#
and DAVID E. POWELLyy
yChemical Properties Research Group, Department of Chemistry, Trent University, Peterborough, Ontario, Canada
zDepartment of Chemistry, Queens University, Kingston, Ontario, Canada
xL.S. McCarty Scientiﬁc Research Consulting, Newmarket, Ontario, Canada
kArnot Research and Consulting (ARC), Toronto, Ontario, Canada
#Department of Physical and Environmental Sciences, University of Toronto Scarborough, Toronto, Ontario, Canada
yyDow Corning Corporation, Health and Environmental Sciences, Auburn, Michigan, USA
(Submitted 14 July 2016; Returned for Revision 2 September 2016; Accepted 27 October 2016)
Abstract: A 1-compartment toxicokinetic model is used to characterize the chemical exposure toxicity space (CETS), providing a novel
graphic tool that can aid in the design of aquatic toxicity tests for ﬁsh and for interpreting their results. The graph depicts the solution to
the differential equation describing the uptake kinetics of a chemical by a modeled ﬁsh under conventional bioassay conditions. The
model relates the exposure concentration in the water to a dimensionless time and the onset of toxicity as determined by an estimated or
assumed critical body residue or incipient lethal aqueous concentration. These concentration graphs are speciﬁc to each chemical and
exposure and organism parameters and clearly demonstrate differences in toxicity between chemicals and how factors such as
hydrophobicity inﬂuence the toxic endpoint. The CETS plots can also be used to assess bioconcentration test conditions to ensure that
concentrations are well below toxic levels. Illustrative applications are presented using a recent set of high-quality toxicity data.
Conversion of concentrations to chemical activities in the plots enables results for different baseline toxicants to be superimposed. For
chemicals that have different modes of toxic action, the increased toxicity then becomes apparent. Implications for design and
interpretation of aquatic toxicity tests are discussed. The model, and pictorial visualization of the time-course of aquatic toxicity tests,
may contribute to improvements in test design, implementation, and interpretation, and to reduced animal usage. Environ Toxicol Chem
2017;36:1389–1396. # 2016 The Authors. Environmental Toxicology and Chemistry Published by Wiley Periodicals, Inc. on behalf
of SETAC.
Keywords: Aquatic toxicology
Bioconcentration
Toxicokinetics
INTRODUCTION
experimental design and execution. Such design may be
In conventional acute aquatic toxicity tests such as those
assisted by employing a predictive model of the uptake
promulgated by the Organisation for Economic Co-operation
process, thus relating planned aqueous concentration to
and Development (OECD) test guideline 203 [1], test
time and to a body burden or incipient lethal concentration
organisms (ﬁsh) are exposed to deﬁned concentrations of
that is postulated to result in lethality. The model need not be
the subject chemical in aqueous solution for a speciﬁed time
highly accurate, but even an approximate model can suggest
that is typically 96 h. The objective of the experiment is to
whether conditions are likely to be successful. The model
establish the concentration that causes death of a deﬁned
can also contribute to the design of bioconcentration factor
percentage (usually 50%) of the ﬁsh at the conclusion of the
(BCF) tests in which the aim is to measure the uptake and
test, namely the LC50. Alternatively, a potentially lethal
accumulation of chemical from water, but under sublethal
concentration may be prescribed, and the time-to-death of
conditions. If there is insufﬁcient time to achieve steady
50% of the ﬁsh is measured. Such tests have demanding
state, a correction can be applied employing an estimate of
requirements, especially if the results are to be used for
the loss rate constant (i.e., using an uptake rate constant
regulatory purposes. Test conditions are normally tightly
[k1]–loss rate constant [k2] approach such as that described
controlled in terms of concentration, temperature, oxygen
by Hendriks et al. [2]).
saturation, and number and condition of the test organisms,
In a series of reviews of aquatic toxicity testing, McCarty
resulting in considerable expense. There is thus an incentive
et al. [3] pointed out that regulatory demands have increased,
to avoid excessive and unproductive testing by careful
while the quality of many data in well-accepted databases is often
inadequate. This is because, in the translation from conceptual
models to operational testing models, not all important toxicity
This article includes online-only Supplemental Data.
modifying factors inﬂuencing linkages between fundamental
This is an open access article under the terms of the Creative Commons
thermodynamics and kinetic processes and adverse toxic effects
Attribution NonCommercial NoDerivs License, which permits use and
distribution in any medium, provided the original work is properly cited,
associated with conventional dose metrics are considered.
the use is non-commercial and no modifications or adaptations are made.
We present a predictive model that employs equations and a
* Address correspondence to [email address]
graphic display to describe the dynamic uptake kinetics of a
Published online 16 November 2016 in Wiley Online Library
chemical by a ﬁsh under conventional bioassay conditions, thus
(wileyonlinelibrary.com).
relating exposure concentration in the water to time and onset of
DOI: 10.1002/etc.3668
1389

1390
Environ Toxicol Chem 36, 2017
D. Mackay et al.
toxicity. Furthermore, the display and equations can be
chemicals may exhibit speciﬁc modes of toxic action and have
presented in terms of chemical activity, which has been shown
lower CBRs [4].
to provide predictive capability for baseline nonpolar narcotic
Rearranging Equation 1 to estimate the LC50 concentra-
substances, often referred to as having a toxic mode of action
tion CW that yields a CF or CBR50 at a deﬁned exposure time t
(MOA1) or, in the case of polar substances, MOA2 [4]. One
gives
appealing feature of the model is that the results may be

graphically presented to visualize the approximate relationship
C
¼
W
LC50 mol m3
between the parameters and the outcomes. The corresponding
¼ CBR50= ½
ð 1  exp  fk þ
g
þ
Þ ð
2
kM t  k1=½k2
kM
3Þ
set of equations can describe these relationships more
accurately. Essentially, the graph and model satisfy the need
where k1/(L  KOW) is equal to k2, and the BCF is equal to
for an a priori assessment of the likely success of the proposed
L  KOW. If kM is 0, Equation 3 simpliﬁes to
test design.

C
¼
W
LC50 mol m3
MODEL AND PARAMETERS
¼ CBR50=ð 1
½  expfk ð
Þg
Þ
1t= L  KOW
L  KOW
To develop the model and the graphic display, we employ the
ð4Þ
conventional 1-compartment ﬁrst-order toxicokinetic model as
described by Arnot and Gobas [5] in their AQUAWEB model.
It is convenient to deﬁne the group (1–exp[– {k þ
2
kM}t]) as Ф,
This gives the equation for uptake from water by respiration as it
the fractional approach to steady state or equilibrium. This
applies to standard short-term bioconcentration and toxicity
group is routinely used to correct measured ﬁsh concentrations
tests such as OECD guidelines 305 [6] and 203 [1], which are
at near-equilibrium conditions (tNE) to estimates of steady-state
described by Newman [7].
concentrations (tE); that is, tE is tNE/Ф [1,2]. At equilibrium and
steady state, Ф is 1.0. When t is 0, Ф is also 0, and thus Ф is
C ¼
ð
½ þ
Þ  ð
½ð f þ
g ÞÞ ð
F
CW
k1= k2
kM
1  exp  k2
kM t
1Þ
essentially a dimensionless time that is dependent on both the
organism and chemical. This gives an equation for LC50 as a
where CF is the concentration in the exposed aquatic organism
function of CBR50, Ф, BCFE (for nonmetabolizing chemicals)
(mol m3 wet wt), CW is the dissolved concentration in the
or with BCFM for metabolizing chemicals, and an equation for
water (mol m3), kM is the ﬁrst-order rate constant for
the ratio CBR50/BCF, which is designated in the present
metabolism or biotransformation (h1), and t is time. We
study as the incipient lethal concentration (ILC50) as deﬁned
estimate the respiratory k1 (m3 m3 h1, or h1) as G  E, where
by Sprague [12,13] or the threshold LC50 as deﬁned by
G is the respiration rate (m3 water/m3 ﬁsh h) and E is the
Wurhmann [14]. Newman [7] describes this approach in more
chemical transfer efﬁciency estimated from the chemical’s
detail. Rearranging to express LC50 and Ф as a function of
octanol–water partition coefﬁcient (KOW) [5,8]. Conventionally
ILC50 gives
k1 is expressed in units such as mmol kg1 h1, but the unit of
h1 gives identical numerical values assuming the ﬁsh density is
CBR50=BCF ¼ ILC50 ¼ LC50  F
ð5Þ
equal to that of water, 1000 kg m3.
The respiratory rate loss constant k2 (h1) is estimated as
Equation 5 suggests that the combined toxicity and
k1/(L  KOW) where L is the fractional lipid or effective octanol
bioconcentration of the substance can be conveniently ex-
content of the ﬁsh. For a more rigorous treatment of
pressed as CBR50/BCF, or ILC50. The ILC50 is the
biotransformation, it may be necessary to apply Michaelis–
concentration in water (mol m3) that results in the organism
Menten kinetics. Applying Equation 1 implies that the ﬁsh are
reaching its critical body residue at steady state, as would apply
not fed, and thus there is no chemical intake in food, growth
after long exposure times when Ф equals 1.0. It is thus the
dilution is negligible, and there are no losses by fecal egestion.
minimum aqueous concentration necessary to cause toxicity.
When t >>1/(k þ
2
kM), a steady-state concentration CF
The time to reach steady state is determined by the chemical
corresponding to a BCF (m3 m3) is approached
half-life in the ﬁsh, and is longer for more hydrophobic and
more persistent chemicals because total elimination rates
C
¼
ð þ
Þ
ð
F=CW
BCF ¼ k1= k2
kM
2Þ
become slower (i.e., k
>
2 is reduced). When kM
k2, the BCF,
and hence the relationship between the external exposure
When k

M
approaches 0, CF approaches CW
(k1/k2) or
concentration and the internal exposure concentration nearer the
C
W
BCFE, where BCFE is a steady-state and equilibrium
site of toxic action, can be lower than the thermodynamic
BCF that can be estimated as L  KOW for many nonpolar,
equilibrium BCFE.
neutral chemicals. If metabolism occurs, the lower steady-
These equations capture the expectation that the LC50
state but nonequilibrium bioconcentration factor BCFM is
depends on the aqueous exposure concentration, time of
k
þ
1/(k2
kM). Alternatively, estimates can be made of k1 and k2
exposure, and toxicity of the substance expressed in the CBR.
and their ratio equated to the BCF. The LC50 is assumed to
There is a hyperbolic relationship between exposure concentra-
occur when CF reaches a critical body residue corresponding
tion (LC50) and time expressed by Ф, their product being the
to an effect on 50% of the organisms (CBR50), usually in the
ILC50.
range of 2 mol m3 to 8 mol m3 or, equivalently, mmol kg1
These algebraic concepts are relatively simple and transpar-
by assuming a ﬁsh density equaling that of water, 1000 kg
ent, but it can be difﬁcult to visualize their relationships and
m3. This range applies to CBRs for acute baseline neutral
signiﬁcance. Accordingly, there is an incentive to present them
narcosis of hydrophobic chemicals in small aquatic organisms
in graphic form in what we call a chemical exposure toxicity
with approximately 5% lipid content [9] and is consistent with
space (CETS) plot analogous to the chemical partitioning space
the Redman et al. [10] or Di Toro et al. [11] analyses
diagrams of the air–water partition coefﬁcient (KAW), the KOW,
employing the target lipid model. Biochemically reactive
and the octanol–air partition coefﬁcient (KOA) used to identify

The chemical exposure toxicity space (CETS) model
Environ Toxicol Chem 36, 2017
1391
the likely multimedia partitioning tendencies of chemicals [15].
hydrophobic phases, especially membranes that are the probable
When the concepts are presented visually, the diagrams may
target site of neutral narcosis toxicity (MOA1 and MOA2). The
convey the relationships among exposure concentration, time,
assumption that BCF is L KOW also becomes invalid for
and toxicity more clearly. In the present study, we display
hydrophilic chemicals that partition appreciably into aqueous
the concepts graphically, providing a method by which the
phases in the ﬁsh. Our aim in the present study is only to present
problems associated with long uptake half-times of chemicals
the concept of the CETS in the expectation that future versions
and chemicals of various hydrophobicity can be identiﬁed. The
may include these reﬁnements.
basis of a CETS plot is Equation 5, in which the quantities
determining toxicity, CBR, and BCF or their ratio ILC50 are
Constraints
related to the product of CW or the LC50 and exposure time
A ﬁrst obvious constraint for an aquatic bioassay is that the
expressed as Ф. This is graphically depicted by plotting LC50 as
LC50 or CW must be less than the solubility of the test chemical.
a function of Ф (x-axis) and CW (y-axis), as illustrated in
The area on the CETS plot where CW is greater than the
Figure 1; Ф ranges from 0 to 1.0, and the corresponding times
solubility may thus be identiﬁed as being out of bounds, as
from 0 to 1. The half-time for uptake occurs when k2t is 0.693
depicted by the red zone in Figure 1. The minimum feasible
and Ф is 0.5. The locus of the intersections of the horizontal CW
LC50 for a test chemical may be estimated as the ILC50. For a
and vertical Ф values is a hyperbolic curve or boundary
test to be feasible, CW is thus limited to an upper limit of the
condition of constant LC50 and represents the combinations of
solubility and a lower limit of the ILC50 under typical
water concentrations and times to death that result in 50% or
conditions of exposure.
another deﬁned percentage of mortality. Increasing the toxicity
A second constraint for an aquatic bioassay is that the
has the effect of reducing the LC50 and moving the curve closer
maximum test time must be reasonable (because of feeding
to the origin, resulting in a reduction in LC50 values or shorter
constraints, etc.). The area on the CETS plot where Ф is greater
times-to death, or both.
than the maximum test time may be identiﬁed as being not
To construct a CETS plot, the ILC50 is deﬁned as the
experimentally feasible because of excessive test duration, as
aqueous concentration at Ф ¼ 1, and the line corresponding to
depicted by the blue zone in Figure 1. Only the remaining green
the hyperbolic boundary condition is established for CBR50/
region in Figure 1 represents a feasible combination of CW and
BCF by deﬁning individual water concentrations (or LC50s) as
test duration as speciﬁed by the 2 constraints and the ILC50
ILC50/Ф. For a speciﬁed time of exposure (Ф), the LC50 is
boundary condition. The time (h) corresponding to each Ф is
equal to C
–
W at the intersection of Ф and the CBR50/BCF or
ln(1–Ф)/k2.
ILC50 boundary condition. Conversely, the time to death for a
In addition to these aquatic bioassay constraints, we
speciﬁed LC50 is equal to Ф at the intersection of CW and the
emphasize that the domain of applicability is a 1-compartment
ILC50 boundary condition. If both exposure time and LC50 are
approximation of multicompartment organisms where all
known, the intersection of CW and Ф on the ILC50 boundary
kinetics are assumed to be ﬁrst order, all compartments achieve
condition deﬁnes the ILC50. It may be desirable to add a
a reasonable approximation of steady state within the test
secondary scale giving time directly under the Ф axis. If the test
duration, there is no signiﬁcant biotransformation, and only 1
yields time-to-death data for a single or speciﬁc ﬁsh, the same
mode of toxic action dominates the adverse effect associated
relationships hold but the LC50 becomes an LC. The CBR and
with the response endpoint being employed.
ILC can then be measured directly.
We emphasize that this is a simple 1-compartment model
Interpretations and implications of the CETS plot
that does not treat the time-dependent distribution of the
In the interests of simplicity, we ﬁrst discuss a series of
chemical within the ﬁsh, nor in its present form does it address
conditions assuming kM to be negligible. Increasing chemical
differences between partitioning to octanol and various lipid/
hydrophobicity then has 2 effects and results in the shrinking
of the green zone for feasible test conditions (Figure 1).
Speciﬁcally, if the hydrophobicity (i.e., KOW) of the test
chemical doubles, it is likely that solubility of the test chemical
is reduced by a factor of 2 and the horizontal solubility limit line
falls (i.e., the area of the red zone increases). If a test duration
such as 96 h is prescribed, then for a given k1 for the test
organism, k2 falls by a factor of 2 and Ф is reduced by a similar
factor and the vertical maximum test time decreases (i.e., the
area of the blue zone increases). The ILC line also falls. The net
effect is that the green zone of feasible test conditions shrinks
and can approach a single point or even vanish when ILC50/Ф is
equal to the chemical’s solubility. Figure 2 illustrates the
condition that corresponds to an aquatic bioassay being
conducted at the solubility of the solid or liquid state chemical
as appropriate, and toxicity occurs at the prescribed test time.
For some chemicals, the green zone of feasible test
conditions may not include the ILC50 boundary condition,
Figure 1. Illustrative chemical exposure toxicity space (CETS) plot of
and thus it is impossible to demonstrate a toxic effect because
concentration in water versus exposure duration, showing the region of
both constraints (i.e., low solubility and excessive exposure
feasible toxicity tests in green. The red and blue regions are experimentally
duration) apply, as shown by the purple zone in Figure 3. The
inaccessible. An increase in chemical toxicity causes the critical body
locations of the solubility and the ILC50 boundary condition are
residue corresponding to an effect on 50% of the organisms/bioconcentra-
tion factor (CBR50/BCF) line to drop vertically. ILC50 ¼ incipient lethal
ﬁxed by toxicity and BCF, as determined by KOW and lipid
concentration that causes 50% mortality.
content L, respectively. Thus, the only option for conducting a

1392
Environ Toxicol Chem 36, 2017
D. Mackay et al.
chlorinated alkanes or parafﬁns, cyclic and linear permethyl
siloxanes, and dyes. These constraints may also apply to
certain highly hydrophobic aromatics such as PCBs, dioxins,
and brominated ﬁre retardants, but these substances may
display other modes of toxic action. Experimental determina-
tion of the aquatic toxicity of these hydrophobic substances
is thus fraught with difﬁculties relating to the necessarily
low aqueous concentrations, possible losses by evaporation,
decreased bioavailability in the water phase caused by the
inevitable presence of organic matter, and excessively long
exposure times, which prevent the organisms from being
maintained in reasonable condition. To achieve a toxic
endpoint, it may be necessary to feed the organisms with a
diet of contaminated food similar in principle to that suggested
in the modiﬁed OECD aqueous and dietary exposure
Figure 2. Illustrative chemical exposure toxicity space (CETS) plot in
bioaccumulation test (OECD guideline 305 [6]). There are
which a toxicity test is feasible at only 1 set of test conditions; that is, the
obvious economic, scientiﬁc, animal usage, and regulatory
locus of feasible test conditions is reduced to a single point corresponding to
the solubility and the prescribed test time. CBR50 ¼ critical body residue
incentives to determine in advance the existence and area of
corresponding to an effect on 50% of the organisms; BCF ¼ bioconcentra-
the green zone, thus avoiding nonproductive tests.
tion factor; ILC50 ¼ incipient lethal concentration that causes 50%
Although it is not considered in any detail in the present
mortality.
study, the domain of applicability of this model is not limited
to hydrophobic chemicals. An examination of hydrophilic
chemicals (log K
<
OW
1) is likely to suggest that standard ﬁxed-
successful bioassay is to create a green zone by employing a
duration exposure testing such as 96 h or 24 h may be
smaller organism with larger values of k1 and k2, the ratio of
excessively long for some chemicals and organisms. This is
which (BCF) is constant and dictated by L and KOW. This
consistent with the long-standing advice of Sprague [12,13] to
effectively moves the ILC50 boundary condition to the left and
conduct and report aquatic toxicity tests to incipient, threshold,
thus closer to the origin.
or steady state rather than at ﬁxed exposure durations.
An obvious strategy when designing an aquatic toxicity test
When biotransformation occurs and kM contributes to the
is to compile a CETS diagram similar to Figure 1 but for the
overall chemical elimination rate constant and the metabolites
intended organism and chemical of interest and determine the
are relatively nontoxic, this increases the total elimination rate
extent of the green zone, especially if it includes the ILC50 line,
constant (k þ
2
kM), reducing the BCF, increasing the ILC50, and
assuming ﬁrst that the chemical is only a baseline toxicant
reducing Ф and the half-time for uptake. If biotransformation
(MOA1 or MOA2). If a chemical is believed to have an
rate constant estimates are available (e.g., from in vivo, in vitro,
additional MOA, then the CBR and ILC50 can be reduced by the
or in silico methods), the estimates should be included to better
toxic ratio.
characterize the range of experimental conditions.
Chemicals with no green zone are likely to be hydrophobic,
with low solubilities in water and low values of k
Logarithmic version
2,
necessitating long exposure times to achieve appreciable
The curvature of the LC50 boundary condition in Figures 1–3
body burdens. Chemicals of this type include highly
can be eliminated by taking logarithms of the x and y parameters
hydrophobic
substances
such
as
long-chain
alkanes,
in Equation 5, yielding Equation 6:
logðCBR50=BCFÞ ¼ logðILC50Þ ¼ logðLC50Þ þ logF ð6Þ
This linearizes the LC50 boundary condition on the CETS plot,
as shown in Figure 4. An attractive option is to draw this
diagram with a series of parallel LC50 lines of slope –1
representing different CBRs or BCFs. The ILC50 can be
determined by extrapolating the boundary condition to Ф of 1.0
or correspondingly to log Ф of 0.
ILLUSTRATION
To demonstrate the use and value of the CETS plot, we
reproduce the bioassay conditions described by van der Heidjen
et al. [16], who have reported a set of high-quality toxicity tests
for 6 organic chemicals conducted under carefully controlled
conditions by using passive dosing from a polymer phase to
Figure 3. Illustrative chemical exposure toxicity space (CETS) plot as in
establish and maintain constant water concentrations. In the
Figures 1 and 2, but showing that lethal conditions are not achievable
present study we focus on their results for the guppy (Poecilia
because of an excessively low solubility and an inadequate duration; that is,
reticulata), with approximately 10 mg to 110 mg weight range,
in the purple region, both constraints apply. CBR50 ¼ critical body residue
approximately 5% whole body lipid, and approximate tempera-
corresponding to an effect on 50% of the organisms; BCF ¼ bioconcentra-
ture of 20
tion factor; ILC50 ¼ incipient lethal concentration that causes 50%
8C. The chemicals, properties and selected test results
mortality.
are given in Table 1, namely, 1,2,4-trichlorobenzene, 1,2,3,4-

The chemical exposure toxicity space (CETS) model
Environ Toxicol Chem 36, 2017
1393
time death occurred. These times-to-death and resulting CBRs
for individual ﬁsh were measured [16].
For a speciﬁc chemical with a known KOW, the BCF can be
estimated as the product of lipid content (5%) and KOW (i.e.,
L  KOW which is also k1/k2). At the time of death (t), the ﬁsh is
undersaturated with chemical; that is, CBR < C
W
BCF and the
fraction of equilibrium Ф is CBR/(C
W
BCF). The ILC is, by
deﬁnition, equal to both CBR/BCF and CW/Ф. The loss rate
constant k2 can then be calculated as –ln(1–Ф)/t and k1 as
k
2
BCF. A k1 can then be calculated for each data point and an
average deduced for each of the low, medium, and high
concentration data sets. Average values of CBR, k1, k2, and ILC
are calculated and are given in Table 1. An ILA is also calculated
as ILC/SW where SW is the chemical’s (liquid or subcooled)
Figure 4. Logarithmic chemical exposure toxicity space (CETS) plot of log
solubility in water (mol m3). In practice, it is preferable to give
Cw versus log Ф showing that the hyperbolic ILC50 line is linearized and its
greatest weight to points for which Ф is relatively small (i.e.,
slope is 1.0. Cw ¼ dissolved concentration in the water; CBR50 ¼ critical
body residue corresponding to an effect on 50% of the organisms;
<0.2). Under these conditions, most distant from equilibrium, k1
BCF ¼ bioconcentration factor; ILC50 ¼ incipient lethal concentration that
can also be estimated as approximately CBR/(C
W
t). It is
causes 50% mortality.
expected that the chemical-to-chemical variation in ILA is
smaller than that of the ILC. The reason is that ILCs are CBR/
BCF and can vary greatly because of differences in BCF as
tetrachlorobenzene, pentylbenzene (QCB), 2,3,4-trichloroani-
inﬂuenced by KOW. The ILC is thus not a good metric for
line (2,3,4-TCA), 2,3,5,6-tetrachloroaniline (2,3,5,6-TeCA),
comparing relative toxicity of different chemicals, the ILA being
and 4-chloro-3-methylphenol (4-Cl-3-MP). Biotransformation
a better metric. This is because ILA is ILC/SW and is thus CBR/
rates are assumed to be negligible in these calculations. The
(BCF  S

W) or CBR/(L  KOW
SW). The group KOW
SW is
solubility and KOW values used in the present study were
an estimate of the solubility of the chemical in octanol, which,
obtained from a handbook by Mackay et al. [17], except for
according to the general solubility equation of Ran and
2,3,4-TCA and 2,3,5,6-TeCA, which were obtained from EPI
Yalkowsky [20], is fairly constant and approximately 3000 mol
SuiteTM [18]. They are slightly different from the solubility
m3. Because K

OW
SW is fairly constant, ILA is proportional to
values reported by van der Heijden et al. [16], which were
the lipid normalized CBR.
calculated using SPARC software [19]. Table 1 also contains
As CBR is also expected to be fairly constant for a group of
chemical properties including estimated incipient lethal con-
chemicals with similar modes of toxic action; therefore, ILA
centrations (ILCs) and incipient lethal activities (ILAs).
should also be fairly constant. This constancy in CBR is evident
Three water concentrations were prescribed (low, medium,
for the entire group, but not for the ILA of 4-Cl-3-MP, which
and high) at levels corresponding to chemical activities (ratios
displays apparently anomalous behavior with a much lower
of concentration to water solubility of the liquid state chemical)
estimation of k1. The estimated solubilities in octanol (SO) range
of approximately 10% to 50%, and thus toxicity was inevitable.
from 1900 mol m3 to 2800 mol m3 for the ﬁrst 5 chemicals,
The experimental technique was to measure the time-to-death of
but a much larger value of 91,700 mol m3 apparently applies to
the ﬁsh at the 3 exposure levels at 20 8C. The exposure regimes
4-Cl-3-MP because of the high solubility in water. van der
used by van der Heijden et al. [16] corresponded to moving the
Heijden et al. [16] give a lower value of SW by a factor of 2,
test condition point along the horizontal concentration line of a
which still gives a very high value of SO. This high solubility in
CETS plot until it reached the ILC boundary condition, at which
water is reﬂected in a very low ILA. Ionization is not the cause
Table 1. Chemicals tested by van der Heijden et al. [16]: 1,2,4-TCB, 1,2,3,4-TeCB, QCB, 2,3,4-TCA, 2,3,5,6-TeCA, and 4-Cl-3-MPa
Property
1,2,4-TCB
1,2,3,4-TeCB
QCB
2,3,4-TCA
2,3,5,6-TeCA
4-Cl-3-MP
Log KOW
4.1
4.5
4.9
3.33
4.10
3.10
SW (liquid state,
0.22
0.06
0.026
0.95
0.17
72.83
mol m3)
S ¼

O
KOW
SW
2775
1897
2065
2031
2140
91 700
BCF ¼ L  KOW
629.5
1 581
3972
106.9
629.45
62.95
CBR (mmol/kg)
8.76 (4.9–18.3)
8.36 (7.0–31.9)
16.1 (9.8–22.9)
5.59 (1.7–8.2)
2.14 (0.6–2.8)
4.96 (1.4–5.8)
k ¼

1
k2
BCF (h1)
36.37 (24.1–57.9)
62.72 (40.9–132.1)
73.04 (39.2–97.7)
31.19 (3.9–80.7)
34.23 (10.0–70.2)
7.47 (1.0–17.2)
k ¼
2
ln(1 – Ф)/t (h1)
0.0587
0.0397
0.0184
0.291
0.0544
0.119
(0.04–0.09)
(0.03–0.08)
(0.01–0.02)
(0.04–0.76)
(0.02–0.11)
(0.02–0.27)
ILC ¼ CBR/BCF
0.014
0.0055
0.0042
0.049
0.0032
0.066
(mol m3)
(0.008–0.03)
(0.004–0.007)
(0.002–0.006)
(0.016–0.077)
(0.001–0.004)
(0.022–0.093)
ILA ¼ ILC/SW (unitless)
0.064
0.092
0.16
0.052
0.018
0.0009
(0.035–0.132)
(0.074–0.124)
(0.095–0.222)
(0.017–0.081)
(0.006–0.026)
(0.0003–0.001)
a Values for log KOW and SW were taken from Mackay et al. [17] if available; otherwise EPISuite [18] was used. Values for lipid content (L) and CBR are from
van der Heijden et al. [16]. Values for k1, k2, ILC, and ILA were calculated for each guppy trial reported by van der Heijden et al. [16], and then an average value
was calculated for each chemical (including low, medium, and high concentrations) and reported in this table. Ranges for values are in parentheses.
TCB ¼ trichlorobenzene; TeCB ¼ tetrachlorobenzene; QCB ¼ pentylbenzene; TCA ¼ trichloroaniline; TeCA ¼ tetrachloroaniline; 4-Cl-3-MP ¼ 4-chloro-3-
methylphenol; K
¼
¼
¼
OW
octanol–water partition coefﬁcient; SW
solubility of the chemical in water; SO
solubility of the chemical in octanol;
BCF ¼ bioconcentration factor; L ¼ lipid content; CBR ¼ critical body residue; k ¼
¼
1
uptake rate constant; k2
loss rate constant; ILC ¼ incipient lethal
concentration; ILA ¼ incipient lethal activities.

1394
Environ Toxicol Chem 36, 2017
D. Mackay et al.
of this high solubility because the dissociation constant exceeds
the experimental pH. Because 4-Cl-3-MP is more polar than the
other chemicals in Table 1, this and physical and biological
degradation issues may inﬂuence solubility estimation. It is
believed that the chemical activity coefﬁcient in water estimated
from the solubility is not equal to the activity coefﬁcient at dilute
conditions as exist during the test and during measurements of
KOW. A full discussion of the reasons for this apparently
anomalous behavior is beyond the scope of the present study;
but for less hydrophobic substances such as 4-Cl-3-MP, care
must then be taken when interpreting CBR, ILC, and ILA as
predictive metrics of toxicity.
Single-chemical CETS plot
Figure 5 gives the CETS plot for 1,2,4-trichlorobenzene. The
solubility limit is depicted, and the 3 horizontal dashed lines
correspond to the low, medium, and high exposure concen-
Figure 5. Linear chemical exposure toxicity space (CETS) concentration
plot for the guppy and 1,2,4-trichlorobenzene tested by van der Heijden et al.
trations. As the test exposure time increases, the point
[16]. The upper red line demarcates the solubility, and the 3 green dashed
corresponding to the exposure conditions moves along the
horizontal lines represent the 3 exposure concentrations. The ILC line is
horizontal line and approaches the blue ILC/Ф line (in this case,
shown by the lower horizontal red line. The gray and orange lines are
ILC is calculated to be 0.014 mmol/L), and death occurs close to
estimated limits for factors of 0.5 and 1.5, respectively, in the time-to-death.
the corresponding time (Ф). Because of natural variability in ﬁsh
Cw ¼ dissolved concentration in the water; ILC50 ¼ incipient lethal
concentration
that
causes
50%
mortality;
LC50 ¼ median
lethal
sensitivity and uptake rates, the times-to-death vary as
concentration.
designated by the data points. The short vertical lines represent
the average values of Ф for each test. To illustrate the sensitivity
of the ILC line to Ф, the gray and orange hyperbolic lines are
extrapolated value of activity at Ф equal to 1.0 or log Ф equal to
estimates corresponding to approximate factors of 0.5 and 1.5 in
0. The upper lines suggest a log ILA ranging from –0.71 to –1.03
the range of measured Ф. Figure 6 gives the same data in a
(ILA range, 0.195–0.093) for all chemicals but lower values of
logarithmic plot. Clearly, the model and diagram are success-
–1.72 (ILA, 0.019) for the chloroaniline and –3.04 (ILA,
fully describing the general onset of toxicity as a function of
0.00091) for the chlorophenol. These activities are within the
time and concentration. The logarithmic plot has the advantage
expected range for baseline narcotics. The chloroaniline and the
that the data can be extrapolated to a value of Ф of 1.0 to give an
chlorophenol have lower lethal activities of 0.019 and 0.0009,
estimate of the log ILC, namely, 1.82 mmol/L in this case.
respectively, because of the high solubilities. van der Heijden
Multiple-chemical CETS plots
et al. [16] have discussed possible reasons for ambiguity in the
relationship between CBR and time-to-death in cases of high
It is instructive to include in Figure 7 data for all the
exposure concentrations. The higher log ILA for QCB of
chemicals tested in a common CETS plot of the type shown in
approximately 0.2 may be suspect because it is the most
Figure 5. Clearly the chemicals differ considerably in lethal
hydrophobic, and has the longest uptake half-life and time-to-
concentrations and times-to-death. The question then arises as
death requiring considerable extrapolation of Ф.
to the causes of these differences. The ILC/Ф lines differ
because the BCFs differ, being dependent on KOW. Clearly, the
2 most polar chemicals corresponding to the upper lines, 2,3,4-
trichloroaniline and 4-chloro-5-methylphenol, are behaving
differently from the more hydrophobic chemicals because of the
high solubility and different afﬁnities for the assumed target
site; that is, KOW may not equal the membrane–water partition
coefﬁcient, and thus different CBRs and modes of toxic action
may apply. These illustrations also ignore any inﬂuence of
biotransformation.
Interpretation using chemical activity
Insight into differences in chemical toxicity can be obtained
by converting the concentrations to chemical activity, as
discussed recently by Thomas et al. [21]. Concentrations of
organic chemicals in water are readily converted to chemical
activities by dividing by the appropriate liquid state water
solubility, rendering the absolute values dimensionless. This
conversion affects the y-axis concentrations on CETS plots, and
the ILC becomes an ILA. The x-axis times and values of Ф are
unaffected. The data from the CETS plot in Figure 7 are shown
in a CETS activity plot in Figure 8 on a linear activity basis and
Figure 6. Logarithmic chemical exposure toxicity space (CETS) concen-
tration plot for the data from Figure 5 that results in the curved incipient
in Figure 9 on a logarithmic activity basis.
lethal concentration (ILC) line becoming linear, facilitating extrapolation to
Figure 9 provides a convenient method of obtaining the log
the limit of Ф ¼ 1.0 or log Ф ¼ 0 when CW equals the ILC. Cw ¼ dissolved
ILA for each chemical (which has a common CBR) as the
concentration in the water.

The chemical exposure toxicity space (CETS) model
Environ Toxicol Chem 36, 2017
1395
Figure 7. Log chemical exposure toxicity space (CETS) concentration plots
Figure 9. Logarithmic chemical exposure toxicity space (CETS) activity
for the guppy and all chemicals tested by van der Heijden et al. [16].
plot for the van der Heijden et al. [16] data for the guppy and all chemicals.
Cw ¼ dissolved concentration in the water.
The 2 most polar chemical are now the 2 lower lines.
(Lumbriculus variegatus and Hyalella azteca), and similar
DISCUSSION
CETS plots can be obtained but are not addressed in the present
The pictorial CETS plot clearly deﬁnes the concentration
study.
and exposure times required to achieve a toxic endpoint.
The CETS plot example used in the present study employed
Unrealistic conditions such as concentrations exceeding the
experimental data for organism properties including CBRs,
water solubility and excessive test times become immediately
organism size, time-to-death, and lipid contents; in the absence
apparent. The use of activity also shows the solubility limit
of these data, however, approximate values can be used, such as
directly. The plots can help identify cases in which it is
typical CBR ranges for a given MOA. This also applies to
difﬁcult or even impossible to achieve a toxic endpoint. This
chemicals that have less well-documented physical–chemical
is most likely for very hydrophobic substances that have low
properties, in which case approximate or estimated values can
solubility limits and long uptake half-times. Particularly
be used for KOW or SW. In such cases, when using approximate
difﬁcult are symmetrical hydrophobic substances such as
values to construct the CETS plot, the results may be less
hexachlorobenzene and anthracene that have low solid-state
accurate; however, even an approximate model can help in
solubilities because of their high melting points. A useful
developing successful test conditions.
feature of the CETS plots is that they clearly identify potential
Converting the aqueous concentrations to chemical activities
experimental difﬁculties in testing hydrophobic chemicals by
can provide additional insights into the test results. Chemicals
aqueous respiratory exposure.
with similar modes of toxic action will group together.
The CETS approach is particularly appropriate for assessing
However, general confounding inﬂuences, including bioavail-
time-to-death data obtained at constant water concentrations,
ability, metabolic degradation, and different modes of toxic
such as those obtained in the tests designed by van der Heijden
action, must be considered. The plots thus have the potential to
et al. [16]. Tests of this type are regarded as models for future
discriminate differences in mode of action that are not apparent
bioassays. The nature of the increasing exposure with time is
from LC50 data. The magnitude of the differences between
readily visualized from the plots, as is the onset of time-to-death.
modes of action can be regarded as a toxic ratio, as discussed by
In the van der Heijden study [16], 2 other organisms were tested
Meader et al. [22]. These ratios can be read directly from the
logarithmic activity plots analogous to Figure 9. Toxicity tests
are feasible only if ILA50/Ф < 1.0 or ILA50 < Ф. The likely
value of ILA50 can be estimated for baseline narcotics as 0.01 to
0.03 [21]. The implication is that Ф must exceed this range for a
successful test. The corresponding range of exposure times can
be estimated and compared with feasible values.
The general conclusion is that as hydrophobicity increases,
testing aquatic toxicity by the respiratory route becomes more
difﬁcult until it ultimately becomes impossible, especially if it is
not biotransformed. Other routes of exposure by diet are then
necessary. More rigorous evaluation of the differences in
toxicity of these and other chemicals requires careful
consideration of solubilities, lipid contents, biotransformation
rate constants, ﬁsh-speciﬁc values of k1 and k2, KOW, and
partition coefﬁcients that are only approximately related to
KOW. The most promising and rigorous approach for detailed
interpretation of toxicity is to follow the approach of van der
Heijden et al. [16] and seek data on the partition coefﬁcients of
the chemical to the various sites of accumulation including the
Figure 8. Linear chemical exposure toxicity space (CETS) activity plot for
the guppy and all chemicals tested by van der Heijden et al. [16].
target site of the toxic action using experimental data or

1396
Environ Toxicol Chem 36, 2017
D. Mackay et al.
predictive methods such as Abraham-type polyparameter linear
test-no-305-bioaccumulation-in-ﬁsh-aqueous-and-dietary-
free energy relationship or by fundamental quantum chemical-
exposure_9789264185296-en
based estimation using approaches such as COSMO-RS. Our
7. Newman MC. 2010. Fundamentals of Ecotoxicology, 3rd ed. CRC,
Boca Raton, FL, USA.
primary purpose in the present study has been to present the
8. Arnot JA, Mackay D, Bonnell M. 2008. Estimating metabolic
concept of the CETS plot as a step toward evaluation of
biotransformation rates in ﬁsh from laboratory data. Environ Toxicol
chemical-speciﬁc toxicity. The implications for effective and
Chem 27:341–351.
economic testing and minimizing animal usage are obvious.
9. McCarty LS, Mackay D. 1993. Body residues and modes of toxic
action. Environ Sci Technol 27:1719–1728.
10. Redman AD, Parkerton TF, McGrath JA, Di Toro DM. 1982.
Supplemental Data—The Supplemental Data are available on the Wiley
PETROTOX: An aquatic toxicity model for petroleum substances.
Online Library at DOI: 10.1002/etc.3668.
Environ Toxicol Chem 31:2498–506.
11. Di Toro DM, McGrath JA, Hansen DJ. 2000. Technical basis for
Acknowledgment—Financial support was kindly provided by Dow Corning,
narcotic chemicals and polycyclic aromatic hydrocarbon criteria, I.
DMER, and a Natural Sciences and Engineering Research Council
Water and tissue. Environ Toxicol Chem 19:1951–1970.
(NSERC) Discovery Grant to J.M. Parnis. J.A. Arnot acknowledges
12. Sprague JB, Ramsay BA. 1965. Lethal levels of mixed copper-zinc
ﬁnancial support from the European Chemical Industry Council (CEFIC)
solutions for juvenile salmon. J Fish Res Board Can 22:425–432.
LRI ECO16 Project: Critical Body Residue Validation for Aquatic
13. Sprague JB. 1969. Measurement of pollutant toxicity to ﬁsh I. Bioassay
Organisms Exposed to Chemicals Causing Toxicity by Baseline Narcosis.
methods for acute toxicity. Water Res 3:793–821.
Open access publication of this paper was provided by the Silicones
14. Wuhrmann K, Woker H. 1948. Experimentelle Untersuchungen uber
Environmental, Health, and Safety Center (SEHSC), a sector group of the
die ammoniak und blausaurevergiftung. Schweiz Z Hydrol 11:
American Chemistry Council (ACC).
210–244.
15. Gouin T, Mackay D, Webster E, Wania F. 2000. Screening
Data Availability—Data are available in van der Heijden et al. [16], and
chemicals for persistence in the environment. Environ Sci Technol
model equations may be found in Arnot and Gobas [5].
34:881–884.
16. Van der Heijden SA, Hermens JLM, Sinnige TL, Mayer P, Gilbert D,
REFERENCES
Jonker MTO. 2015. Determining high-quality critical body residues for
multiple species and chemicals by applying improved experimental
1. Organisation for Economic Co-operation and Development. 1992. Test no.
design and data interpretation concepts. Environ Sci Technol 49:
203: Fish, acute toxicity test. In Guideline for the Testing of Chemicals.
1879–1887.
Paris, France. [cited 2016 April 25]. Available from: http://www.oecd-
17. Mackay D, Shiu WY, Ma KC, Lee SC. 2006. Physical Chemical
ilibrary.org/environment/test-no-203-ﬁsh-acute-toxicity-
Properties and Environmental Fate for Organic Chemicals. Taylor &
test_9789264069961-en
Francis/CRC Group, Boca Raton, FL, USA.
2. Hendriks AJ, van der Linde A, Cornelissen G, Sijm D. 2001. The power
18. US Environmental Protection Agency. 2000–2012. EPISuite. Ofﬁce of
of size. 1. Rate constants and equilibrium ratios for accumulation of
Pollution Prevention Toxics and Syracuse Research Corporation
organic substances related to octanol-water partition ratio and species
(SRC), Washington, DC.
weight. Environ Toxicol Chem 20:1399–1420.
19. Hilal SH, Karickhoff SW, Carreira LA. 2004. Prediction of the
3. McCarty LS, Arnot JA, Mackay D. 2013. Evaluation of critical body
solubility, activity coefﬁcient and liquid/liquid partition coefﬁcient of
residue data for acute narcosis in aquatic organisms. Environ Toxicol
organic compounds. QSAR Comb Sci 23:709–720.
Chem 32:2301–2314.
20. Ran Y, Yalkowsky SH 2001. Prediction of drug solubility by the
4. Verhaar HJM, Ramos EU, Hermens, JLM. 1996. Classifying
general solubility equation (GSE). J Chem Inf Comp Sci 41:
environmental pollutants. 2: Separation of class 1 (baseline toxicity)
354–357.
and class 2 (‘polar narcosis’) type compounds based on chemical
21. Thomas P, Dawick J, Lampi M, Lemaire P, Presow S, van Egmond R,
descriptors. J Chemometrics 10:149–162.
Arnot JA, Mackay D, Mayer P, Burgos MG. 2015. Application of the
5. Arnot J, Gobas F. 2004. A food web bioaccumulation model for organic
activity framework for assessing aquatic ecotoxicology data for organic
chemicals in aquatic ecosystems. Environ Toxicol Chem 23:2343–2355.
chemicals. Environ Sci Technol 49:12289–12296.
6. Organisation for Economic Co-operation and Development. 2012. Test
22. Maeder V, Escher BI, Scheringer M, Hungerbuhler K. 2004. Toxic ratio
no. 305: Bioaccumulation in ﬁsh: Aqueous and dietary exposure. In
as an indicator of the intrinsic toxicity in the assessment of persistent,
Guideline for the Testing of Chemicals. Paris, France. [cited 2016
bioaccumulative, and toxic chemicals. Environ Sci Technol 38:
April 25]. Available from: http://www.oecd-ilibrary.org/environment/
3659–3666.